From jharris at widomaker.com Thu Apr 1 03:24:00 2004 From: jharris at widomaker.com (Jason Harris) Date: Thu Apr 1 03:21:25 2004 Subject: key count, 2004-03-31 Message-ID: <20040401012400.GQ10980@pm1.ric-41.lft.widomaker.com> As of Thu Apr 1 01:16:48 2004 UTC, there are 171342 v3 pubkeys, 1817518 v4 pubkeys, 1831395 subkeys, and 1739 duplicate (short) keyids on kjsl.com. -- Jason Harris | NIC: JH329, PGP: This _is_ PGP-signed, isn't it? jharris@widomaker.com _|_ web: http://keyserver.kjsl.com/~jharris/ Got photons? (TM), (C) 2004 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: not available Url : /pipermail/attachments/20040331/fe6862f2/attachment.bin From ml at arend-whv.info Thu Apr 1 02:06:18 2004 From: ml at arend-whv.info (Thomas Arend) Date: Thu Apr 1 04:48:09 2004 Subject: Urgent In-Reply-To: <404C8037.8030308@sagraluzzatto.com.br> References: <404C8037.8030308@sagraluzzatto.com.br> Message-ID: <200404010206.21916.ml@arend-whv.info> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am Montag, 8. M?rz 2004 15:16 schrieb Rodrigo Padula - Editora Sagra Luzzatto: > I have problena to use PHP with Gnupg. > > Helpe-me!!! > I tried hard to locate your brain and get in contact to read it. Sorry, I failed. So, would you minded to use the conventional way and write your problem down? Regards Thomas > > ASS: Rodrigo Padula > > > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users - -- Verschl?sselte und signiert Mails bevorzugt. Encrypted and signed mails prefered. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.3.2 (GNU/Linux) iD8DBQFAa1z72TqsmTFMxwkRAhptAJ9ibu0IkvFgQU14CbxmHAhOvwo7jACfXM9o rptxvuFFcQHdCR2Bxohv4jU= =7YG/ -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Thu Apr 1 09:23:38 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Thu Apr 1 09:20:54 2004 Subject: key count, 2004-03-31 In-Reply-To: <20040401012400.GQ10980@pm1.ric-41.lft.widomaker.com> References: <20040401012400.GQ10980@pm1.ric-41.lft.widomaker.com> Message-ID: <200404010923.42488@fortytwo.ch> On Thursday 01 April 2004 03.24, Jason Harris wrote: > As of Thu Apr 1 01:16:48 2004 UTC, there are 171342 v3 pubkeys, > 1817518 v4 pubkeys, 1831395 subkeys, and 1739 duplicate (short) keyids > on kjsl.com. How complicated would it be to add a category 'keys with only self-signatures' to the count? cheers -- vbi -- The content of this message may or may not reflect the opinion of me, my employer, my girlfriend, my cat or anybody else, regardless of the fact whether such an employer, girlfriend, cat, or anybody else exists. I (or my employer, girlfriend, cat or whoever) disclaim any legal obligations resulting from the above message. You, as the reader of this message, may or may not have the permission to redistribute this message as a whole or in parts, verbatim or in modified form, or to distribute any message at all. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040401/b2a9185a/attachment.bin From avbidder at fortytwo.ch Thu Apr 1 09:29:01 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Thu Apr 1 09:26:15 2004 Subject: keyserver, which to use? In-Reply-To: <406AD572.10097.1B4EC3B@localhost> References: <406AD572.10097.1B4EC3B@localhost> Message-ID: <200404010929.04234@fortytwo.ch> Please do not send HTML email to the list. Thanks. On Wednesday 31 March 2004 14.28, Stephan Reichenbach wrote: > we have been using pksd 0.96 for some time and I try to get a second > keyserver running, that will synchronize with pksd 0.96 in an intranet. pksd has numerous bugs. I advise against using it. SKS is a good keyserver under active development, and it can synchronize with pksd. If you decide to use pksd, be sure to download the newest version from the sourceforge project. pgp-keyserver-folk@flame.org is a list that you may be interested in. > the main thing is, that they keyserver must be able to receive keys > send bei "winpt" (author timo schulz). WinPT itself does not send and receive keys - it just remote-controls gnupg. In any case, both pksd and sks are written to operate with gnupg, so there should be no problem. > it should be installed under suse linux 8 or 9 SKS is written in ocaml, I don't know how much ocaml support there is in SuSE. But you should be able to download and install ocaml without problems (Or you may switch to Debian, which is better anyway :-) and supports ocaml out of the box). pksd should have no problem. cheers -- vbi -- Con el tiempo y la paciencia se adquiere la ciencia. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040401/fdf7be5e/attachment.bin From boldyrev+nospam at cgitftp.uiggm.nsc.ru Thu Apr 1 08:50:55 2004 From: boldyrev+nospam at cgitftp.uiggm.nsc.ru (Ivan Boldyrev) Date: Thu Apr 1 10:39:15 2004 Subject: Newbie Question References: <200403311414.21729.ml@tbulka.org> Message-ID: On 8700 day of my life Thomas Bulka wrote: > Hi, > as a GnuPG-Newbie I want to ask a very basic general > understanding question. > If I generate a key, is it supposed to work only > with the email-Address I entered during the generation > process? Really e-mail addresses are important only when someone encrypts message to you from mail client or from any other program that uses e-mail for selecting a key. > I mean, can I sign mails sent from another adress with this > key or am I supposed to create another key for every address I use? As you was told, you can add more user IDs. -- Ivan Boldyrev XML -- new language of ML family. From gnupg at ml0402.albert.uni.cc Thu Apr 1 12:00:08 2004 From: gnupg at ml0402.albert.uni.cc (Albert) Date: Thu Apr 1 11:58:42 2004 Subject: keyserver, which to use? In-Reply-To: <200404010929.04234@fortytwo.ch> References: <406AD572.10097.1B4EC3B@localhost> <200404010929.04234@fortytwo.ch> Message-ID: <200404011200.08286.gnupg@ml0402.albert.uni.cc> Am Donnerstag, 1. April 2004 09:29 schrieb Adrian 'Dagurashibanipal' von Bidder: > SKS is written in ocaml, I don't know how much ocaml support > there is in SuSE. For SuSE 8.2: pin 0.27 - package info for ocaml ------------------------------------------------------------------ *** no rpm named ocaml installed ------------------------------------------------------------------ ------------------------------------------------------------------ *** zgrep ocaml /var/lib/pin/ARCHIVES.gz ------------------------------------------------------------------ ---> ./CD1/suse/i586/ocaml-3.06-77.i586.rpm ---> ./CD2/suse/src/ocaml-3.06-77.src.rpm ./CD1/suse/i586/lablgtk-1.2.5-75.i586.rpm: drwxr-xr-x root root 5256 Mar 17 15:53 /usr/lib/ocaml/lablgtk ... Albert From jharris at widomaker.com Thu Apr 1 15:17:34 2004 From: jharris at widomaker.com (Jason Harris) Date: Thu Apr 1 15:15:21 2004 Subject: key count, 2004-03-31 In-Reply-To: <200404010923.42488@fortytwo.ch> References: <20040401012400.GQ10980@pm1.ric-41.lft.widomaker.com> <200404010923.42488@fortytwo.ch> Message-ID: <20040401131734.GA81455@pm1.ric-06.lft.widomaker.com> On Thu, Apr 01, 2004 at 09:23:38AM +0200, Adrian 'Dagurashibanipal' von Bidder wrote: > On Thursday 01 April 2004 03.24, Jason Harris wrote: > > As of Thu Apr 1 01:16:48 2004 UTC, there are 171342 v3 pubkeys, > > 1817518 v4 pubkeys, 1831395 subkeys, and 1739 duplicate (short) keyids > > on kjsl.com. > > How complicated would it be to add a category 'keys with only > self-signatures' to the count? That is the domain of keyanalyze: http://dtype.org/keyanalyze/ -- Jason Harris | NIC: JH329, PGP: This _is_ PGP-signed, isn't it? jharris@widomaker.com _|_ web: http://keyserver.kjsl.com/~jharris/ Got photons? (TM), (C) 2004 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: not available Url : /pipermail/attachments/20040401/2aa8d3bf/attachment.bin From avbidder at fortytwo.ch Thu Apr 1 16:03:23 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Thu Apr 1 16:00:42 2004 Subject: key count, 2004-03-31 In-Reply-To: <20040401131734.GA81455@pm1.ric-06.lft.widomaker.com> References: <20040401012400.GQ10980@pm1.ric-41.lft.widomaker.com> <200404010923.42488@fortytwo.ch> <20040401131734.GA81455@pm1.ric-06.lft.widomaker.com> Message-ID: <200404011603.26380@fortytwo.ch> On Thursday 01 April 2004 15.17, Jason Harris wrote: > On Thu, Apr 01, 2004 at 09:23:38AM +0200, Adrian 'Dagurashibanipal' von Bidder wrote: > > How complicated would it be to add a category 'keys with only > > self-signatures' to the count? > > That is the domain of keyanalyze: > > http://dtype.org/keyanalyze/ No I feel really, really old. Of course I eagerly download msd-sorted.txt.bz2 every two weeks and look at my ranking... I just never look at the summary page anymore. Oh, well.. :-/ -- vbi -- One picture is worth 128K words. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040401/707d149e/attachment.bin From ml at tbulka.org Thu Apr 1 14:56:14 2004 From: ml at tbulka.org (Thomas Bulka) Date: Thu Apr 1 18:47:04 2004 Subject: Newbie Question In-Reply-To: References: <200403311414.21729.ml@tbulka.org> Message-ID: <200404011456.14682.ml@tbulka.org> Am Donnerstag, 1. April 2004 08:50 schrieb Ivan Boldyrev: > On 8700 day of my life Thomas Bulka wrote: > > Hi, > > as a GnuPG-Newbie I want to ask a very basic general > > understanding question. > > If I generate a key, is it supposed to work only > > with the email-Address I entered during the generation > > process? > > Really e-mail addresses are important only when someone encrypts > message to you from mail client or from any other program that > uses e-mail for selecting a key. > > > I mean, can I sign mails sent from another adress with this > > key or am I supposed to create another key for every address I > > use? > > As you was told, you can add more user IDs. I sent another email to the list yesterday but somehow it got lost. I just wanted to thank you for your help, this works fine for me. Thomas From andreas.bergen at in-jesus.de Thu Apr 1 22:55:31 2004 From: andreas.bergen at in-jesus.de (Andreas Bergen) Date: Fri Apr 2 08:22:43 2004 Subject: hierarchical keys? Encryption to two subkeys impossible In-Reply-To: <20040331214959.GB15185@Update.UU.SE> References: <200403301911.58942.andreas.bergen@in-jesus.de> <200403312153.15154.andreas.bergen@in-jesus.de> <20040331214959.GB15185@Update.UU.SE> Message-ID: <200404012257.21514.andreas.bergen@in-jesus.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > This could possibly be accomplished by using subkeys. > I do not think they are intended to be used that way but it might be > possible. > > People have talked about having a master (key-)signing key on a secure > machine and exporting subkeys from that key to use on less secure > machines. That way the private key that collects signatures is safe and > it is still possible to sign with a well known key on less secure or > potentially uncontrolled machines (like at work) without risking the > real key. > > By using the primary key of an OpenPGP key as master and generating > subkeys for the delegated keys you should get the first level of > delegation, i do not know how to get a second level of delegation this way. I tried that and it worked. Thanks for the help. This raises another question, though. I tried encrypting with two encryption-keys (they're always subkeys, so there's no implicit master-slave-key-relation, but I think that's not necessary, as a user can't be forced to encrypt something so that someone else can decrypt it -- it's purely voluntary -- but I think it should be possible!). As of version 1.2.2 gpg seems to always encrypt to the last available encryption-subkey. There seems to be no way to encrypt to both (valid) subkeys, even when using the respective key-id. Here is an example: bergen@paulus<441> gpg --list-keys testkey pub 1024D/D2DFACE3 2004-04-01 testkey (Nur f?r Testzwecke) sub 1024g/5DB18EA4 2004-04-01 sub 1024g/EC2B2B4B 2004-04-01 cwd: /home/bergen/test/crypt/gpg bergen@paulus<442> gpg -e passwd Sie haben keine User-ID angegeben (Sie k?nnen die Option "-r" verwenden). Geben Sie die User-ID ein. Beenden mit einer leeren Zeile: 5DB18EA4 Added 1024g/EC2B2B4B 2004-04-01 "testkey (Nur f?r Testzwecke) " Geben Sie die User-ID ein. Beenden mit einer leeren Zeile: EC2B2B4B gpg: ?bersprungen: ?ffentlicher Schl?ssel bereits gesetzt Geben Sie die User-ID ein. Beenden mit einer leeren Zeile: As you can see, I enter 5DB18EA4 as key-ID to encrypt to, but it gets encrypted to EC2B2B4B. I think there should be a way to decrypt to both encryption-keys. Thanks for any help. Yours Andreas Bergen - -- Andreas Bergen E-Mail: andreas dot bergen at in-jesus dot de PGP/GnuPG-encrypted / -signed Email welcome. PGP-key-ID: 8CDEC18F Gott ist Liebe, und wer in der Liebe bleibt, bleibt in Gott und Gott in ihm. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQFAbIHd/28tHYzewY8RAiXMAJ4w4XCIvj+WSx4qQL27waPnU9NBAQCePHt5 Vs6W37bLa5WkWHwm/EUP6hs= =CUAM -----END PGP SIGNATURE----- From eleuteri at myrealbox.com Fri Apr 2 08:43:46 2004 From: eleuteri at myrealbox.com (David Picon Alvarez) Date: Fri Apr 2 08:41:03 2004 Subject: hierarchical keys? Encryption to two subkeys impossible References: <200403301911.58942.andreas.bergen@in-jesus.de><200403312153.15154.andreas.bergen@in-jesus.de><20040331214959.GB15185@Update.UU.SE> <200404012257.21514.andreas.bergen@in-jesus.de> Message-ID: <004801c4187d$da94f3f0$b9453cd5@enterprise> > As you can see, I enter 5DB18EA4 as key-ID to encrypt to, but it gets > encrypted to EC2B2B4B. I think there should be a way to decrypt to both > encryption-keys. Not sure if you've tried this, but try writing a bang (!) after the UID thus: gpg -bla-bla-bla-list-of-options 01234567! HTH, --David. From jason.harper at ge.com Fri Apr 2 15:56:10 2004 From: jason.harper at ge.com (Harper, Jason (GE Consumer Finance)) Date: Fri Apr 2 15:54:42 2004 Subject: GPG - Invalid packet Message-ID: <7730B3FE2184A5499A4A8601F598BB1B279027@ALPMLVEM01.e2k.ad.ge.com> <> Dear Group, I've been using gpg to do file encryption with partners of ours, but we just ran into two odd issues: 1. a GPG "don't know: invalid packet (ctb-1f) & (ctb-40) 2. a GPG "problem reading source" I've done the google thing on this, but I'm not making much progress. Does the list have any pointers for me to look at? We've not run into this before. The source file is being encrypted, and compressed and signed by an NAI 7.11 Business PGP system, and we are opening on a Solaris 8 box with gpg 1.2.3. I've sanitized some of the data to protect the innocent.. Sample one: gpg --output ce331 --decrypt CE_Con_1040326.dat gpg: WARNING: using insecure memory! gpg: please see http://www.gnupg.org/faq.html for more information gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created 2001-08-23 "KEYNAME " gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created 2004-03-25 "key@sample.com (COMPANY decription key) " gpg: Signature made Fri Mar 26 14:20:29 2004 EST using DSA key ID xxxxxxxx gpg: Good signature from "COMPANY " gpg: aka "COMPANY " gpg: [don't know]: invalid packet (ctb=1f) gpg: WARNING: message was not integrity protected gpg: [don't know]: invalid packet (ctb=40) Then it just ends, but we end up with a plain text file. Accuracy of the file is unkown. What is the CTB=1f and CTB=40 Sample two: ->> gpg --output au331 --decrypt AU_Con_1040331.dat gpg: WARNING: using insecure memory! gpg: please see http://www.gnupg.org/faq.html for more information gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created 2001-08-23 "COMPANY " gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created 2004-03-25 "key@sample.com (COMPANY decription key) " gpg: Signature made Fri Mar 26 14:20:46 2004 EST using DSA key ID xxxxxxx gpg: Good signature from "COMPANY " gpg: aka "COMPANY " File `au331' exists. Overwrite (y/N)? yes gpg: Problem reading source (2386467048 bytes remaining) gpg: handle plaintext failed: file read error gpg: WARNING: message was not integrity protected Thanks for any and all assistance. jason From jrhendri at maine.rr.com Fri Apr 2 16:19:37 2004 From: jrhendri at maine.rr.com (Jim Hendrick) Date: Fri Apr 2 16:17:11 2004 Subject: GPG - Invalid packet In-Reply-To: <7730B3FE2184A5499A4A8601F598BB1B279027@ALPMLVEM01.e2k.ad.ge.com> Message-ID: <002401c418bd$882f18d0$24211e0a@keane7926> do you know if it was encrypted with ascii-armor? I have seen problems with non-ascii systems sending files with wierd characters. Jim > -----Original Message----- > From: gnupg-users-bounces@gnupg.org > [mailto:gnupg-users-bounces@gnupg.org]On Behalf Of Harper, Jason (GE > Consumer Finance) > Sent: Friday, April 02, 2004 8:56 AM > To: gnupg-users@gnupg.org > Subject: GPG - Invalid packet > > > < some reason] > not subscribed on the correct e-mail address to post the first time>> > > Dear Group, > > I've been using gpg to do file encryption with partners of > ours, but we > just ran into two odd issues: > > 1. a GPG "don't know: invalid packet (ctb-1f) & (ctb-40) > 2. a GPG "problem reading source" > > I've done the google thing on this, but I'm not making much progress. > Does the list have any pointers for me to look at? We've not run into > this before. > > The source file is being encrypted, and compressed and signed > by an NAI > 7.11 Business PGP system, and we are opening on a Solaris 8 > box with gpg > 1.2.3. I've sanitized some of the data to protect the innocent.. > > Sample one: > > gpg --output ce331 --decrypt CE_Con_1040326.dat > gpg: WARNING: using insecure memory! > gpg: please see http://www.gnupg.org/faq.html for more information > gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created > 2001-08-23 > "KEYNAME " > gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created > 2004-03-25 > "key@sample.com (COMPANY decription key) " > gpg: Signature made Fri Mar 26 14:20:29 2004 EST using DSA key ID > xxxxxxxx > gpg: Good signature from "COMPANY " > gpg: aka "COMPANY " > gpg: [don't know]: invalid packet (ctb=1f) > gpg: WARNING: message was not integrity protected > gpg: [don't know]: invalid packet (ctb=40) > > Then it just ends, but we end up with a plain text file. > Accuracy of the > file is unkown. What is the CTB=1f and CTB=40 > > Sample two: > > ->> gpg --output au331 --decrypt AU_Con_1040331.dat > gpg: WARNING: using insecure memory! > gpg: please see http://www.gnupg.org/faq.html for more information > gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created > 2001-08-23 > "COMPANY " > gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created > 2004-03-25 > "key@sample.com (COMPANY decription key) " > gpg: Signature made Fri Mar 26 14:20:46 2004 EST using DSA key ID > xxxxxxx > gpg: Good signature from "COMPANY " > gpg: aka "COMPANY " > File `au331' exists. Overwrite (y/N)? yes > gpg: Problem reading source (2386467048 bytes remaining) > gpg: handle plaintext failed: file read error > gpg: WARNING: message was not integrity protected > > Thanks for any and all assistance. > > jason > > > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users > -------------- next part -------------- A non-text attachment was scrubbed... Name: winmail.dat Type: application/ms-tnef Size: 3620 bytes Desc: not available Url : /pipermail/attachments/20040402/cafa57d3/winmail-0001.bin From dshaw at jabberwocky.com Fri Apr 2 16:29:04 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 2 16:26:21 2004 Subject: GPG - Invalid packet In-Reply-To: <7730B3FE2184A5499A4A8601F598BB1B279027@ALPMLVEM01.e2k.ad.ge.com> References: <7730B3FE2184A5499A4A8601F598BB1B279027@ALPMLVEM01.e2k.ad.ge.com> Message-ID: <20040402142904.GB10518@jabberwocky.com> On Fri, Apr 02, 2004 at 08:56:10AM -0500, Harper, Jason (GE Consumer Finance) wrote: > I've been using gpg to do file encryption with partners of ours, but we > just ran into two odd issues: > > 1. a GPG "don't know: invalid packet (ctb-1f) & (ctb-40) > 2. a GPG "problem reading source" > > I've done the google thing on this, but I'm not making much progress. > Does the list have any pointers for me to look at? We've not run into > this before. > > The source file is being encrypted, and compressed and signed by an NAI > 7.11 Business PGP system, and we are opening on a Solaris 8 box with gpg > 1.2.3. I've sanitized some of the data to protect the innocent.. Could you run 'gpg --list-packets' on those files? It might shed some light. How large are the files (both input, and expected output sizes) ? Are the input files ASCII armor ("---BEGIN PGP MESSAGE---", etc) or binary? David From jason.harper at ge.com Fri Apr 2 16:32:54 2004 From: jason.harper at ge.com (Harper, Jason (GE Consumer Finance)) Date: Fri Apr 2 16:30:46 2004 Subject: GPG - Invalid packet Message-ID: <7730B3FE2184A5499A4A8601F598BB1B27902A@ALPMLVEM01.e2k.ad.ge.com> Jim, No ascii-armor was used. The encryption command that was used on the NAI system was: pgp --encrypt --user mypublickey --encrypt-to-self datafile.dat jason -----Original Message----- From: Jim Hendrick [mailto:jrhendri@maine.rr.com] Sent: Friday, April 02, 2004 9:20 AM To: Harper, Jason (GE Consumer Finance); gnupg-users@gnupg.org Subject: RE: GPG - Invalid packet do you know if it was encrypted with ascii-armor? I have seen problems with non-ascii systems sending files with wierd characters. Jim > -----Original Message----- > From: gnupg-users-bounces@gnupg.org > [mailto:gnupg-users-bounces@gnupg.org]On Behalf Of Harper, Jason (GE > Consumer Finance) > Sent: Friday, April 02, 2004 8:56 AM > To: gnupg-users@gnupg.org > Subject: GPG - Invalid packet > > > < reason] not subscribed on the correct e-mail address to post the first > time>> > > Dear Group, > > I've been using gpg to do file encryption with partners of ours, but > we just ran into two odd issues: > > 1. a GPG "don't know: invalid packet (ctb-1f) & (ctb-40) 2. a GPG > "problem reading source" > > I've done the google thing on this, but I'm not making much progress. > Does the list have any pointers for me to look at? We've not run into > this before. > > The source file is being encrypted, and compressed and signed by an > NAI > 7.11 Business PGP system, and we are opening on a Solaris 8 box with > gpg 1.2.3. I've sanitized some of the data to protect the innocent.. > > Sample one: > > gpg --output ce331 --decrypt CE_Con_1040326.dat > gpg: WARNING: using insecure memory! > gpg: please see http://www.gnupg.org/faq.html for more information > gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created > 2001-08-23 > "KEYNAME " > gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created > 2004-03-25 > "key@sample.com (COMPANY decription key) " > gpg: Signature made Fri Mar 26 14:20:29 2004 EST using DSA key ID > xxxxxxxx > gpg: Good signature from "COMPANY " > gpg: aka "COMPANY " > gpg: [don't know]: invalid packet (ctb=1f) > gpg: WARNING: message was not integrity protected > gpg: [don't know]: invalid packet (ctb=40) > > Then it just ends, but we end up with a plain text file. > Accuracy of the > file is unkown. What is the CTB=1f and CTB=40 > > Sample two: > > ->> gpg --output au331 --decrypt AU_Con_1040331.dat > gpg: WARNING: using insecure memory! > gpg: please see http://www.gnupg.org/faq.html for more information > gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created > 2001-08-23 > "COMPANY " > gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created > 2004-03-25 > "key@sample.com (COMPANY decription key) " > gpg: Signature made Fri Mar 26 14:20:46 2004 EST using DSA key ID > xxxxxxx > gpg: Good signature from "COMPANY " > gpg: aka "COMPANY " > File `au331' exists. Overwrite (y/N)? yes > gpg: Problem reading source (2386467048 bytes remaining) > gpg: handle plaintext failed: file read error > gpg: WARNING: message was not integrity protected > > Thanks for any and all assistance. > > jason > > > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users > From jason.harper at ge.com Fri Apr 2 16:56:41 2004 From: jason.harper at ge.com (Harper, Jason (GE Consumer Finance)) Date: Fri Apr 2 16:54:50 2004 Subject: GPG - Invalid packet Message-ID: <7730B3FE2184A5499A4A8601F598BB1B27902C@ALPMLVEM01.e2k.ad.ge.com> David, Here is the list-packets on the file that made up the first sample: gpg --list-packets CE_Con_1040326.dat gpg: WARNING: using insecure memory! gpg: please see http://www.gnupg.org/faq.html for more information :marker packet: 50 47 50 :pubkey enc packet: version 3, algo 16, keyid xxxxxxxxxxxxxxxxx data: [3072 bits] data: [3071 bits] :pubkey enc packet: version 3, algo 16, keyid xxxxxxxxxxxxxxxxx data: [1019 bits] data: [1023 bits] :encrypted data packet: length: 1549 gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxxx, created 2001-08-23 "Company " gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created 2004-03-25 "Company2 (Company2 decription key) " :compressed packet: algo=1 :onepass_sig packet: keyid xxxxxxxxxxxxxxxxxxx version 3, sigclass 00, digest 2, pubkey 17, last=1 :literal data packet: mode b, created 0, name="datadump20040109.dat", raw data: 0 bytes gpg: WARNING: message was not integrity protected The input files, prior to encyrption, are textual dumps of data. The CE file here, when we get it is 27kb. -----Original Message----- From: gnupg-users-bounces@gnupg.org [mailto:gnupg-users-bounces@gnupg.org] On Behalf Of David Shaw Sent: Friday, April 02, 2004 9:29 AM To: gnupg-users@gnupg.org Subject: Re: GPG - Invalid packet On Fri, Apr 02, 2004 at 08:56:10AM -0500, Harper, Jason (GE Consumer Finance) wrote: > I've been using gpg to do file encryption with partners of ours, but > we just ran into two odd issues: > > 1. a GPG "don't know: invalid packet (ctb-1f) & (ctb-40) 2. a GPG > "problem reading source" > > I've done the google thing on this, but I'm not making much progress. > Does the list have any pointers for me to look at? We've not run into > this before. > > The source file is being encrypted, and compressed and signed by an > NAI > 7.11 Business PGP system, and we are opening on a Solaris 8 box with > gpg 1.2.3. I've sanitized some of the data to protect the innocent.. Could you run 'gpg --list-packets' on those files? It might shed some light. How large are the files (both input, and expected output sizes) ? Are the input files ASCII armor ("---BEGIN PGP MESSAGE---", etc) or binary? David _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users From dshaw at jabberwocky.com Fri Apr 2 17:23:43 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 2 17:20:59 2004 Subject: GPG - Invalid packet In-Reply-To: <7730B3FE2184A5499A4A8601F598BB1B27902C@ALPMLVEM01.e2k.ad.ge.com> References: <7730B3FE2184A5499A4A8601F598BB1B27902C@ALPMLVEM01.e2k.ad.ge.com> Message-ID: <20040402152343.GA11892@jabberwocky.com> On Fri, Apr 02, 2004 at 09:56:41AM -0500, Harper, Jason (GE Consumer Finance) wrote: > David, > > Here is the list-packets on the file that made up the first sample: > > gpg --list-packets CE_Con_1040326.dat > gpg: WARNING: using insecure memory! > gpg: please see http://www.gnupg.org/faq.html for more information > :marker packet: > 50 47 50 > :pubkey enc packet: version 3, algo 16, keyid xxxxxxxxxxxxxxxxx > data: [3072 bits] > data: [3071 bits] > :pubkey enc packet: version 3, algo 16, keyid xxxxxxxxxxxxxxxxx > data: [1019 bits] > data: [1023 bits] > :encrypted data packet: > length: 1549 > gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxxx, created 2001-08-23 > "Company " > gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created 2004-03-25 > "Company2 (Company2 decription key) " > :compressed packet: algo=1 > :onepass_sig packet: keyid xxxxxxxxxxxxxxxxxxx > version 3, sigclass 00, digest 2, pubkey 17, last=1 > :literal data packet: > mode b, created 0, name="datadump20040109.dat", > raw data: 0 bytes > gpg: WARNING: message was not integrity protected > > The input files, prior to encyrption, are textual dumps of data. The CE > file here, when we get it is 27kb. Is it a binary or ASCII armor file? That is, if you read the encrypted file, does it begin with "----BEGIN PGP MESSAGE----" or is it binary gibberish? David From dshaw at jabberwocky.com Fri Apr 2 17:26:56 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 2 17:24:13 2004 Subject: GPG - Invalid packet In-Reply-To: <7730B3FE2184A5499A4A8601F598BB1B27902A@ALPMLVEM01.e2k.ad.ge.com> References: <7730B3FE2184A5499A4A8601F598BB1B27902A@ALPMLVEM01.e2k.ad.ge.com> Message-ID: <20040402152656.GB11892@jabberwocky.com> On Fri, Apr 02, 2004 at 09:32:54AM -0500, Harper, Jason (GE Consumer Finance) wrote: > Jim, > > No ascii-armor was used. The encryption command that was used on the NAI > system was: > > pgp --encrypt --user mypublickey --encrypt-to-self datafile.dat How was the file transferred to you? It it was FTP, make sure you didn't transfer it in text mode. That'll mangle the file in a way similar to what you posted. David From jason.harper at ge.com Fri Apr 2 17:46:31 2004 From: jason.harper at ge.com (Harper, Jason (GE Consumer Finance)) Date: Fri Apr 2 17:46:36 2004 Subject: GPG - Invalid packet Message-ID: <7730B3FE2184A5499A4A8601F598BB1B27902E@ALPMLVEM01.e2k.ad.ge.com> We get the file via a program called "Connect Direct" from Sterling Software, which I 'believe' to be a binary transfer program (will have to check). Also the file itself does not have the neat "----BEGIN PGP MESSAGE----". When I look at it in VIM it is total gibberish, but there is a noticeable "pgp" within the first 6 characters (e.g. "? PGP??N ????e?l? ) -----Original Message----- From: David Shaw [mailto:dshaw@jabberwocky.com] Sent: Friday, April 02, 2004 10:27 AM To: Harper, Jason (GE Consumer Finance) Cc: gnupg-users@gnupg.org Subject: Re: GPG - Invalid packet On Fri, Apr 02, 2004 at 09:32:54AM -0500, Harper, Jason (GE Consumer Finance) wrote: > Jim, > > No ascii-armor was used. The encryption command that was used on the > NAI system was: > > pgp --encrypt --user mypublickey --encrypt-to-self datafile.dat How was the file transferred to you? It it was FTP, make sure you didn't transfer it in text mode. That'll mangle the file in a way similar to what you posted. David From sbutler at fchn.com Fri Apr 2 17:49:43 2004 From: sbutler at fchn.com (Steve Butler) Date: Fri Apr 2 17:55:44 2004 Subject: GPG - Invalid packet Message-ID: <9A86613AB85FF346BB1321840DB42B4B046D42F3@jupiter.fchn.com> I had a similar problem with files coming from a mainframe environment. They threw some switch that made the file a multiple of their sector length. The trailing bytes became all hex zeros. I edited the file to remove the trailing bytes and GnuPG worked fine. The sender never did admit that the problem was theirs but subsequent files did not end at a sector boundary. See if the file size is a multiple of 256 bytes. The other thing that bites (or should I say bytes <>) is when somebody ships the file to us in ASCII mode without using the --ascii-armor. Since our FTP server is running on Linux, any CR/LF combinations are transformed into just the LF. That will stop the decryption with all sorts of strange results (depending on where the bytes were dropped). Right now none of our senders are using the --ascii-armor switch and we are not using it for our recipients. Been doing this for a couple of years. Once it is set up and automated (on both ends) then it works like a charm. But, everybody has to handle the file as binary. Some folks have a hard time grasping that concept -- especially one un-named vendor up here in the USA Pacific Northwest (not Billy boy). -----Original Message----- From: Harper, Jason (GE Consumer Finance) [mailto:jason.harper@ge.com] Sent: Friday, April 02, 2004 6:33 AM To: Jim Hendrick; gnupg-users@gnupg.org Subject: RE: GPG - Invalid packet Jim, No ascii-armor was used. The encryption command that was used on the NAI system was: pgp --encrypt --user mypublickey --encrypt-to-self datafile.dat jason -----Original Message----- From: Jim Hendrick [mailto:jrhendri@maine.rr.com] Sent: Friday, April 02, 2004 9:20 AM To: Harper, Jason (GE Consumer Finance); gnupg-users@gnupg.org Subject: RE: GPG - Invalid packet do you know if it was encrypted with ascii-armor? I have seen problems with non-ascii systems sending files with wierd characters. Jim > -----Original Message----- > From: gnupg-users-bounces@gnupg.org > [mailto:gnupg-users-bounces@gnupg.org]On Behalf Of Harper, Jason (GE > Consumer Finance) > Sent: Friday, April 02, 2004 8:56 AM > To: gnupg-users@gnupg.org > Subject: GPG - Invalid packet > > > < reason] not subscribed on the correct e-mail address to post the first > time>> > > Dear Group, > > I've been using gpg to do file encryption with partners of ours, but > we just ran into two odd issues: > > 1. a GPG "don't know: invalid packet (ctb-1f) & (ctb-40) 2. a GPG > "problem reading source" > > I've done the google thing on this, but I'm not making much progress. > Does the list have any pointers for me to look at? We've not run into > this before. > > The source file is being encrypted, and compressed and signed by an > NAI > 7.11 Business PGP system, and we are opening on a Solaris 8 box with > gpg 1.2.3. I've sanitized some of the data to protect the innocent.. > > Sample one: > > gpg --output ce331 --decrypt CE_Con_1040326.dat > gpg: WARNING: using insecure memory! > gpg: please see http://www.gnupg.org/faq.html for more information > gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created > 2001-08-23 > "KEYNAME " > gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created > 2004-03-25 > "key@sample.com (COMPANY decription key) " > gpg: Signature made Fri Mar 26 14:20:29 2004 EST using DSA key ID > xxxxxxxx > gpg: Good signature from "COMPANY " > gpg: aka "COMPANY " > gpg: [don't know]: invalid packet (ctb=1f) > gpg: WARNING: message was not integrity protected > gpg: [don't know]: invalid packet (ctb=40) > > Then it just ends, but we end up with a plain text file. > Accuracy of the > file is unkown. What is the CTB=1f and CTB=40 > > Sample two: > > ->> gpg --output au331 --decrypt AU_Con_1040331.dat > gpg: WARNING: using insecure memory! > gpg: please see http://www.gnupg.org/faq.html for more information > gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created > 2001-08-23 > "COMPANY " > gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created > 2004-03-25 > "key@sample.com (COMPANY decription key) " > gpg: Signature made Fri Mar 26 14:20:46 2004 EST using DSA key ID > xxxxxxx > gpg: Good signature from "COMPANY " > gpg: aka "COMPANY " > File `au331' exists. Overwrite (y/N)? yes > gpg: Problem reading source (2386467048 bytes remaining) > gpg: handle plaintext failed: file read error > gpg: WARNING: message was not integrity protected > > Thanks for any and all assistance. > > jason > > > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users > _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. From dshaw at jabberwocky.com Fri Apr 2 18:47:52 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 2 18:45:12 2004 Subject: GPG - Invalid packet In-Reply-To: <7730B3FE2184A5499A4A8601F598BB1B27902E@ALPMLVEM01.e2k.ad.ge.com> References: <7730B3FE2184A5499A4A8601F598BB1B27902E@ALPMLVEM01.e2k.ad.ge.com> Message-ID: <20040402164752.GC11892@jabberwocky.com> On Fri, Apr 02, 2004 at 10:46:31AM -0500, Harper, Jason (GE Consumer Finance) wrote: > We get the file via a program called "Connect Direct" from Sterling > Software, which I 'believe' to be a binary transfer program (will > have to check). Also the file itself does not have the neat > "----BEGIN PGP MESSAGE----". When I look at it in VIM it is total > gibberish, but there is a noticeable "pgp" within the first 6 > characters (e.g. "?? PGP??N ?????e?l? ) That's normal for a binary message. So, it could be binary being treated as text somewhere. This is a really common problem. An easy check is to send the file you received back to the sender. If it doesn't match, then it got mangled by the transfer. Another thing to try is to get the sender to turn on ASCII armor. If your transfer program wants text, give it text :) David From vedaal at hush.com Fri Apr 2 19:26:52 2004 From: vedaal at hush.com (vedaal@hush.com) Date: Fri Apr 2 19:24:04 2004 Subject: GPG - Invalid packet Message-ID: <200404021726.i32HQqiL042314@mailserver2.hushmail.com> >Message: 3 >Date: Fri, 2 Apr 2004 09:56:41 -0500 >From: "Harper, Jason \(GE Consumer Finance\)" >Subject: RE: GPG - Invalid packet >To: "David Shaw" , >Message-ID: > <7730B3FE2184A5499A4A8601F598BB1B27902C@ALPMLVEM01.e2k.ad.ge.com >Here is the list-packets on the file that made up the first sample: >:pubkey enc packet: version 3, algo 16, keyid xxxxxxxxxxxxxxxxx > data: [3072 bits] > data: [3071 bits] >:pubkey enc packet: version 3, algo 16, keyid xxxxxxxxxxxxxxxxx > data: [1019 bits] > data: [1023 bits] >:encrypted data packet: > length: 1549 >gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxxx, created 2001- >08-23 > "Company " >gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created 2004- >03-25 > "Company2 (Company2 decription key) " the encryption used an ADK does gnupg support decryption of a message when an adk is used, or does it not recognize such keys? vedaal Concerned about your privacy? Follow this link to get FREE encrypted email: https://www.hushmail.com/?l=2 Free, ultra-private instant messaging with Hush Messenger https://www.hushmail.com/services.php?subloc=messenger&l=434 Promote security and make money with the Hushmail Affiliate Program: https://www.hushmail.com/about.php?subloc=affiliate&l=427 From jason.harper at ge.com Fri Apr 2 20:28:54 2004 From: jason.harper at ge.com (Harper, Jason (GE Consumer Finance)) Date: Fri Apr 2 20:26:40 2004 Subject: GPG - Invalid packet Message-ID: <7730B3FE2184A5499A4A8601F598BB1B279034@ALPMLVEM01.e2k.ad.ge.com> We're going to attempt the ascii-armor now. We're also going to ask the sender to retrieve the file and see if he can decrypt it. -----Original Message----- From: David Shaw [mailto:dshaw@jabberwocky.com] Sent: Friday, April 02, 2004 11:48 AM To: Harper, Jason (GE Consumer Finance) Cc: gnupg-users@gnupg.org Subject: Re: GPG - Invalid packet On Fri, Apr 02, 2004 at 10:46:31AM -0500, Harper, Jason (GE Consumer Finance) wrote: > We get the file via a program called "Connect Direct" from Sterling > Software, which I 'believe' to be a binary transfer program (will have > to check). Also the file itself does not have the neat "----BEGIN PGP > MESSAGE----". When I look at it in VIM it is total gibberish, but > there is a noticeable "pgp" within the first 6 characters (e.g. "?? > PGP??N ?????e?l? ) That's normal for a binary message. So, it could be binary being treated as text somewhere. This is a really common problem. An easy check is to send the file you received back to the sender. If it doesn't match, then it got mangled by the transfer. Another thing to try is to get the sender to turn on ASCII armor. If your transfer program wants text, give it text :) David From jason.harper at ge.com Fri Apr 2 21:35:45 2004 From: jason.harper at ge.com (Harper, Jason (GE Consumer Finance)) Date: Fri Apr 2 21:33:44 2004 Subject: GPG - Invalid packet Message-ID: <7730B3FE2184A5499A4A8601F598BB1B279038@ALPMLVEM01.e2k.ad.ge.com> There were a total of three files, and the two we just tested with ASCII-Armor turned on have decrypted perfectly. Thank you all very much for your assistance. jason -----Original Message----- From: Harper, Jason (GE Consumer Finance) Sent: Friday, April 02, 2004 1:29 PM To: 'David Shaw' Cc: gnupg-users@gnupg.org Subject: RE: GPG - Invalid packet We're going to attempt the ascii-armor now. We're also going to ask the sender to retrieve the file and see if he can decrypt it. -----Original Message----- From: David Shaw [mailto:dshaw@jabberwocky.com] Sent: Friday, April 02, 2004 11:48 AM To: Harper, Jason (GE Consumer Finance) Cc: gnupg-users@gnupg.org Subject: Re: GPG - Invalid packet On Fri, Apr 02, 2004 at 10:46:31AM -0500, Harper, Jason (GE Consumer Finance) wrote: > We get the file via a program called "Connect Direct" from Sterling > Software, which I 'believe' to be a binary transfer program (will have > to check). Also the file itself does not have the neat "----BEGIN PGP > MESSAGE----". When I look at it in VIM it is total gibberish, but > there is a noticeable "pgp" within the first 6 characters (e.g. "?? > PGP??N ?????e?l? ) That's normal for a binary message. So, it could be binary being treated as text somewhere. This is a really common problem. An easy check is to send the file you received back to the sender. If it doesn't match, then it got mangled by the transfer. Another thing to try is to get the sender to turn on ASCII armor. If your transfer program wants text, give it text :) David From andreas.bergen at in-jesus.de Fri Apr 2 22:02:49 2004 From: andreas.bergen at in-jesus.de (Andreas Bergen) Date: Fri Apr 2 23:24:49 2004 Subject: hierarchical keys? Encryption to two subkeys impossible Message-ID: <200404022202.49953.andreas.bergen@in-jesus.de> Thanks. it works. -- Andreas Bergen E-Mail: andreas dot bergen at in-jesus dot de PGP/GnuPG-encrypted / -signed Email welcome. PGP-key-ID: 8CDEC18F Gott ist Liebe, und wer in der Liebe bleibt, bleibt in Gott und Gott in ihm. From jdbeyer at exit109.com Sun Apr 4 21:43:40 2004 From: jdbeyer at exit109.com (Jean-David Beyer) Date: Sun Apr 4 21:41:01 2004 Subject: GnuCash gnucash-1.8.8-0.9 under RHEL 3 ES Message-ID: <4070656C.7040904@exit109.com> With lots of help from people on this board, I finally have it running without segmentation faults, etc.. I.e., it seems to work as well as whatever came with Red Hat Linux 7.3. Basically, I had to install a bunch of libraries that came with RHEL 3 ES but do not load during the installation procedures. Then I had to supplement them with a bunch from Red Hat Linux 9. Then I had to get one or two from Fedora, and one from rpmfind that I could not find anywhere else. Mark Eackloff sent me this list: You need Guppi-0.40.3-13 from RH9. You need bonobo-1.0.22-4 from RH9 which requires oaf-0.5.1 (0.6.10-5 from RH9 works) and gnome-print-0.37.4 from RH9 which requires libgnomeprint15-0.37.4-1 from RH9 You need g-wrap-1.3.4-3 from RH9 which requires guile-1.6.4-8.2 from RH9.03 Beta Updates (guile-1.6.4-8 from RHEL3) 1.6.4-8 will allow you to install the gnucash rpm but you will get run time errors on startup. You need gtkhtml-1.1.9-0.9.1 from RH9 which requires libcapplet0-1.4.0.1-11.1 from RH9 and GConf-1.0.9-10 from RH9 You need libgal21-0.23-1 from RH9 which requires gal-0.23-1 from RH9 I had trouble with some, as I said in my reply to him: The ones I had trouble with were: libgnomeprint15-0.37.4-1 from RH9: On my RHL9 Disk #1 I found only libgnomeprint15-0.37-4, but it worked. I could not find the version you list using Google. gtkhtml-1.1.9-0.9.1 from RH9: On my RHL9 Disk #1, I found only gtkhtml-1.1.8-5. I found one using Google, but IIIRC, it was from Fedora, not RHL9. guile-1.6.4-8.2 from RH9.03 Beta Updates: I could not find "Beta Updates", but I found this as part of Fedora's official release 1. libcapplet0-1.4.0.1-11.1: This is not on my RHL9 Disk #1 either, but libcapplet0-1.4.0.1-11 is, so I used that. -- .~. Jean-David Beyer Registered Linux User 85642. /V\ Registered Machine 241939. /( )\ Shrewsbury, New Jersey http://counter.li.org ^^-^^ 15:35:00 up 3 days, 11:50, 2 users, load average: 3.12, 2.95, 2.40 From jdbeyer at exit109.com Sun Apr 4 23:39:46 2004 From: jdbeyer at exit109.com (Jean-David Beyer) Date: Sun Apr 4 23:36:59 2004 Subject: GnuCash gnucash-1.8.8-0.9 under RHEL 3 ES In-Reply-To: <4070656C.7040904@exit109.com> References: <4070656C.7040904@exit109.com> Message-ID: <407080A2.8070203@exit109.com> Jean-David Beyer wrote: > With lots of help from people on this board, I finally have it running > without segmentation faults, etc.. I.e., it seems to work as well as > whatever came with Red Hat Linux 7.3. > [snip] Sorry I posted that to the wrong list. -- .~. Jean-David Beyer Registered Linux User 85642. /V\ Registered Machine 241939. /( )\ Shrewsbury, New Jersey http://counter.li.org ^^-^^ 17:35:00 up 3 days, 13:50, 5 users, load average: 4.28, 4.22, 4.13 From jharris at widomaker.com Mon Apr 5 00:48:13 2004 From: jharris at widomaker.com (Jason Harris) Date: Mon Apr 5 00:45:40 2004 Subject: new (2004-04-04) keyanalyze results Message-ID: <20040404224812.GU10980@pm1.ric-41.lft.widomaker.com> New keyanalyze results are available at: http://keyserver.kjsl.com/~jharris/ka/2004-04-04/ Earlier reports are also available, for comparison: http://keyserver.kjsl.com/~jharris/ka/ Even earlier monthly reports are at: http://dtype.org/keyanalyze/ SHA-1 hashes and sizes for all the "permanent" files: e4dccd596e4470c987d7e42fd63da520fc07163f 13657410 preprocess.keys a30e7f5977e76abf699d5d76c249680ed2638424 10627270 othersets.txt 3d4bad9e80187179919233592d0aac29db5ac46b 2482816 msd-sorted.txt b0f152cbac2bff77aeed70a933fec6d7ac3e7b71 1484 index.html 7579107eceefb9929604451fb248f98ac7917389 2289 keyring_stats 5c5c311b60f7f9bf32be1fbfd2a3435560601c51 984057 msd-sorted.txt.bz2 9217717c8ba186dc9a10b103e7239b8b94a595ac 26 other.txt 87385ffeb2104877038e920960787a18d2ca6936 2073226 othersets.txt.bz2 fe15c161a4555152ccca4851e59486bb7f83a7e7 6102839 preprocess.keys.bz2 75e39091bdf06a38801f2517242ccea3f453c8e3 13396 status.txt b8cb20674074238abd239ddd6ac3b78385846564 212070 top1000table.html 75a92032dd27734c38d839163d6c89d0665a98d1 30656 top1000table.html.gz 12df748fc3254efa99a39a25c91f7969b4354fa3 11072 top50table.html 794d5e72d797a15ffc0b86857d820930f1518124 2444 D3/D39DA0E3 -- Jason Harris | NIC: JH329, PGP: This _is_ PGP-signed, isn't it? jharris@widomaker.com _|_ web: http://keyserver.kjsl.com/~jharris/ Got photons? (TM), (C) 2004 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: not available Url : /pipermail/attachments/20040404/afb1dbdd/attachment.bin From wk at gnupg.org Mon Apr 5 11:32:04 2004 From: wk at gnupg.org (Werner Koch) Date: Mon Apr 5 11:17:01 2004 Subject: GPG - Invalid packet In-Reply-To: <200404021726.i32HQqiL042314@mailserver2.hushmail.com> (vedaal@hush.com's message of "Fri, 2 Apr 2004 09:26:52 -0800") References: <200404021726.i32HQqiL042314@mailserver2.hushmail.com> Message-ID: <87ad1qdb3f.fsf@vigenere.g10code.de> On Fri, 2 Apr 2004 09:26:52 -0800, said: >> gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxxx, created 2001- >> 08-23 >> "Company " >> gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created 2004- >> 03-25 >> "Company2 (Company2 decription key) " > the encryption used an ADK I can't see that from the output. It is a pretty standard messages encrypted to 2 recipients. An ARR (Additional Recipient Request) is probably what you mean with ADK; this is a request to the program (included in a self-signature) to silently add another recipient. GnuPG ignores such requests. Werner From atom-gpg at suspicious.org Wed Apr 7 00:16:57 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 7 00:14:24 2004 Subject: openPGP vs x509 In-Reply-To: References: Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 is there any development in working towards an openPGP ~type~ of framework for web sites (HTTPS) as an alternative to the x509 standard? one shouldn't have to pay for trust ;) ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Politics would be a helluva good business if it weren't for the goddamned people." -- Richard M. Nixon -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAcyxhnCgLvz19QeMRAmw5AJ9Itb/UUKD0Q16v+kGlerT5Z8wkXgCgide3 QWaC3FIPEDQbB4mQ+mueezI= =f2CK -----END PGP SIGNATURE----- From JPClizbe at comcast.net Wed Apr 7 05:33:54 2004 From: JPClizbe at comcast.net (John Clizbe) Date: Wed Apr 7 05:31:38 2004 Subject: openPGP vs x509 In-Reply-To: References: Message-ID: <407376A2.2030408@comcast.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Atom 'Smasher' wrote: > is there any development in working towards an openPGP ~type~ of framework > for web sites (HTTPS) as an alternative to the x509 standard? > > one shouldn't have to pay for trust ;) Check CACert.org. - -- John P. Clizbe Inet: JPClizbe(a)comcast DOT nyet Golden Bear Networks PGP/GPG KeyID: 0x608D2A10 "Most men take the straight and narrow. A few take the road less traveled. I chose to cut through the woods." -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5rc1 (Windows 2000 SP4) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFAc3ahHQSsSmCNKhARAlHzAKCIblnfpBDav2o+v+xb+RFXuhEC4ACg21gv zJ9syl1bSYAG42rz4/dOMzw= =Y/GS -----END PGP SIGNATURE----- From jason at monkeypr0n.org Wed Apr 7 07:12:48 2004 From: jason at monkeypr0n.org (Jason Burnett) Date: Wed Apr 7 08:10:05 2004 Subject: Implementation questions Message-ID: <20040407051248.GA16639@monkeypr0n.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I am working on getting our department and hopefully company converted to using secure communications for email and sensitive document sharing, this is the easy part. The question has been brought up concerning storing secure information (encrypted) on an internal server we have and allowing multiple personnel to be able to decrypt that info. Can you encrypt a file against multiple keys? Is having a departmental or company key that everyone has a good idea (does not seem like it would be)? I am just looking for some suggestions on how to implement this or some examples of how this has been done by members of the list. Thanks in advance. - -- jason@monkeypr0n.org -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAc43QlNj4xMBh32QRAkn7AJ0aOpAfXWQMHQ/FnJsE84nxDtyuJwCff8qc rWIGiVFhwbTNf7Fgj+b+deA= =ufN4 -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Wed Apr 7 08:42:40 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 7 08:39:55 2004 Subject: openPGP vs x509 In-Reply-To: <407376A2.2030408@comcast.net> References: <407376A2.2030408@comcast.net> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Check CACert.org. =================================== it's still based on the x509 PKI (trust us: single point of failure), which i don't like as much as the openPGP PKI (web of trust: tunable to the paranoia and needs of the end user). it doesn't seem hard (in theory) to implement a (user-friendly!) openPGP type of PKI into web browsers... one could import PGP keys into their browser (or just leave the default verisign/thawte/etc keys), assign levels of trust to those keys, and assuming that the user-defined (or default) trust settings are met, everything proceeds as normal... except of course that one's PGP key could be used to sign a web site's certificate and "trust" doesn't have to be issued from a central point. really, it could be done in a way that's invisible to the 99% of end users who don't know or care about crypto/keys/certs (like it is now), but could open a lot of possibilities for people who want to hack around and experiment, and not rely on a single point of failure. of course, there'd still be a need for CAs: some web sites will still prefer to buy their "trust" from a "trusted" source, rather than enter into a "web" of trust (corporate mentality)... and those are the companies that don't mind paying the verisign tax. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "There can be no greater good than the quest for peace, and no finer purpose than the preservation of freedom." -- U.S. President Ronald Reagan -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAc6LlnCgLvz19QeMRAkQgAJwICavEv7uMaz2aC7zSFPudkZfh/ACaAsiN OYt+fOEF631PhuMV4+mzqFI= =CMsk -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Wed Apr 7 09:07:47 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 7 09:05:03 2004 Subject: Implementation questions In-Reply-To: <20040407051248.GA16639@monkeypr0n.org> References: <20040407051248.GA16639@monkeypr0n.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > The question has been brought up concerning storing secure information > (encrypted) on an internal server we have and allowing multiple > personnel to be able to decrypt that info. Can you encrypt a file > against multiple keys? Is having a departmental or company key that > everyone has a good idea (does not seem like it would be)? I am just > looking for some suggestions on how to implement this or some examples > of how this has been done by members of the list. ==================================== PGP/GPG will let you encrypt a message to any number of users at once. last i checked, MIT-PGP (tm) had a feature that's sort of like a key-escrow, so if someone leaves the company, email encrypted to them can be decrypted with the "other" key. GnuPG doesn't support that, but in some workplaces you can do almost the same thing by adding a "recipient" line to a user's config file. keys used by more than one individual person are almost always a bad idea, but you'll really have to assess the security risks, concerns, threats, etc to get a picture of what level of security you want/need and what applications (PGP, symmetric encryption, file-system encryption, etc) will help you get there. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarrely inexplicable. There is another theory which states that this has already happened." -- Douglas Adams -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAc6jInCgLvz19QeMRAqNhAJ9ujvfTQgS285StRDZpC4PnHIJ6ngCdFSCr zLKn7R0jPqWUDFkC8+YhjPE= =xhqZ -----END PGP SIGNATURE----- From wk at gnupg.org Wed Apr 7 09:28:10 2004 From: wk at gnupg.org (Werner Koch) Date: Wed Apr 7 09:12:01 2004 Subject: openPGP vs x509 In-Reply-To: (atom-gpg@suspicious.org's message of "Tue, 6 Apr 2004 18:16:57 -0400 (EDT)") References: Message-ID: <871xn09rhx.fsf@vigenere.g10code.de> On Tue, 6 Apr 2004 18:16:57 -0400 (EDT), Atom 'Smasher' said: > is there any development in working towards an openPGP ~type~ of framework > for web sites (HTTPS) as an alternative to the x509 standard? GNUTLS support OpenPGP keys, however I am not aware whether servers or clients already make use of it. Werner From jason at monkeypr0n.org Wed Apr 7 09:44:32 2004 From: jason at monkeypr0n.org (Jason Burnett) Date: Wed Apr 7 09:41:45 2004 Subject: Implementation questions In-Reply-To: References: <20040407051248.GA16639@monkeypr0n.org> Message-ID: <20040407074432.GA14995@monkeypr0n.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Thanks for the reply, the email being encrypted was not a huge concern, we would only use that for corporate communication and if the recipient left then the sender would know what was in the email. We were more looking along the lines of storing customer info (passwords) where multiple sysadmins could access/decrypt the info and the scenario our suit likes to use "What if your all hit by a bus at the same time?" he would be able to decrypt the passwords and pass them on to the replacment team. Once a file is encrypted to multiple keys is there a way to remove one of the keys from being able to decrypt it? Sort of like revoking a key? on Wed Apr 07 Atom 'Smasher' spoke forth with the blessed manuscript > > The question has been brought up concerning storing secure information > > (encrypted) on an internal server we have and allowing multiple > > personnel to be able to decrypt that info. Can you encrypt a file > > against multiple keys? Is having a departmental or company key that > > everyone has a good idea (does not seem like it would be)? I am just > > looking for some suggestions on how to implement this or some examples > > of how this has been done by members of the list. > ==================================== > PGP/GPG will let you encrypt a message to any number of users at once. > last i checked, MIT-PGP (tm) had a feature that's sort of like a > key-escrow, so if someone leaves the company, email encrypted to them can > be decrypted with the "other" key. GnuPG doesn't support that, but in some > workplaces you can do almost the same thing by adding a "recipient" line > to a user's config file. > keys used by more than one individual person are almost always a bad idea, > but you'll really have to assess the security risks, concerns, threats, > etc to get a picture of what level of security you want/need and what > applications (PGP, symmetric encryption, file-system encryption, etc) will > help you get there. > ...atom > _________________________________________ > PGP key - http://atom.smasher.org/pgp.txt > 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 > ------------------------------------------------- > "There is a theory which states that if ever anyone discovers > exactly what the Universe is for and why it is here, it will > instantly disappear and be replaced by something even more > bizarrely inexplicable. There is another theory which states > that this has already happened." > -- Douglas Adams > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users - -- PGP key - http://www.monkeypr0n.org/keyring/jason@monkeypr0n 793A 39E7 5655 0415 5C5E 976E 94D8 F8C4 C061 DF64 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAc7FglNj4xMBh32QRAoH0AJ9dHF3qqeCxcDEY46WZ6ZRCyPIzXwCfSq4z 0zR0sS1FEzpaVDP3udAV2wM= =QHHp -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Wed Apr 7 10:07:41 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 7 10:04:58 2004 Subject: openPGP vs x509 In-Reply-To: References: <407376A2.2030408@comcast.net> Message-ID: <200404071007.43883@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wednesday 07 April 2004 08.42, Atom 'Smasher' wrote: > > Check CACert.org. > > =================================== > > it's still based on the x509 PKI (trust us: single point of failure), > which i don't like as much as the openPGP PKI (web of trust: tunable > to the paranoia and needs of the end user). It's not hard to set up a CA of your own, and thus implementing a PGP-like trust structure in the X509 world (at least, it's much easier than to convert all x509-using TLS/SSL aware web browsers to PGP/TLS.) The hard part is to get people to use it. All in all, I think X.509 vs. PGP is mostly just a question of how to encode a public key with associated ownership information. Currently it seems more complicated to implement peer to peer trust models in X.509, but I'm convinced that this is only a user interface thing, and not intrinsic to the technology. (But then, I'm no X.509 expert at all, so I wouldn't know.) > it doesn't seem hard (in theory) to implement a (user-friendly!) > openPGP type of PKI into web browsers. do it. cheers - -- vbi - -- featured link: http://fortytwo.ch/smtp -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkBzts5gGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6S1gAn1XTydlSzRyDw2aACaR7p9Hi ZXl+AKCzhaPHRMjffrMzMv2axDdZ+mkMUQ== =m3QZ -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Wed Apr 7 10:13:31 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 7 10:10:50 2004 Subject: Implementation questions In-Reply-To: <20040407074432.GA14995@monkeypr0n.org> References: <20040407051248.GA16639@monkeypr0n.org> <20040407074432.GA14995@monkeypr0n.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Thanks for the reply, the email being encrypted was not a huge concern, > we would only use that for corporate communication and if the recipient > left then the sender would know what was in the email. ================ experiment: 1) install my public key 2) $ chmod 0 ~/.gnupg/gpg.conf 3) $ ps | gpg -ear 0x3D7D41E3 you're the sender, i'm the recipient. as the sender of that message, see if you can decrypt it... i'll be *very* impressed if you can. my point, here, is that you can't always count on the sender encrypting to themself. and what if the sender and recipient both get hit by a bus? > We were more looking along the lines of storing customer info > (passwords) where multiple sysadmins could access/decrypt the info and ================ data can be ENcrypted on auto-pilot to as many keys as you want, and sent to a file, database, etc.... > the scenario our suit likes to use "What if your all hit by a bus at the > same time?" he would be able to decrypt the passwords and pass them on > to the replacment team. ================ one of the encryption keys could belong to the boss... they never plan on getting hit by a bus. also, you might want to print out your private key, write out your secret password on it, put it in an envelope, and store that in a locked safe, under control of the boss. when all of the techs get hit by a meteorite, the new techs can recover the keys from the safe. > Once a file is encrypted to multiple keys is there a way to remove one > of the keys from being able to decrypt it? Sort of like revoking a key? ================ that's tough.... i can't think of a way to do that. even if you could "revoke" the key, let's say one of the employees emails the database to himself, and then gets fired: he's home with a copy of the db and his secret key. the best thing i can think of (at 0400 hrs) to deal with that, is to have a file/db encrypted to each employee. if an employee leaves, that file/db can be deleted.... but it won't do any good to delete it if they already saved their own copy ;) ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "I hope we shall crush in its birth the aristocracy of our monied corporations which dare already to challenge our government to a trial of strength, and bid defiance to the laws our country." -- Thomas Jefferson, 1816 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAc7gwnCgLvz19QeMRAmJLAKClBBjNsmNTjTc7uBcJ//se6a0K6QCeJie4 tRhmRLEz27VK7xJiruqXVQ4= =r8cs -----END PGP SIGNATURE----- From kai.klesatschke at gmx.de Wed Apr 7 10:29:31 2004 From: kai.klesatschke at gmx.de (Kai Klesatschke) Date: Wed Apr 7 10:27:19 2004 Subject: openPGP vs x509 In-Reply-To: References: Message-ID: <4073BBEB.6070203@gmx.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 For which reasons do you want to use X509 for SSL connections? For the ca of our university I did an inquiry about this issue and didn't find any CA that certifies OpenPGP keys for Servers. But there are many uncommercial CAs which certifies without any costs. My opninion is, that the hierarchical X509 structur is more efficient because not every client has to know about a special certificate. On the other side, I have no idea how this should work with pgp-keys in reality. To encrypt a connection normaly the opponents publickey is used to encrypt and the secretkey is used to decrypt. This means, that a server have to know all public keys of clients connecting to it. A handshake between the server an client maybe solve this prob. Atom 'Smasher' wrote: | is there any development in working towards an openPGP ~type~ of framework | for web sites (HTTPS) as an alternative to the x509 standard? | | one shouldn't have to pay for trust ;) | | | | ...atom | | _________________________________________ | PGP key - http://atom.smasher.org/pgp.txt | 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 | ------------------------------------------------- | | "Politics would be a helluva good business | if it weren't for the goddamned people." | -- Richard M. Nixon -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (MingW32) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFAc7vrlJVV36J3GRMRApe8AJ9Mm+6E9+oXnne95+AN11oYU8SI6QCggB8N iO2dxPgi9Nv5Xk7vJC1N7/Q= =LPbu -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Wed Apr 7 10:48:06 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 7 10:45:16 2004 Subject: openPGP vs x509 Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > My opninion is, that the hierarchical X509 structur is more efficient > because not every client has to know about a special certificate. On the > other side, ====================== i would argue that it's less efficient, largely because there's a single point of failure (the CA). > I have no idea how this should work with pgp-keys in reality. ======================== here's how i picture it: let's say you connect to https://my-server.com and that certificate is signed with my PGP key. if you have my key "installed" in your browser, and marked as "trusted", then you get a secure connection. otherwise, you get a pop-up box asking you if you'd like to accept, examine or discard the certificate. > To encrypt a connection normaly the opponents publickey is used > to encrypt and the secretkey is used to decrypt. This means, that a > server have to know all public keys of clients connecting to it. A > handshake between the server an client maybe solve this prob. ======================= my understanding of SSL/TLS is that the client can authenticate the server, but the server has no way to authenticate the client (via SSL/TLS). in practice, i connect to https://paypal.com and i want to be VERY sure that i'm connecting to the correct server. the server has no need to authenticate that the key on my end ~really~ belongs to me... that's what my login credentials (username/password) are for. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "You have just dined, and however scrupulously the slaughterhouse is concealed in the graceful distance of miles, there is complicity." -- Ralph Waldo Emerson, 1870 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAc7/mnCgLvz19QeMRAonWAJ9N4+YlIfrmw5NHBoCpzjOmjrmMUwCdFEqp 3YD9lG8pmvXsZNw9XAbVRmY= =Ky8C -----END PGP SIGNATURE----- From jason at monkeypr0n.org Wed Apr 7 10:54:39 2004 From: jason at monkeypr0n.org (Jason Burnett) Date: Wed Apr 7 10:51:50 2004 Subject: Implementation questions In-Reply-To: References: <20040407051248.GA16639@monkeypr0n.org> <20040407074432.GA14995@monkeypr0n.org> Message-ID: <20040407085439.GA30640@monkeypr0n.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 on Wed Apr 07 Atom 'Smasher' spoke forth with the blessed manuscript > > Thanks for the reply, the email being encrypted was not a huge concern, > > we would only use that for corporate communication and if the recipient > > left then the sender would know what was in the email. > ================ > experiment: > 1) install my public key > 2) $ chmod 0 ~/.gnupg/gpg.conf > 3) $ ps | gpg -ear 0x3D7D41E3 > you're the sender, i'm the recipient. as the sender of that message, see > if you can decrypt it... i'll be *very* impressed if you can. > my point, here, is that you can't always count on the sender encrypting to > themself. and what if the sender and recipient both get hit by a bus? You are correct, but hopefully we would be able to educate the users enough to perform this function. > > We were more looking along the lines of storing customer info > > (passwords) where multiple sysadmins could access/decrypt the info and > ================ > data can be ENcrypted on auto-pilot to as many keys as you want, and sent > to a file, database, etc.... I can't seem to find any docs on encrypting to multiple keys, I encrypt my important files all the time, do you have a URL or some info on how to encrypt to multiple keys? or am I overanalyzing this function? I have been up all night trying to rebuild a crashed server, so overlooking simple info is not out of the question right now. > > the scenario our suit likes to use "What if your all hit by a bus at the > > same time?" he would be able to decrypt the passwords and pass them on > > to the replacment team. > ================ > one of the encryption keys could belong to the boss... they never plan on > getting hit by a bus. also, you might want to print out your private key, > write out your secret password on it, put it in an envelope, and store > that in a locked safe, under control of the boss. when all of the techs > get hit by a meteorite, the new techs can recover the keys from the safe. > > Once a file is encrypted to multiple keys is there a way to remove one > > of the keys from being able to decrypt it? Sort of like revoking a key? > ================ > that's tough.... i can't think of a way to do that. even if you could > "revoke" the key, let's say one of the employees emails the database to > himself, and then gets fired: he's home with a copy of the db and his > secret key. This is something we will not be able to overcome, the employees need access to the information and could make copies of it before they let us know they are leaving. We have to trust that our HR department performs proper background checks before hiring someone (they said they did). > the best thing i can think of (at 0400 hrs) to deal with that, is to have > a file/db encrypted to each employee. if an employee leaves, that file/db > can be deleted.... but it won't do any good to delete it if they already > saved their own copy ;) Our main goal is to make sure that, as long our employees maintain proper personal security, that the data is fairly safe from prying eyes. I appreciate the help and I feel your pain about trying to think straight at 0400, I really didnt expect a response to this tonight, nice surprise, something for me to do while harddrives format. > ...atom > _________________________________________ > PGP key - http://atom.smasher.org/pgp.txt > 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 > ------------------------------------------------- > "I hope we shall crush in its birth the aristocracy of our > monied corporations which dare already to challenge our > government to a trial of strength, and bid defiance to > the laws our country." > -- Thomas Jefferson, 1816 > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users - -- PGP key - http://www.monkeypr0n.org/keyring/jason@monkeypr0n 793A 39E7 5655 0415 5C5E 976E 94D8 F8C4 C061 DF64 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAc8HPlNj4xMBh32QRAtpRAJ0fgP145uQXhBIchcs8O9f0xvADaACcC5mA t8xl0Q7CRF3Jm7GaxxpDOFs= =Iu/k -----END PGP SIGNATURE----- From wk at gnupg.org Wed Apr 7 13:04:37 2004 From: wk at gnupg.org (Werner Koch) Date: Wed Apr 7 12:46:53 2004 Subject: openPGP vs x509 In-Reply-To: (atom-gpg@suspicious.org's message of "Wed, 7 Apr 2004 04:48:06 -0400 (EDT)") References: Message-ID: <87lll882wq.fsf@vigenere.g10code.de> On Wed, 7 Apr 2004 04:48:06 -0400 (EDT), Atom 'Smasher' said: > my understanding of SSL/TLS is that the client can authenticate the > server, but the server has no way to authenticate the client (via > SSL/TLS). TLS provides such a mechanism but sane users don't hand their money over to Verisign for a user certificate. I don't know whether browsers support this at all. Werner From Holger.Sesterhenn at smgwtest.aachen.utimaco.de Wed Apr 7 13:22:56 2004 From: Holger.Sesterhenn at smgwtest.aachen.utimaco.de (Holger Sesterhenn) Date: Wed Apr 7 13:20:56 2004 Subject: TLS client authentication, Re: openPGP vs x509 In-Reply-To: <87lll882wq.fsf@vigenere.g10code.de> References: <87lll882wq.fsf@vigenere.g10code.de> Message-ID: <4073E490.7040001@smgwtest.aachen.utimaco.de> Hello, Werner Koch wrote: > TLS provides such a mechanism but sane users don't hand their money > over to Verisign for a user certificate. I don't know whether > browsers support this at all. It's called 'client authentication' and you have to import this special certifacte into your browser. IE, NS and Mozilla do support such behaviour since ages. I'am working with such a configuration every day. Of course we have created our own certificate hierachy. No need for Verisign. Best Regards, Holger Sesterhenn --- Internet http://www.utimaco.com From gnupg at dossen.dk Wed Apr 7 14:07:30 2004 From: gnupg at dossen.dk (Mads Laursen) Date: Wed Apr 7 14:04:46 2004 Subject: TLS client authentication, Re: openPGP vs x509 In-Reply-To: <4073E490.7040001@smgwtest.aachen.utimaco.de> References: <87lll882wq.fsf@vigenere.g10code.de> <4073E490.7040001@smgwtest.aachen.utimaco.de> Message-ID: <20040407120730.GA3390@horse10.daimi.au.dk> On 07/04/04 13.22, Holger Sesterhenn wrote: > Hello, > > Werner Koch wrote: > > > TLS provides such a mechanism but sane users don't hand their money > > over to Verisign for a user certificate. I don't know whether > > browsers support this at all. > > It's called 'client authentication' and you have to import this special > certifacte into your browser. IE, NS and Mozilla do support such > behaviour since ages. > > I'am working with such a configuration every day. Of course we have > created our own certificate hierachy. No need for Verisign. Just adding a data-point: This is also in use for government<->citizen communication in Denmark (e.g. I used it to file my tax 'papers'), and it works pretty good. /dossen -- Common sense is the collection of prejudices acquired by age eighteen. -- Albert Einstein -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : /pipermail/attachments/20040407/4524a69c/attachment.bin From avbidder at fortytwo.ch Wed Apr 7 14:14:24 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 7 14:11:43 2004 Subject: openPGP vs x509 In-Reply-To: <4073BBEB.6070203@gmx.de> References: <4073BBEB.6070203@gmx.de> Message-ID: <200404071414.26114@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wednesday 07 April 2004 10.29, Kai Klesatschke wrote: > My opninion is, that the hierarchical X509 structur is more efficient > because not every client has to know about a special certificate. On No, I think you can just do exactly the same as with X.509: ship a few PGP keys of some CAs which are trusted by default for casual users. Or don't, and just go in 'encrypted/untrusted' mode by default. Serious users can then install PGP keys of servers they trust. I fail to see the difference here between operation with X509 certificates, or operation with PGP keys. > the other side, I have no idea how this should work with pgp-keys in > reality. To encrypt a connection normaly the opponents publickey is > used to encrypt and the secretkey is used to decrypt. This means, > that a server have to know all public keys of clients connecting to > it. A handshake between the server an client maybe solve this prob. I'm guessing here, but if I were implementing TLS/GPG, I would use the OpenPGP keys for authentication only, and create a session key on connect. Again, same as with X509/TLS, you can operate in an assymmetric mode where only the server authenticates itself with a public key, and the client does not present a public key to the server. cheers - -- vbi - -- Q: How many IBM 370's does it take to execute a job? A: Four, three to hold it down, and one to rip its head off. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkBz8KBgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6giYAoIDZdBLZoNn0El7D7QpHhIrY A+L2AKCWzOu2BfckjSfGqcrqH3wqUSybnw== =w+FY -----END PGP SIGNATURE----- From newton at hammet.net Wed Apr 7 15:31:23 2004 From: newton at hammet.net (Newton Hammet) Date: Wed Apr 7 15:28:32 2004 Subject: Implementation questions/what to do when someone leaves In-Reply-To: References: <20040407051248.GA16639@monkeypr0n.org><20040407074432.GA14995@monkeypr0n.org> Message-ID: <5584.192.35.232.241.1081344683.squirrel@webmail.io.com> On the question of a file encrypted to many keys, and one of the recipients leaves, gets convicted, etc. then simply de-crypt the document (which anyone of the other recipients can do) and then re-encrypt it with N-1 keys (the list of keys remaining with key deleted for the person for whom access is to be withdrawn). And, the re-encryption can be accomplished also by anyone of the recipients. Regards, Newton > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > >> Thanks for the reply, the email being encrypted was not a huge concern, >> we would only use that for corporate communication and if the recipient >> left then the sender would know what was in the email. > ================ > > experiment: > 1) install my public key > 2) $ chmod 0 ~/.gnupg/gpg.conf > 3) $ ps | gpg -ear 0x3D7D41E3 > > you're the sender, i'm the recipient. as the sender of that message, see > if you can decrypt it... i'll be *very* impressed if you can. > > my point, here, is that you can't always count on the sender encrypting to > themself. and what if the sender and recipient both get hit by a bus? > > >> We were more looking along the lines of storing customer info >> (passwords) where multiple sysadmins could access/decrypt the info and > ================ > > data can be ENcrypted on auto-pilot to as many keys as you want, and sent > to a file, database, etc.... > > >> the scenario our suit likes to use "What if your all hit by a bus at the >> same time?" he would be able to decrypt the passwords and pass them on >> to the replacment team. > ================ > > one of the encryption keys could belong to the boss... they never plan on > getting hit by a bus. also, you might want to print out your private key, > write out your secret password on it, put it in an envelope, and store > that in a locked safe, under control of the boss. when all of the techs > get hit by a meteorite, the new techs can recover the keys from the safe. > > >> Once a file is encrypted to multiple keys is there a way to remove one >> of the keys from being able to decrypt it? Sort of like revoking a key? > ================ > > that's tough.... i can't think of a way to do that. even if you could > "revoke" the key, let's say one of the employees emails the database to > himself, and then gets fired: he's home with a copy of the db and his > secret key. > > the best thing i can think of (at 0400 hrs) to deal with that, is to have > a file/db encrypted to each employee. if an employee leaves, that file/db > can be deleted.... but it won't do any good to delete it if they already > saved their own copy ;) > > > ...atom > > _________________________________________ > PGP key - http://atom.smasher.org/pgp.txt > 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 > ------------------------------------------------- > > "I hope we shall crush in its birth the aristocracy of our > monied corporations which dare already to challenge our > government to a trial of strength, and bid defiance to > the laws our country." > -- Thomas Jefferson, 1816 > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.4 (FreeBSD) > Comment: What is this gibberish? - > http://atom.smasher.org/links/#digital_signatures > > iD8DBQFAc7gwnCgLvz19QeMRAmJLAKClBBjNsmNTjTc7uBcJ//se6a0K6QCeJie4 > tRhmRLEz27VK7xJiruqXVQ4= > =r8cs > -----END PGP SIGNATURE----- > > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users > =========== public key: (find at http://www.pgp.net) pub 4096R/136FC036 2004-02-09 Newton Hammet finger print: 0x93cae808136fc036 From sbutler at fchn.com Wed Apr 7 17:17:39 2004 From: sbutler at fchn.com (Steve Butler) Date: Wed Apr 7 17:15:42 2004 Subject: Implementation questions Message-ID: <9A86613AB85FF346BB1321840DB42B4B046D433B@jupiter.fchn.com> Having worked as a system admin for a large newspaper company, if a critical employee is walked out the door (or leaves) all passwords known to them (or even presumed to be known by them) are immediately changed. Therefore, any database of passwords they may have copied at home (even if encrypted) are immediately invalid. The new list is distributed as needed. If the super user password changes, a copy is placed in the VP's office safe. -----Original Message----- From: Atom 'Smasher' [mailto:atom-gpg@suspicious.org] Sent: Wednesday, April 07, 2004 1:14 AM To: gnupg-users@gnupg.org Subject: Re: Implementation questions -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Thanks for the reply, the email being encrypted was not a huge concern, > we would only use that for corporate communication and if the recipient > left then the sender would know what was in the email. ================ experiment: 1) install my public key 2) $ chmod 0 ~/.gnupg/gpg.conf 3) $ ps | gpg -ear 0x3D7D41E3 you're the sender, i'm the recipient. as the sender of that message, see if you can decrypt it... i'll be *very* impressed if you can. my point, here, is that you can't always count on the sender encrypting to themself. and what if the sender and recipient both get hit by a bus? > We were more looking along the lines of storing customer info > (passwords) where multiple sysadmins could access/decrypt the info and ================ data can be ENcrypted on auto-pilot to as many keys as you want, and sent to a file, database, etc.... > the scenario our suit likes to use "What if your all hit by a bus at the > same time?" he would be able to decrypt the passwords and pass them on > to the replacment team. ================ one of the encryption keys could belong to the boss... they never plan on getting hit by a bus. also, you might want to print out your private key, write out your secret password on it, put it in an envelope, and store that in a locked safe, under control of the boss. when all of the techs get hit by a meteorite, the new techs can recover the keys from the safe. > Once a file is encrypted to multiple keys is there a way to remove one > of the keys from being able to decrypt it? Sort of like revoking a key? ================ that's tough.... i can't think of a way to do that. even if you could "revoke" the key, let's say one of the employees emails the database to himself, and then gets fired: he's home with a copy of the db and his secret key. the best thing i can think of (at 0400 hrs) to deal with that, is to have a file/db encrypted to each employee. if an employee leaves, that file/db can be deleted.... but it won't do any good to delete it if they already saved their own copy ;) ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "I hope we shall crush in its birth the aristocracy of our monied corporations which dare already to challenge our government to a trial of strength, and bid defiance to the laws our country." -- Thomas Jefferson, 1816 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAc7gwnCgLvz19QeMRAmJLAKClBBjNsmNTjTc7uBcJ//se6a0K6QCeJie4 tRhmRLEz27VK7xJiruqXVQ4= =r8cs -----END PGP SIGNATURE----- _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. From rmalayter at bai.org Wed Apr 7 17:51:54 2004 From: rmalayter at bai.org (Ryan Malayter) Date: Wed Apr 7 17:49:35 2004 Subject: Implementation questions Message-ID: <792DE28E91F6EA42B4663AE761C41C2A021C8BA0@cliff.bai.org> [Steve Butler] > Having worked as a system admin for a large newspaper > company, if a critical > employee is walked out the door (or leaves) all passwords > known to them (or > even presumed to be known by them) are immediately changed. > Therefore, any > database of passwords they may have copied at home (even if > encrypted) are > immediately invalid. > > The new list is distributed as needed. If the super user > password changes, > a copy is placed in the VP's office safe. We use a "secret sharing" scheme where N VP level personnel are given a share of a secret, and N-M of them are required to get together to reconstruct our root passwords. That way, if all IT personnel are "hit by the same bus", management can still Do a Google search for "secret splitting" and "secret sharing", and you'll see plenty of (provably secure) ways to do this. We use a simple geometric method and excel spreadsheets, distributed on business-card style CDs to the executives. But there are plenty of other ways to do it and software tools to get you there. From mwood at IUPUI.Edu Wed Apr 7 19:10:06 2004 From: mwood at IUPUI.Edu (Mark H. Wood) Date: Wed Apr 7 19:07:28 2004 Subject: openPGP vs x509 In-Reply-To: References: Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 You're asking the wrong list. The necessary work is all in the browsers. You need to add a different type of certificate evaluation to each browser, together with a new user interface to control it, one which will be somewhat more complex than the existing "I trust this certificate implicitly to certify any web page from whatever domain it says it's for" UI. There's no way to express user-specified trust levels in existing browsers. BTW, certificates can be had for free, but trust is *always* one of the costliest things you can desire. - -- Mark H. Wood, Lead System Programmer mwood@IUPUI.Edu Open-source executable: $0.00. Source: $0.00 Control: priceless! -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) Comment: pgpenvelope 2.10.2 - http://pgpenvelope.sourceforge.net/ iD8DBQFAdDX3s/NR4JuTKG8RAhsWAJ9THW/G+jqKAk9LlOJj1mpmuofvBACdHhUa IbDHYBoeztnxxW0XPk9/LJA= =o8lS -----END PGP SIGNATURE----- From wk at gnupg.org Wed Apr 7 20:26:34 2004 From: wk at gnupg.org (Werner Koch) Date: Wed Apr 7 20:11:54 2004 Subject: TLS client authentication, Re: openPGP vs x509 In-Reply-To: <4073E490.7040001@smgwtest.aachen.utimaco.de> (Holger Sesterhenn's message of "Wed, 07 Apr 2004 13:22:56 +0200") References: <87lll882wq.fsf@vigenere.g10code.de> <4073E490.7040001@smgwtest.aachen.utimaco.de> Message-ID: <87ad1n7ig5.fsf@vigenere.g10code.de> On Wed, 07 Apr 2004 13:22:56 +0200, Holger Sesterhenn said: > I'am working with such a configuration every day. Of course we have > created our own certificate hierachy. No need for Verisign. I was more thinking about "why is it not in common uses" or why does ebay/gmx/... don't use it. Agreed, OpenPGP won't help either in this case. Werner From wk at gnupg.org Wed Apr 7 20:34:14 2004 From: wk at gnupg.org (Werner Koch) Date: Wed Apr 7 20:16:53 2004 Subject: openPGP vs x509 In-Reply-To: <200404071414.26114@fortytwo.ch> (Adrian von Bidder's message of "Wed, 7 Apr 2004 14:14:24 +0200") References: <4073BBEB.6070203@gmx.de> <200404071414.26114@fortytwo.ch> Message-ID: <8765cb7i3d.fsf@vigenere.g10code.de> On Wed, 7 Apr 2004 14:14:24 +0200, Adrian 'Dagurashibanipal' von Bidder said: > Serious users can then install PGP keys of servers they trust. I fail to > see the difference here between operation with X509 certificates, or > operation with PGP keys. For a casual user it is far easier to create a PGP key than an X.509 key/certificate. You can instantly do this with OpenPGP tools (ssh is similar) whereas you have to go into lengthy discussions with OpenSSL to create even just a self-signed certificate containing a usable email attribute. Werner From atom-gpg at suspicious.org Wed Apr 7 21:02:16 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 7 20:59:30 2004 Subject: Implementation questions In-Reply-To: <9A86613AB85FF346BB1321840DB42B4B046D433B@jupiter.fchn.com> References: <9A86613AB85FF346BB1321840DB42B4B046D433B@jupiter.fchn.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Having worked as a system admin for a large newspaper company, if a critical > employee is walked out the door (or leaves) all passwords known to them (or > even presumed to be known by them) are immediately changed. Therefore, any > database of passwords they may have copied at home (even if encrypted) are > immediately invalid. =============================== a db full of old passwords won't do them any good, but a db full of customer information can't be updated every time someone leaves the company. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "This is Radio Clash On pirate satellite Orbiting your living room Everybody hold on tight" -- The Clash -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdFA8nCgLvz19QeMRAmmwAJ9v97MxS8GPIvtc8EMg6s8wZ+K13QCghIKs PvEl9pydt2FnoUcGKW9VlT0= =MOb4 -----END PGP SIGNATURE----- From gnupg at kubieziel.de Wed Apr 7 21:02:30 2004 From: gnupg at kubieziel.de (Jens Kubieziel) Date: Wed Apr 7 20:59:45 2004 Subject: --refresh-keys stops working while processing Message-ID: <20040407190230.GC1872@kubieziel.de> Hi, I wanted to update my keyring and did a "gpg --refresh-keys" without other options. After processing 194 keys it stopped. I got no error message and the last line which was spit out is: gpg: Schl?ssel C5836F2A: "Christian Franke (Der Spinner) " Nicht ge?ndert After waiting some 10-20 minutes I killed the process. When I retried it, the gpg-process stopped at approximately 110 keys. I can reproduce this error for every "--refresh-keys". What can be the reason for this? (I was only the whole time.) Thanks for any hints -- Jens Kubieziel http://www.kubieziel.de Any sufficiently advanced technology is indistinguishable from a rigged demo. From atom-gpg at suspicious.org Wed Apr 7 21:03:51 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 7 21:00:59 2004 Subject: Implementation questions Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > I can't seem to find any docs on encrypting to multiple keys, I encrypt > my important files all the time, do you have a URL or some info > on how to encrypt to multiple keys? or am I overanalyzing this function? > I have been up all night trying to rebuild a crashed server, so > overlooking simple info is not out of the question right now. ========================= this encrypts to 0x12345678 and 0x09876543: cat secret-file | gpg -ea -r 0x12345678 -r 0x09876543 ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Love all. Trust few. Do wrong to none." -- William Shakespeare -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdE+DnCgLvz19QeMRAj5lAJ4zaFTNEnQr1KmhsmSiR0+LzRZypACgqcs5 MCtN++oYvpTAdz3zYeS/hC0= =ScI5 -----END PGP SIGNATURE----- From jkasperer at bluewin.ch Wed Apr 7 21:19:01 2004 From: jkasperer at bluewin.ch (Kasperer) Date: Wed Apr 7 21:17:05 2004 Subject: PGP decompression error, GPG is OK! Message-ID: <002001c41cd5$30bd2350$86744151@M> This is not strictly a GnuPG question, however I believe you might shed light on it. The message that follows can be decrypted by GnuPG (v1.2.2), but PGP v8.0.3 says "error decompressing data": http://www.megacryption.cc/download/PGPerror.asc The file has been symmetrically encrypted with passphrase "a", CAST algorithm and ZIP compression (RFC1951), normally supported by both GnuPG and PGP. It has something to do with the last part of the compressed packet, the rest of the data is correctly decrypted by PGP. Any guess appreciated. Thanks in advance. John Kasperer From linux at codehelp.co.uk Wed Apr 7 21:31:32 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Wed Apr 7 21:28:47 2004 Subject: Implementation questions In-Reply-To: <20040407085439.GA30640@monkeypr0n.org> References: <20040407051248.GA16639@monkeypr0n.org> <20040407085439.GA30640@monkeypr0n.org> Message-ID: <200404072031.32673.linux@codehelp.co.uk> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wednesday 07 Apr 2004 9:54, Jason Burnett wrote: > on Wed Apr 07 Atom 'Smasher' spoke forth with the blessed manuscript > I can't seem to find any docs on encrypting to multiple keys, I encrypt > my important files all the time, do you have a URL or some info > on how to encrypt to multiple keys? or am I overanalyzing this function? Just specify more than one keyid on the encrypt command line: $ gpg -ear keyid1 keyid2 keyid3 keyid4 keyidN As atom hinted, you can specify recipient keys in gpg.conf and if you are in control of user gpg.conf files, as you claimed, you could specify the list there. Administrating this kind of system will require exponentially more input the more users you enlist. > Our main goal is to make sure that, as long our employees maintain > proper personal security, that the data is fairly safe from prying eyes. Wouldn't an encrypted filesystem be an alternative? - -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAdFcUiAEJSii8s+MRApyOAKDBCKKMbBlDBgOUCFyp1RrXkyqu7ACgiJF1 4nxPALVsyMxNNB2ptDzKVa8= =wH8n -----END PGP SIGNATURE----- From jason at monkeypr0n.org Wed Apr 7 22:27:41 2004 From: jason at monkeypr0n.org (Jason Burnett) Date: Wed Apr 7 22:24:57 2004 Subject: Implementation questions In-Reply-To: <200404072031.32673.linux@codehelp.co.uk> References: <20040407051248.GA16639@monkeypr0n.org> <20040407085439.GA30640@monkeypr0n.org> <200404072031.32673.linux@codehelp.co.uk> Message-ID: <20040407202741.GA12357@monkeypr0n.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 on Wed Apr 07 Neil Williams spoke forth with the blessed manuscript > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > On Wednesday 07 Apr 2004 9:54, Jason Burnett wrote: > > on Wed Apr 07 Atom 'Smasher' spoke forth with the blessed manuscript > > I can't seem to find any docs on encrypting to multiple keys, I encrypt > > my important files all the time, do you have a URL or some info > > on how to encrypt to multiple keys? or am I overanalyzing this function? > Just specify more than one keyid on the encrypt command line: > $ gpg -ear keyid1 keyid2 keyid3 keyid4 keyidN > As atom hinted, you can specify recipient keys in gpg.conf and if you are in > control of user gpg.conf files, as you claimed, you could specify the list > there. Thanks for all the suggestions, I have been experimenting with the suggestions today and I am very satisfied with the results so far. > Administrating this kind of system will require exponentially more input the > more users you enlist. > > Our main goal is to make sure that, as long our employees maintain > > proper personal security, that the data is fairly safe from prying eyes. > Wouldn't an encrypted filesystem be an alternative? This is another option that is being discussed. > - -- > Neil Williams > ============= > http://www.codehelp.co.uk/ > http://www.dclug.org.uk/ > http://www.isbn.org.uk/ > http://sourceforge.net/projects/isbnsearch/ > http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.4 (GNU/Linux) > iD8DBQFAdFcUiAEJSii8s+MRApyOAKDBCKKMbBlDBgOUCFyp1RrXkyqu7ACgiJF1 > 4nxPALVsyMxNNB2ptDzKVa8= > =wH8n > -----END PGP SIGNATURE----- > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users - -- PGP key - http://www.monkeypr0n.org/keyring/jason@monkeypr0n 793A 39E7 5655 0415 5C5E 976E 94D8 F8C4 C061 DF64 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAdGQ9lNj4xMBh32QRAs1qAJ45E3UfgYvaYteUnnHxH2F3/njYEQCfYfsA lJeLBe4mTCCETqec//OJSmY= =RkQZ -----END PGP SIGNATURE----- From jkasperer at bluewin.ch Wed Apr 7 23:13:03 2004 From: jkasperer at bluewin.ch (Kasperer) Date: Wed Apr 7 23:10:49 2004 Subject: PGP decompression error, GPG is OK! Message-ID: <007601c41ce5$1d52f140$86744151@M> Never mind, I realized that garbage data was added to the compressed data, before encryption. Glad to see that GPG behaves better than PGP in this occasion. From shavital at mac.com Thu Apr 8 05:26:18 2004 From: shavital at mac.com (Charly Avital) Date: Thu Apr 8 05:23:45 2004 Subject: --refresh-keys stops working while processing In-Reply-To: <20040407190230.GC1872@kubieziel.de> References: <20040407190230.GC1872@kubieziel.de> Message-ID: Might it be related to the status of the keyserver from which gpg is refreshing the keys? I just ran the same line, the process completed without any problems, at the end I got the following sum up: ---------- gpg: 3 marginal(s) needed, 1 complete(s) needed, classic trust model gpg: depth: 0 valid: 22 signed: 44 trust: 0-, 0q, 0n, 0m, 0f, 22u gpg: depth: 1 valid: 44 signed: 20 trust: 2-, 1q, 0n, 8m, 33f, 0u gpg: depth: 2 valid: 15 signed: 9 trust: 1-, 0q, 0n, 4m, 10f, 0u gpg: next trustdb check due at 2004-05-22 gpg: Total number processed: 244 gpg: w/o user IDs: 1 gpg: unchanged: 95 gpg: new user IDs: 78 gpg: new subkeys: 40 gpg: new signatures: 4384 -------------------------------------------- There a few incidences a key not found in the particular keyserver I was using (keyserver.kjsl.com), e.g.: gpgkeys: key [...finger print, or long ID...] not found on keyserver. Running GnuPG 1.3.5, under Macintosh OS 10.3.3 Charly At 9:02 PM +0200 4/7/04, Jens Kubieziel wrote: >Hi, > >I wanted to update my keyring and did a "gpg --refresh-keys" without >other options. After processing 194 keys it stopped. I got no error >message and the last line which was spit out is: >gpg: Schl?ssel C5836F2A: "Christian Franke (Der Spinner) " Nicht ge?ndert > >After waiting some 10-20 minutes I killed the process. When I retried it, >the gpg-process stopped at approximately 110 keys. I can reproduce this >error for every "--refresh-keys". >What can be the reason for this? (I was only the whole time.) > >Thanks for any hints >-- >Jens Kubieziel http://www.kubieziel.de >Any sufficiently advanced technology is indistinguishable from a rigged demo. From johanw at vulcan.xs4all.nl Wed Apr 7 23:28:27 2004 From: johanw at vulcan.xs4all.nl (Johan Wevers) Date: Thu Apr 8 17:41:42 2004 Subject: --refresh-keys stops working while processing In-Reply-To: <20040407190230.GC1872@kubieziel.de> from Jens Kubieziel at "Apr 7, 2004 09:02:30 pm" Message-ID: <200404072128.XAA00643@vulcan.xs4all.nl> Jens Kubieziel wrote: >After waiting some 10-20 minutes I killed the process. When I retried it, >the gpg-process stopped at approximately 110 keys. I can reproduce this >error for every "--refresh-keys". >What can be the reason for this? (I was only the whole time.) I have the same. I assume it's a long delay in the keyserver response. When I wait very long the process will finish. -- ir. J.C.A. Wevers // Physics and science fiction site: johanw@vulcan.xs4all.nl // http://www.xs4all.nl/~johanw/index.html PGP/GPG public keys at http://www.xs4all.nl/~johanw/pgpkeys.html From russ at coldstonelabs.org Fri Apr 9 04:04:04 2004 From: russ at coldstonelabs.org (Russell Valentine) Date: Fri Apr 9 04:01:37 2004 Subject: Use of public key servers Message-ID: <40760494.5080109@coldstonelabs.org> Hi, I've posted my public key on a public key server a while ago. A friend of mind was trying to explain to me that public key servers are useless and bad. His arguments where: 1) You can't prove a key from the public key server is really that persons public key, you still have to validate it some way. So you can get the key personally from that person in some way and at that time validate the key. Instead of using the key even thought it might not be his. 2) It allows evil people get to get data from the public key servers and do malicious things with it (Such as spammers with email addresses, or possibly know a username for a account on some server that hosts the email.) I don't really know what to say to him, and am wondering if anyone could tell me or point me to some good reasons as to why you would want to post your public key on a public key server? Thanks. Russell Valentine -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 256 bytes Desc: OpenPGP digital signature Url : /pipermail/attachments/20040408/70df2f70/signature.bin From dshaw at jabberwocky.com Fri Apr 9 04:57:57 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 9 04:55:17 2004 Subject: Use of public key servers In-Reply-To: <40760494.5080109@coldstonelabs.org> References: <40760494.5080109@coldstonelabs.org> Message-ID: <20040409025757.GB25930@jabberwocky.com> On Thu, Apr 08, 2004 at 09:04:04PM -0500, Russell Valentine wrote: > Hi, > > I've posted my public key on a public key server a while ago. A friend > of mind was trying to explain to me that public key servers are useless > and bad. His arguments where: > > 1) You can't prove a key from the public key server is really that > persons public key, you still have to validate it some way. So you can > get the key personally from that person in some way and at that time > validate the key. Instead of using the key even thought it might not be his. All true. However, note the existence of phone books and directory assistance. It is convenient to be able to look people up in one place, rather than asking everyone for their key individually. > 2) It allows evil people get to get data from the public key servers and > do malicious things with it (Such as spammers with email addresses, or > possibly know a username for a account on some server that hosts the email.) Hasn't happened yet. Isn't likely to happen for various technical reasons, the simplest being that for a spammer, scraping addresses off of the web or usenet is so trivial, why bother with something difficult? David -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 330 bytes Desc: not available Url : /pipermail/attachments/20040408/9ac67ba8/attachment.bin From dshaw at jabberwocky.com Fri Apr 9 04:59:02 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 9 04:56:20 2004 Subject: --refresh-keys stops working while processing In-Reply-To: <20040407190230.GC1872@kubieziel.de> References: <20040407190230.GC1872@kubieziel.de> Message-ID: <20040409025902.GC25930@jabberwocky.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, Apr 07, 2004 at 09:02:30PM +0200, Jens Kubieziel wrote: > Hi, > > I wanted to update my keyring and did a "gpg --refresh-keys" without > other options. After processing 194 keys it stopped. I got no error > message and the last line which was spit out is: > gpg: Schl?ssel C5836F2A: "Christian Franke (Der Spinner) " Nicht ge?ndert > > After waiting some 10-20 minutes I killed the process. When I retried it, > the gpg-process stopped at approximately 110 keys. I can reproduce this > error for every "--refresh-keys". > What can be the reason for this? (I was only the whole time.) Do you have any Elgamal signing keys in your keyring? Refreshing one of those can take a very long time. David -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.3.6-cvs (GNU/Linux) Comment: Key available at http://www.jabberwocky.com/david/keys.asc iHEEARECADEFAkB2EXYqGGh0dHA6Ly93d3cuamFiYmVyd29ja3kuY29tL2Rhdmlk L2tleXMuYXNjAAoJEOJmXIdJ4cvJgRkAnRv2gXmjIXJQ7pEN+nzrLWbG1GTDAJ9m uqtiracGceWGBtF21rrDfVyHcg== =NEQL -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Fri Apr 9 05:03:42 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 9 05:00:59 2004 Subject: Use of public key servers In-Reply-To: <40760494.5080109@coldstonelabs.org> References: <40760494.5080109@coldstonelabs.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > 1) You can't prove a key from the public key server is really that > persons public key, you still have to validate it some way. So you can > get the key personally from that person in some way and at that time > validate the key. Instead of using the key even thought it might not be his. ====================================== if me and you know and trust each other, we can sign each others keys (after properly validating them). from then on, if i download a key from a key-server that's signed by you, i can reasonably assume that the key belongs to the person who claims to own that key: if you download a key that's signed by me, you can reasonably assume the same. getting around that requires forging a signature, which is generally considered to be infeasible. type "web of trust" into your favorite search engine. > 2) It allows evil people get to get data from the public key servers and > do malicious things with it (Such as spammers with email addresses, or > possibly know a username for a account on some server that hosts the email.) ====================================== the only information they would get is an email address, user name, comment, and *public* key information.... * email address: there has been no evidence of spammers harvesting email from key servers. it might seem like a hot target, but it's really not. in any case, install a spam filter. * user name: this is (typically) the user name associated with that address, so there are easier ways to find out that i'm "Atom Smasher". * comment: this optional field can include anything you want, but some common sense will tell you what not to put in it (like a password, or social security number {in the US}). * public key: a big part of public key cryptography depends on making the public key widely available. if you can gain information about one's private key by having access to their public key, we'd all like to know about it. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "[The] feminist agenda is not about equal rights for women. It is about a socialist, anti-family political movement that encourages women to leave their husbands, kill their children, practice witchcraft, destroy capitalism and become lesbians." -- Rev. Pat Robertson, 1992 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdhKSnCgLvz19QeMRAgmoAJ4mELCCqcCNTOZsyMSGcdk/kX7TzwCeK/af 4dX6Ii1zrkR7W/WPyGiWJTY= =U9GD -----END PGP SIGNATURE----- From agreene at pobox.com Fri Apr 9 05:15:01 2004 From: agreene at pobox.com (Anthony E. Greene) Date: Fri Apr 9 05:11:20 2004 Subject: Use of public key servers In-Reply-To: <40760494.5080109@coldstonelabs.org> References: <40760494.5080109@coldstonelabs.org> Message-ID: <20040409031500.GB16043@cp5340.localdomain> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 08-Apr-2004/21:04 -0500, Russell Valentine wrote: >I've posted my public key on a public key server a while ago. A friend >of mind was trying to explain to me that public key servers are useless >and bad. His arguments where: > >1) You can't prove a key from the public key server is really that >persons public key, you still have to validate it some way. So you can >get the key personally from that person in some way and at that time >validate the key. Instead of using the key even thought it might not be his. That only matters if you really need to know that a certain key really belongs to a certain person. If I only need to communicate securely with russ@coldstonelabs.org, I don't necessarily care whether the key I use actually belongs to a real world person named Russell Valentine. For example, if you wish to continue this discussion off-list, do you really care whether 0x6C94239D belongs to Anthony E. Greene, or do you just need to know that you can use it to send a secure message to agreene@pobox.com? Keyservers make casual use of cryptography between strangers easier. They are not intended to solve the Web-of-Trust problem. >2) It allows evil people get to get data from the public key servers and >do malicious things with it (Such as spammers with email addresses, or >possibly know a username for a account on some server that hosts the email.) There are much easier ways to get email addresses (which also implies much easier ways to get usernames). In the early days of PGP, cryptography was so geeky and spam so against the prevailing attitude, that spammers would have been inviting retaliation by targeting PGP users. These days there are too many easy/cheap/fast ways to get hundreds of thousands of email addresses to worry about harvesting keyservers. Tony - -- Anthony E. Greene AOL/Yahoo Messenger: TonyG05 HomePage: OpenPGP Key: 0x6C94239D/7B3D BD7D 7D91 1B44 BA26 C484 A42A 60DD 6C94 239D Linux. The choice of a GNU generation -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Anthony E. Greene 0x6C94239D iD8DBQFAdhUvpCpg3WyUI50RArD6AJ4llE5VwNFC4MWlkENOI6lo/WeENACcClhm gsAE1fWY+mkSXZzP1G5992Y= =WhHB -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Fri Apr 9 06:32:12 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 9 06:29:30 2004 Subject: 5.2.3.15. Notation Data In-Reply-To: References: <40760494.5080109@coldstonelabs.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 (using gpg 1.2.4) can someone show me some real-world examples of using: --sig-notation --cert-notation --set-notation trying this: $ gpg --cert-notation "%k=xyz" -u smasher --sign-key 4697BE6F gives me this error: gpg: a user notation name must contain the '@' character that seems *sort-of* wrong, after reading the man page. but i did get this to work: gpg --cert-notation "xxx@xxx.xxx=xyz" -u smasher --sign-key 4697BE6F is there a way to include notation data in self-signatures? would that be a good way to specify something like: "This key is not to be used for legally binding signatures." if that can't be done (in a self-signature) with a notation packet, what's a good way to make such intended use of a key clear? what are good reasons for using notation data? what are bad reasons? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Every great advance in natural knowledge has involved the absolute rejection of authority." -- Julian Huxley -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdidRnCgLvz19QeMRAtbhAJ9zHmKrBtzPUpkuF8fz3Q5VncFh9QCfQvMN 2oeE+4HpeU9YJjv/75rYym4= =FCPF -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Fri Apr 9 06:55:55 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 9 06:53:14 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <40760494.5080109@coldstonelabs.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 using a cleartext signature while attempting to include a policy URL or notation data gives me these errors: gpg: can't put a policy URL into v3 (PGP 2.x style) signature and gpg: can't put notation data into v3 (PGP 2.x style) signatures i'm surprised to see this limitation... is that correct? i don't see where rfc 2440 prohibits this... ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "I would rather fail in a cause that will ultimately triumph than triumph in a cause that will ultimately fail." -- Woodrow Wilson -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdizgnCgLvz19QeMRAiizAJ9vJ/4XeoAzccMJiIbW6VvLSIRcggCfTr1I pLNi8vEyE8u2GyDlcE+OiRI= =s1X/ -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Fri Apr 9 07:23:41 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 9 07:20:56 2004 Subject: notation data & policy URL In-Reply-To: References: <40760494.5080109@coldstonelabs.org> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 notation data (5.2.3.15) and policy URL (5.2.3.19) seem to be the place for human readable notes that don't fit neatly anywhere else.... is there anything wrong with using the policy URL for non-URL text? example: i sign my brothers key while meeting for lunch... i could ~just~ sign the key and leave it at that, or... i could specify the conditions of the signature: This is my brother. I've known him since 1976 and verified his key in person. is there a recommended way to include that information directly in my signature on his key? obviously, i could put that information on a web page, and use a policy URL to direct people to that page (that page should be clearsigned txt?). would it be a bad idea to just include that text in the policy URL space? is there a good way to use the notation data for that type of information? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "As soon as men decide that all means are permitted to fight an evil, then their good becomes indistinguishable from the evil that they set out to destroy." -- Christopher Dawson, The Judgment of Nations, 1942 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdjNhnCgLvz19QeMRArWjAKCDTPl/PztxL3j3Ys+60VJvyGbWlwCfSLfT hw4ms4FsQ+C0Oj2lsY+hHv4= =/Uei -----END PGP SIGNATURE----- From jason.harper at ge.com Thu Apr 1 21:26:22 2004 From: jason.harper at ge.com (Harper, Jason (GE Consumer Finance)) Date: Fri Apr 9 11:19:12 2004 Subject: GPG - Invalid packet Message-ID: <7730B3FE2184A5499A4A8601F598BB1B279018@ALPMLVEM01.e2k.ad.ge.com> Dear Group, I've been using gpg to do file encryption with partners of ours, but we just ran into two odd issues. I've done the google thing on this, but I'm not making much progress. Does the list have any pointers for me to look at? We've not run into this before. The source file is being encrypted, and compressed and signed by an NAI 7.11 Business PGP system, and we are opening on a Solaris 8 box with gpg 1.2.3. I've sanitized some of the data to protect the innocent.. Sample one: gpg --output ce331 --decrypt CE_Con_1040326.dat gpg: WARNING: using insecure memory! gpg: please see http://www.gnupg.org/faq.html for more information gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created 2001-08-23 "KEYNAME " gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created 2004-03-25 "key@sample.com (COMPANY decription key) " gpg: Signature made Fri Mar 26 14:20:29 2004 EST using DSA key ID xxxxxxxx gpg: Good signature from "COMPANY " gpg: aka "COMPANY " gpg: [don't know]: invalid packet (ctb=1f) gpg: WARNING: message was not integrity protected gpg: [don't know]: invalid packet (ctb=40) Then it just ends, but we end up with a plain text file. Accuracy of the file is unkown. What is the CTB=1f and CTB=40 Sample two: ->> gpg --output au331 --decrypt AU_Con_1040331.dat gpg: WARNING: using insecure memory! gpg: please see http://www.gnupg.org/faq.html for more information gpg: encrypted with 3072-bit ELG-E key, ID xxxxxxxx, created 2001-08-23 "COMPANY " gpg: encrypted with 1024-bit ELG-E key, ID xxxxxxxx, created 2004-03-25 "key@sample.com (COMPANY decription key) " gpg: Signature made Fri Mar 26 14:20:46 2004 EST using DSA key ID xxxxxxx gpg: Good signature from "COMPANY " gpg: aka "COMPANY " File `au331' exists. Overwrite (y/N)? yes gpg: Problem reading source (2386467048 bytes remaining) gpg: handle plaintext failed: file read error gpg: WARNING: message was not integrity protected Thanks for any and all assistance. jason From dthomson_hms at yahoo.com Tue Apr 6 22:05:37 2004 From: dthomson_hms at yahoo.com (Doug Thomson) Date: Fri Apr 9 11:19:17 2004 Subject: I have keys - now I want to encrypt a file Message-ID: <20040406200537.97433.qmail@web12404.mail.yahoo.com> Hi all, I have just installed gpg on my Solaris system. My requirement is to encrypt small flat files and send them to outside recipients. I have successfully imported two public keys from these recipients. Now I want to encrypt 'testfile' to prepare it for FTPing to one of the recipients. What syntax should I use? I have thus far tried gpg --encrypt-files testfile --recipient somebody@location.com keyname but I get the message 'you did not specify a user ID' suggestions? I have checked the FAQ and do not find the answer there. thanks ! Doug Thomson Minneapolis, MN USA dthomson_hms@yahoo.com __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/ From anonymous at remailer.metacolo.com Tue Apr 6 22:41:02 2004 From: anonymous at remailer.metacolo.com (Anonymous Sender) Date: Fri Apr 9 11:19:20 2004 Subject: GPG - Invalid packet Message-ID: <5a889d950e62b97c710e5a9ab72c634e@remailer.metacolo.com> > The source file is being encrypted, and compressed and signed by an NAI > 7.11 Business PGP system, and we are opening on a Solaris 8 box with gpg > 1.2.3. How was the encrypted file sent to you? By e-mail, FTP, HTTP? From linux at codehelp.co.uk Fri Apr 9 12:29:55 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Fri Apr 9 12:26:37 2004 Subject: I have keys - now I want to encrypt a file In-Reply-To: <20040406200537.97433.qmail@web12404.mail.yahoo.com> References: <20040406200537.97433.qmail@web12404.mail.yahoo.com> Message-ID: <200404091130.03051.linux@codehelp.co.uk> On Tuesday 06 April 2004 9:05, Doug Thomson wrote: > Hi all, > I have just installed gpg on my Solaris system. > My requirement is to encrypt small flat files and send > them to outside recipients. I have successfully > imported two public keys from these recipients. Now I > want to encrypt 'testfile' to prepare it for FTPing to > one of the recipients. What syntax should I use? I > have thus far tried > > gpg --encrypt-files testfile --recipient > somebody@location.com keyname Use the keyid for the recipient - what do you mean by keyname? Matching by email address is possible but commonly ends up with more than one key available. e.g., I have two keys that would match my email address on keyservers and more that have not been used publicly. --encrypt-files is a special case: from man gpg The command expects the files to be encrypted either on the command line or reads the filenames from stdin; each name must be on separate line. The command is intended for a quick encryption of multiple files. Normally, just use --encrypt or just -e To send by FTP, it may be best to use ASCII armour and send as an ASCII transfer. The final command becomes: $ gpg --recipient 0xDEADBEEF --armor --encrypt testfile or just $ gpg -ar 0xDEADBEEF -e testfile Replace DEADBEEF with the keyid of the recipient, in my case 28BCB3E3. Look for the keyid from this output: $ gpg --list-key keyname Unless the recipient is an automated script, it's best to also sign the encrypted file: $ gpg -sar 0xDEADBEEF -r keyid (to help your recipient have some confidence about the contents of the encrypted file). > but I get the message 'you did not specify a user ID' You didn't specify a UID, you specified an email address that could match more than one UID or none at all. > I have checked the FAQ and do not find the answer > there. Check the man page, learn how to specify a keyid. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040409/56d8cdd5/attachment.bin From JPClizbe at comcast.net Fri Apr 9 12:47:04 2004 From: JPClizbe at comcast.net (John Clizbe) Date: Fri Apr 9 12:45:06 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <40760494.5080109@coldstonelabs.org> Message-ID: <40767F28.4080201@comcast.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Atom 'Smasher' wrote: > using a cleartext signature while attempting to include a policy URL or > notation data gives me these errors: > > gpg: can't put a policy URL into v3 (PGP 2.x style) signature > and > gpg: can't put notation data into v3 (PGP 2.x style) signatures > > i'm surprised to see this limitation... is that correct? > > i don't see where rfc 2440 prohibits this... > RFC 2440 prohibiting their use probably isn't the issue. RFC 2440 dates from Nov. 1998. The first PGP version after that would have been PGP 6.0. That would be the earliest that support for RFC 2440 features could or should be expected. One cannot realisticly expect full 2440 behavior with software as old as PGP 2.x. Nor should those features be expected to be included or to work in a 2.x compatibility mode. This is a case of GnuPG protecting you from yourself. - -- John P. Clizbe Inet: JPClizbe(a)comcast DOT nyet Golden Bear Networks PGP/GPG KeyID: 0x608D2A10 "Most men take the straight and narrow. A few take the road less traveled. I chose to cut through the woods." -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5rc1 (Windows 2000 SP4) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFAdn8mHQSsSmCNKhARAsQgAKDYC4Ya70iUWHPeY+lssMuJqS5iSQCdH0Ql 7V/V2JmdsElGu820p0KGh7o= =BbVx -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Fri Apr 9 14:09:46 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 9 14:07:01 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <40760494.5080109@coldstonelabs.org> Message-ID: <20040409120946.GE25930@jabberwocky.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, Apr 09, 2004 at 12:55:55AM -0400, Atom 'Smasher' wrote: > using a cleartext signature while attempting to include a policy URL or > notation data gives me these errors: > > gpg: can't put a policy URL into v3 (PGP 2.x style) signature > and > gpg: can't put notation data into v3 (PGP 2.x style) signatures > > i'm surprised to see this limitation... is that correct? > > i don't see where rfc 2440 prohibits this... It's not a question of prohibition. It's a question of impossibility. V3 signatures have no subpackets - there is no place to put the policy URL. David -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.3.6-cvs (GNU/Linux) Comment: Key available at http://www.jabberwocky.com/david/keys.asc iHEEARECADEFAkB2kooqGGh0dHA6Ly93d3cuamFiYmVyd29ja3kuY29tL2Rhdmlk L2tleXMuYXNjAAoJEOJmXIdJ4cvJSSEAniYoDtzBDdst0C3WWIuDNdN9NVWyAKC7 w14gcZY63xUKhOqY8mKUh6VJfQ== =1fZt -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Fri Apr 9 14:12:27 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 9 14:09:54 2004 Subject: notation data & policy URL In-Reply-To: References: <40760494.5080109@coldstonelabs.org> Message-ID: <20040409121227.GF25930@jabberwocky.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, Apr 09, 2004 at 01:23:41AM -0400, Atom 'Smasher' wrote: > notation data (5.2.3.15) and policy URL (5.2.3.19) seem to be the place > for human readable notes that don't fit neatly anywhere else.... > > is there anything wrong with using the policy URL for non-URL text? Yes. Policy URLs are URLs. > example: i sign my brothers key while meeting for lunch... i could ~just~ > sign the key and leave it at that, or... i could specify the conditions of > the signature: > This is my brother. I've known him since 1976 and verified his key > in person. > > is there a recommended way to include that information directly in my > signature on his key? Notations. Still, if you include all that information in the signature, you're making the signature larger. A policy URL at least is small. David -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.3.6-cvs (GNU/Linux) Comment: Key available at http://www.jabberwocky.com/david/keys.asc iHEEARECADEFAkB2kysqGGh0dHA6Ly93d3cuamFiYmVyd29ja3kuY29tL2Rhdmlk L2tleXMuYXNjAAoJEOJmXIdJ4cvJrFoAn2aSqH1kPai/U1pwiLeTKFQ/OUFSAKDT bNW+F0O6yKcUUQgD4b46Eie2IQ== =ellu -----END PGP SIGNATURE----- From sbutler at fchn.com Fri Apr 9 16:19:43 2004 From: sbutler at fchn.com (Steve Butler) Date: Fri Apr 9 16:17:34 2004 Subject: I have keys - now I want to encrypt a file Message-ID: <9A86613AB85FF346BB1321840DB42B4B046D4362@jupiter.fchn.com> Change the order a little. gpg --recipient "$1" --output "$2.gpg" --encrypt "$2" -----Original Message----- From: Doug Thomson [mailto:dthomson_hms@yahoo.com] Sent: Tuesday, April 06, 2004 1:06 PM To: gnupg-users@gnupg.org Subject: I have keys - now I want to encrypt a file Hi all, I have just installed gpg on my Solaris system. My requirement is to encrypt small flat files and send them to outside recipients. I have successfully imported two public keys from these recipients. Now I want to encrypt 'testfile' to prepare it for FTPing to one of the recipients. What syntax should I use? I have thus far tried gpg --encrypt-files testfile --recipient somebody@location.com keyname but I get the message 'you did not specify a user ID' suggestions? I have checked the FAQ and do not find the answer there. thanks ! Doug Thomson Minneapolis, MN USA dthomson_hms@yahoo.com __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/ _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. From atom-gpg at suspicious.org Fri Apr 9 18:01:58 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 9 17:59:13 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: <20040409120946.GE25930@jabberwocky.com> References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > It's not a question of prohibition. It's a question of > impossibility. V3 signatures have no subpackets - there is no place > to put the policy URL. =============================== i guess i never realized that clearsigning was a v3-only animal. is that something that might warrant an update in a pending draft rfc? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "The real truth of the matter is, as you and I know, that a financial element in the large centers has owned the government of the U.S. since the days of Andrew Jackson." -- Franklin Delano Roosevelt, November 21st, l933 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdsj6nCgLvz19QeMRAvztAJ4vVsfcHOyXa6pUpnEBQJrWdOYqfwCeOY5s iIRJZ1Td/vIyTjXkKo2vIQk= =BhHi -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Fri Apr 9 18:49:14 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 9 18:46:29 2004 Subject: notation data & policy URL In-Reply-To: <20040409121227.GF25930@jabberwocky.com> References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > is there anything wrong with using the policy URL for non-URL text? > > Yes. Policy URLs are URLs. > > > example: i sign my brothers key while meeting for lunch... i could ~just~ > > sign the key and leave it at that, or... i could specify the conditions of > > the signature: > > This is my brother. I've known him since 1976 and verified his key > > in person. > > > > is there a recommended way to include that information directly in my > > signature on his key? > > Notations. > > Still, if you include all that information in the signature, you're > making the signature larger. A policy URL at least is small. ================================= i still don't 'get' why the notation data is in the form: foo=bar why not have just 'foo', or 'bar'? also, can a policy url and/or notation data be included in a self signature? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "States of war are also understood to require the opposition in the legislature to moderate its otherwise essential functions of criticism. Calls are issued to stand behind the political leadership and to display unity, with the implication that the enemy is watching and that failure to unite is tantamount to treason. These are not healthy conditions for a democracy; indeed, they are the opposite of democracy." -- Philip E. Agre, Department of Information Studies, University of California, Los Angeles -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdtQOnCgLvz19QeMRAqVjAJ9n4fe/ykTgLsVsp/P92h8TjUBj6wCgjk04 G7kFz9QWfowaxSsfMf5Awek= =NT1/ -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Fri Apr 9 20:26:39 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 9 20:23:59 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> Message-ID: <20040409182638.GA2867@jabberwocky.com> On Fri, Apr 09, 2004 at 12:01:58PM -0400, Atom 'Smasher' wrote: > > It's not a question of prohibition. It's a question of > > impossibility. V3 signatures have no subpackets - there is no place > > to put the policy URL. > =============================== > > i guess i never realized that clearsigning was a v3-only animal. is that > something that might warrant an update in a pending draft rfc? Huh? Clearsigning is not a v3 only thing. David From atom-gpg at suspicious.org Fri Apr 9 20:35:40 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 9 20:32:55 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: <20040409182638.GA2867@jabberwocky.com> References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> <20040409182638.GA2867@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > > It's not a question of prohibition. It's a question of > > > impossibility. V3 signatures have no subpackets - there is no place > > > to put the policy URL. > > =============================== > > > > i guess i never realized that clearsigning was a v3-only animal. is that > > something that might warrant an update in a pending draft rfc? > > Huh? Clearsigning is not a v3 only thing. ========================= is there a v4 definition of clearsigning that allows subpackets? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Those who profess to favor freedom, and yet deprecate agitation, are men who want rain without thunder and lightning. They want the ocean without the roar of its many waters." -- Frederick Douglass -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdu0AnCgLvz19QeMRAo1kAJ9b9NuTvCNacDjYPcZJQiIAvhzDowCeJVJr SdcQ2TH3lNzcxYnbT/2xmqY= =N092 -----END PGP SIGNATURE----- From dgc at uchicago.edu Fri Apr 9 20:50:41 2004 From: dgc at uchicago.edu (David Champion) Date: Fri Apr 9 20:48:21 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> <20040409182638.GA2867@jabberwocky.com> Message-ID: <20040409185041.GG14052@dust.uchicago.edu> * On 2004.04.09, in , * "Atom 'Smasher'" wrote: > > > i guess i never realized that clearsigning was a v3-only animal. is that > > > something that might warrant an update in a pending draft rfc? > > > > Huh? Clearsigning is not a v3 only thing. > > is there a v4 definition of clearsigning that allows subpackets? I'm not a good one to talk about such things, but to throw in a little support here: v4 and v3 only refer to the structure of the key data. You can clearsign with any type of key, because "clearsign" only describes the datastream of the signed material. GnuPG will clearsign with v3 or v4 keys. Only v4 keys allow subpackets; v3 as a key format predates subpacket support in PGP. Allowing them in v3 keys would make the keys incompatible across different v3-supporting applications, which would pretty much negate the purpose of naming one key version "v3" and another "v4". You're presumably getting this error not because you can't clearsign, but because you can't have a policy URL or notation in a v3 key. -- -D. dgc@uchicago.edu NSIT University of Chicago From atom-gpg at suspicious.org Fri Apr 9 21:14:00 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 9 21:11:18 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: <20040409185041.GG14052@dust.uchicago.edu> References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> <20040409182638.GA2867@jabberwocky.com> <20040409185041.GG14052@dust.uchicago.edu> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > You're presumably getting this error not because you can't clearsign, > but because you can't have a policy URL or notation in a v3 key. ======================== but i should be able to include a policy URL in a data signature, and i am using a v4 key. key with subpackets (v4): %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% $ gpg --list-key smasher pub 1024D/3D7D41E3 2003-10-04 Atom Smasher sig 3 3D7D41E3 2004-02-22 Atom Smasher uid Atom Smasher sig 3 3D7D41E3 2004-02-22 Atom Smasher sub 2048g/1E88BF71 2003-10-04 [expires: 2005-01-26] sig 3D7D41E3 2003-10-04 Atom Smasher %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% trying to include a policy URL (XYZ) into a clearsigned message tells me: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% $ date | gpg --sig-policy-url XYZ --clearsign You need a passphrase to unlock the secret key for user: "Atom Smasher " 1024-bit DSA key, ID 3D7D41E3, created 2003-10-04 gpg: writing to stdout - -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Fri Apr 9 15:08:54 EDT 2004 gpg: can't put a policy URL into v3 (PGP 2.x style) signatures gpg: DSA signature from: "3D7D41E3 Atom Smasher " - -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFAdvTKnCgLvz19QeMRAiZNAJ44xLs1X8J4oaXGF0PcVUZZ6unjFQCfQeGk Aoyqh/skv/lanScpeVzNiSA= =UQ4V - -----END PGP SIGNATURE----- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% there error is: gpg: can't put a policy URL into v3 (PGP 2.x style) signatures ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Patriotism is your conviction that this country is superior to all other countries because you were born in it." -- George Bernard Shaw -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdvX8nCgLvz19QeMRAtVBAKCRJ5gK828G9SWzroVTXt6ZFutHwACeLwQl 158KflfpLqXH21d0WjXKpEc= =zPI7 -----END PGP SIGNATURE----- From DougB at dougbarton.net Fri Apr 9 22:22:14 2004 From: DougB at dougbarton.net (Doug Barton) Date: Fri Apr 9 21:20:14 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> <20040409182638.GA2867@jabberwocky.com> <20040409185041.GG14052@dust.uchicago.edu> Message-ID: <407705F6.3020907@dougbarton.net> Atom 'Smasher' wrote: > $ date | gpg --sig-policy-url XYZ --clearsign I'm not sure "XYZ" qualifies as a URL, but I think that the actual problem might be that you have something in your gpg.conf that says "produce v3 signatures." Doug > You need a passphrase to unlock the secret key for > user: "Atom Smasher " > 1024-bit DSA key, ID 3D7D41E3, created 2003-10-04 > > gpg: writing to stdout > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Fri Apr 9 15:08:54 EDT 2004 > gpg: can't put a policy URL into v3 (PGP 2.x style) signatures > gpg: DSA signature from: "3D7D41E3 Atom Smasher " > - _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users From dshaw at jabberwocky.com Fri Apr 9 21:42:09 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 9 21:39:25 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> <20040409182638.GA2867@jabberwocky.com> <20040409185041.GG14052@dust.uchicago.edu> Message-ID: <20040409194209.GB2867@jabberwocky.com> On Fri, Apr 09, 2004 at 03:14:00PM -0400, Atom 'Smasher' wrote: > > You're presumably getting this error not because you can't clearsign, > > but because you can't have a policy URL or notation in a v3 key. > ======================== > > but i should be able to include a policy URL in a data signature, and i am > using a v4 key. You probably have --force-v3-sigs or one of the --pgpX options set. David From dshaw at jabberwocky.com Fri Apr 9 21:51:08 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 9 21:48:24 2004 Subject: notation data & policy URL In-Reply-To: References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> Message-ID: <20040409195108.GC2867@jabberwocky.com> On Fri, Apr 09, 2004 at 12:49:14PM -0400, Atom 'Smasher' wrote: > > > is there anything wrong with using the policy URL for non-URL text? > > > > Yes. Policy URLs are URLs. > > > > > example: i sign my brothers key while meeting for lunch... i could ~just~ > > > sign the key and leave it at that, or... i could specify the conditions of > > > the signature: > > > This is my brother. I've known him since 1976 and verified his key > > > in person. > > > > > > is there a recommended way to include that information directly in my > > > signature on his key? > > > > Notations. > > > > Still, if you include all that information in the signature, you're > > making the signature larger. A policy URL at least is small. > ================================= > > i still don't 'get' why the notation data is in the form: foo=bar > why not have just 'foo', or 'bar'? Because the standard says so :) > also, can a policy url and/or notation data be included in a self > signature? Sure, why not? Self-signatures are signatures too. David From gnupg at kubieziel.de Fri Apr 9 23:00:18 2004 From: gnupg at kubieziel.de (Jens Kubieziel) Date: Fri Apr 9 22:57:30 2004 Subject: --refresh-keys stops working while processing In-Reply-To: <200404072128.XAA00643@vulcan.xs4all.nl> References: <20040407190230.GC1872@kubieziel.de> <200404072128.XAA00643@vulcan.xs4all.nl> Message-ID: <20040409210018.GB2219@kubieziel.de> * Johan Wevers schrieb am 2004-04-07 23:28:27: > Jens Kubieziel wrote: > > >After waiting some 10-20 minutes I killed the process. When I retried it, > >the gpg-process stopped at approximately 110 keys. I can reproduce this > >error for every "--refresh-keys". > >What can be the reason for this? (I was only the whole time.) > > I have the same. I assume it's a long delay in the keyserver response. > When I wait very long the process will finish. It seemed to be. I chose another keyserver and it worked without further problems. -- Jens Kubieziel http://www.kubieziel.de A budget is something we go without to stay within. From atom-gpg at suspicious.org Sat Apr 10 00:51:48 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 10 00:49:07 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: <407705F6.3020907@dougbarton.net> References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> <20040409182638.GA2867@jabberwocky.com> <20040409185041.GG14052@dust.uchicago.edu> <407705F6.3020907@dougbarton.net> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > I'm not sure "XYZ" qualifies as a URL, but I think that the actual > problem might be that you have something in your gpg.conf that says > "produce v3 signatures." ==================================== nothing of the sort.... the only uncommented lines in my conf are: keyserver hkp://subkeys.pgp.net verbose compress-level 9 default-key 3D7D41E3 comment What is this gibberish? - http://atom.smasher.org/links/#digital_signatures can anyone else duplicate this behavior in 1.2.4? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "They tell us that we live in a great free republic; that our institutions are democratic; that we are a free and self-governing people. That is too much, even for a joke. Wars throughout history have been waged for conquest and plunder. And that is war in a nutshell. The master class has always declared the wars; the subject class has always fought the battles." -- Eugene V. Debs, 1918 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdykInCgLvz19QeMRAthQAJ9PbRa7pCWOBgHA/v2RSIN4aFLKyACfcRK7 Nu5ITFIgYTe3tNvDTxuPe/s= =JpqD -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Sat Apr 10 01:06:27 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Sat Apr 10 01:03:48 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <20040409120946.GE25930@jabberwocky.com> <20040409182638.GA2867@jabberwocky.com> <20040409185041.GG14052@dust.uchicago.edu> <407705F6.3020907@dougbarton.net> Message-ID: <20040409230626.GA3931@jabberwocky.com> On Fri, Apr 09, 2004 at 06:51:48PM -0400, Atom 'Smasher' wrote: > > I'm not sure "XYZ" qualifies as a URL, but I think that the actual > > problem might be that you have something in your gpg.conf that says > > "produce v3 signatures." > ==================================== > > nothing of the sort.... the only uncommented lines in my conf are: > keyserver hkp://subkeys.pgp.net > verbose > compress-level 9 > default-key 3D7D41E3 > comment What is this gibberish? - http://atom.smasher.org/links/#digital_signatures Add: no-force-v3-sigs David From atom-gpg at suspicious.org Sat Apr 10 01:15:28 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 10 01:12:38 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Add: > > no-force-v3-sigs =========================== that does the trick. why does it default to v3? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "There's enough on this planet for everyone's needs but not for everyone's greed" -- Mahatma Gandhi -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdy5GnCgLvz19QeMRAp5XAJ0VG7C6N0Y1KQWw7E3WIgOa2w1gHACfTurk q4pUji/F2MhXg4Crzyrw2hg= =2qxh -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Sat Apr 10 01:30:37 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Sat Apr 10 01:27:53 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: Message-ID: <20040409233037.GB3931@jabberwocky.com> On Fri, Apr 09, 2004 at 07:15:28PM -0400, Atom 'Smasher' wrote: > > Add: > > > > no-force-v3-sigs > =========================== > > that does the trick. > > why does it default to v3? PGP compatibility. No version of PGP before 8 can reliably handle v4 signatures. David From atom-gpg at suspicious.org Sat Apr 10 01:38:37 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 10 01:35:55 2004 Subject: notation data & policy URL In-Reply-To: <20040409195108.GC2867@jabberwocky.com> References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> <20040409195108.GC2867@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > i still don't 'get' why the notation data is in the form: foo=bar > > why not have just 'foo', or 'bar'? > > Because the standard says so :) ======================= ok... next question: why does the standard say so? it seems odd that instead of being able to specify: this is how i verified the key one has to, instead, specify: X@Y=this is how i verified the key > > also, can a policy url and/or notation data be included in a self > > signature? > > Sure, why not? Self-signatures are signatures too. ======================= how would one do that? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "In a way, risking climate change is even more frightening than playing Russian roulette... but with the pistol pointed at the head of one's child..." -- Stephen J. Decanio, The Economics of Climate Change -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAdzQCnCgLvz19QeMRAjvHAJwK7vgRuke9QYHGDQvlZYoZANjWDQCgnyVN Xqqq8it0HpyG+FeqEHgY3oY= =sIvs -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Sat Apr 10 01:48:26 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Sat Apr 10 01:45:45 2004 Subject: notation data & policy URL In-Reply-To: References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> <20040409195108.GC2867@jabberwocky.com> Message-ID: <20040409234826.GA5454@jabberwocky.com> On Fri, Apr 09, 2004 at 07:38:37PM -0400, Atom 'Smasher' wrote: > > > i still don't 'get' why the notation data is in the form: foo=bar > > > why not have just 'foo', or 'bar'? > > > > Because the standard says so :) > ======================= > > ok... next question: why does the standard say so? > > it seems odd that instead of being able to specify: > this is how i verified the key > one has to, instead, specify: > X@Y=this is how i verified the key Because you want to say "this is how I verified the key". Someone else wants to say "this key belongs to me". Someone else wants to embed unprintable binary data. Someone else wants to give his phone number. Thus, a key=value pair. Notations are a general purpose extension mechanism. They are not a policy URL alternative. > > > also, can a policy url and/or notation data be included in a self > > > signature? > > > > Sure, why not? Self-signatures are signatures too. > ======================= > > how would one do that? Same way one would do it for a non self-signature. cert-policy-url or cert-notation. David From boldyrev+nospam at cgitftp.uiggm.nsc.ru Sat Apr 10 06:21:35 2004 From: boldyrev+nospam at cgitftp.uiggm.nsc.ru (Ivan Boldyrev) Date: Sat Apr 10 06:37:02 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature References: <40760494.5080109@coldstonelabs.org> <20040409120946.GE25930@jabberwocky.com> <20040409182638.GA2867@jabberwocky.com> <20040409185041.GG14052@dust.uchicago.edu> <20040409194209.GB2867@jabberwocky.com> Message-ID: On 8710 day of my life David Shaw wrote: > On Fri, Apr 09, 2004 at 03:14:00PM -0400, Atom 'Smasher' wrote: >> > You're presumably getting this error not because you can't clearsign, >> > but because you can't have a policy URL or notation in a v3 key. >> ======================== >> >> but i should be able to include a policy URL in a data signature, and i am >> using a v4 key. > > You probably have --force-v3-sigs or one of the --pgpX options set. AFAIR, --force-v3-sigs is default (at least in some versions). I had to use no-force-v3-sigs in my gpg.conf. -- Ivan Boldyrev Violets are red, Roses are blue. // I'm schizophrenic, And so am I. From atom-gpg at suspicious.org Sat Apr 10 08:59:16 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 10 08:56:35 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: <20040409233037.GB3931@jabberwocky.com> References: <20040409233037.GB3931@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > PGP compatibility. No version of PGP before 8 can reliably handle v4 > signatures. ======================= then this looks like a typo in the man page... --force-v3-sigs --no-force-v3-sigs OpenPGP states that an implementation should generate v4 sig- natures but PGP versions 5 through 7 only recognize v4 signa- tures on key material. This option forces v3 signatures for signatures on data. Note that this option overrides --ask- sig-expire, as v3 signatures cannot have expiration dates. --no-force-v3-sigs disables this option. also, this seems like a deviation from the usual behavior, that if there's a "--force-xyz" and a "--no-force-xyz", the "--no-force-xyz" is usually the default, unless otherwise specified. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "The incidence of disease has increased in proportion to the progress of science." -- Akbarali Jetha -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAd5tInCgLvz19QeMRAgLvAJwMb+E/FbknynA9/HxoYfnVgQrhoACfcI56 tRr4LQXNwcB1pqLCzeA8bH8= =X+/o -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Sat Apr 10 09:15:33 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 10 09:12:47 2004 Subject: notation data & policy URL In-Reply-To: <20040409234826.GA5454@jabberwocky.com> References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> <20040409195108.GC2867@jabberwocky.com> <20040409234826.GA5454@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > > > also, can a policy url and/or notation data be included in a self > > > > signature? > > > > > > Sure, why not? Self-signatures are signatures too. > > ======================= > > > > how would one do that? > > Same way one would do it for a non self-signature. cert-policy-url or > cert-notation. ================================ it took me a while to figure it out using existing keys... if all else fails, use --expert. however, this seems strange... if i generate a key like this: gpg --cert-policy 'http://test-policy' --cert-notation 'A@B=test notation' --gen-key it places the notation and URL on both the signing key and the encryption sub-key: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% $ gpg --list-key --show-policy --show-notation 2A42B922 pub 1024D/2A42B922 2004-04-10 testing sig 3 PN 2A42B922 2004-04-10 testing Signature policy: http://test-policy Signature notation: A@B=test notation sub 1024g/5A5D67E7 2004-04-10 sig PN 2A42B922 2004-04-10 testing Signature policy: http://test-policy Signature notation: A@B=test notation %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% while i do understand the logic (they ~are~ both certification signatures), it seems redundant... shouldn't it be sufficient to only add those packets to the signing key? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Thoughtcrime was not a thing that could be concealed forever. You might dodge successfully for a while, even for years, but sooner or later they were bound to get you." -- George Orwell, "1984" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iD8DBQFAd58anCgLvz19QeMRAkMpAKCYvNVMxa+OFC4e5Y+pI3m6tnFUKwCeNvjp KF1HqSxRts0nShAoLd/H0aY= =lgY9 -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Sat Apr 10 09:39:37 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 10 09:36:52 2004 Subject: notation data & policy URL In-Reply-To: <20040409234826.GA5454@jabberwocky.com> References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> <20040409195108.GC2867@jabberwocky.com> <20040409234826.GA5454@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > ok... next question: why does the standard say so? > > > > it seems odd that instead of being able to specify: > > this is how i verified the key > > one has to, instead, specify: > > X@Y=this is how i verified the key > > Because you want to say "this is how I verified the key". Someone > else wants to say "this key belongs to me". Someone else wants to > embed unprintable binary data. Someone else wants to give his phone > number. > > Thus, a key=value pair. Notations are a general purpose extension > mechanism. They are not a policy URL alternative. =========================================== i still don't get it... is there a formal way that things are supposed to be done within notation data? what are some examples of proper uses (and improper abuses) of notation data? is it meant to someday replace other packets? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Patriotism is your conviction that this country is superior to all other countries because you were born in it." -- George Bernard Shaw -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB3pL4ACgkQnCgLvz19QeP6XgCffSwb+aWd4MfjA56Y8sRDH0a6 dP8An2qen7EHGjLDM9pKyBCDuZpzR1kE =al8o -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Sat Apr 10 13:00:17 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Sat Apr 10 12:57:33 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: References: <20040409233037.GB3931@jabberwocky.com> Message-ID: <20040410110016.GD5454@jabberwocky.com> On Sat, Apr 10, 2004 at 02:59:16AM -0400, Atom 'Smasher' wrote: > > PGP compatibility. No version of PGP before 8 can reliably handle v4 > > signatures. > ======================= > > then this looks like a typo in the man page... > > --force-v3-sigs > --no-force-v3-sigs > OpenPGP states that an implementation should generate v4 sig- > natures but PGP versions 5 through 7 only recognize v4 signa- > tures on key material. This option forces v3 signatures for > signatures on data. Note that this option overrides --ask- > sig-expire, as v3 signatures cannot have expiration dates. > --no-force-v3-sigs disables this option. Where is the typo? > also, this seems like a deviation from the usual behavior, that if there's > a "--force-xyz" and a "--no-force-xyz", the "--no-force-xyz" is usually > the default, unless otherwise specified. The default is just the one that is most appropriate. There is no standard behavior to have the "no" value be the default. Note "--escape-from", "--mangle-dos-filenames", and "--ask-cert-level" are also default-to-yes. David From dshaw at jabberwocky.com Sat Apr 10 13:09:16 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Sat Apr 10 13:06:36 2004 Subject: notation data & policy URL In-Reply-To: References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> <20040409195108.GC2867@jabberwocky.com> <20040409234826.GA5454@jabberwocky.com> Message-ID: <20040410110916.GE5454@jabberwocky.com> On Sat, Apr 10, 2004 at 03:15:33AM -0400, Atom 'Smasher' wrote: > it places the notation and URL on both the signing key and the encryption > sub-key: > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > $ gpg --list-key --show-policy --show-notation 2A42B922 > pub 1024D/2A42B922 2004-04-10 testing > sig 3 PN 2A42B922 2004-04-10 testing > Signature policy: http://test-policy > Signature notation: A@B=test notation > sub 1024g/5A5D67E7 2004-04-10 > sig PN 2A42B922 2004-04-10 testing > Signature policy: http://test-policy > Signature notation: A@B=test notation > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > while i do understand the logic (they ~are~ both certification > signatures), it seems redundant... shouldn't it be sufficient to only add > those packets to the signing key? Nope. It's dangerous for a program to try and be "smart" and guess what you really mean. You asked for a notation in both certs, so you got it. Notations are a general-purpose extension mechanism. To do this, they need to be usable anywhere a signature is generated. David From dshaw at jabberwocky.com Sat Apr 10 13:13:00 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Sat Apr 10 13:10:21 2004 Subject: notation data & policy URL In-Reply-To: References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> <20040409195108.GC2867@jabberwocky.com> <20040409234826.GA5454@jabberwocky.com> Message-ID: <20040410111300.GF5454@jabberwocky.com> On Sat, Apr 10, 2004 at 03:39:37AM -0400, Atom 'Smasher' wrote: > > Because you want to say "this is how I verified the key". Someone > > else wants to say "this key belongs to me". Someone else wants to > > embed unprintable binary data. Someone else wants to give his phone > > number. > > > > Thus, a key=value pair. Notations are a general purpose extension > > mechanism. They are not a policy URL alternative. > =========================================== > > i still don't get it... > > is there a formal way that things are supposed to be done within notation > data? what are some examples of proper uses (and improper abuses) of > notation data? is it meant to someday replace other packets? The draft defines it thus: Notation names are arbitrary strings encoded in UTF-8. They reside two name spaces: The IETF name space and the user name space. The IETF name space is registered with IANA. These names MUST NOT contain the "@" character (0x40) is this is a tag for the user name space. Names in the user name space consist of a UTF-8 string tag followed by "@" followed by a DNS domain name. Note that the tag MUST NOT contain an "@" character. For example, the "sample" tag used by Example Corporation could be "sample@example.com". Names in a user space are owned and controlled by the owners of that domain. Obviously, it's of bad form to create a new name in a DNS space that you don't own. Since the user name space is in the form of an email address, implementers MAY wish to arrange for that address to reach a person who can be consulted about the use of the named tag. Note that due to UTF-8 encoding, not all valid user space name tags are valid email addresses. Once you have a @ character in the key, you are explicitly saying that the format of the value is defined by you. You can thus put anything you like in there. Since the IETF has not defined any "official" keys, GnuPG requires you to have that @ character. David From boldyrev+nospam at cgitftp.uiggm.nsc.ru Sat Apr 10 15:05:20 2004 From: boldyrev+nospam at cgitftp.uiggm.nsc.ru (Ivan Boldyrev) Date: Sat Apr 10 15:13:54 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature References: <20040409233037.GB3931@jabberwocky.com> <20040410110016.GD5454@jabberwocky.com> Message-ID: On 8710 day of my life David Shaw wrote: > On Sat, Apr 10, 2004 at 02:59:16AM -0400, Atom 'Smasher' wrote: >> > PGP compatibility. No version of PGP before 8 can reliably handle v4 >> > signatures. >> ======================= >> >> then this looks like a typo in the man page... >> >> --force-v3-sigs >> --no-force-v3-sigs >> OpenPGP states that an implementation should generate v4 sig- >> natures but PGP versions 5 through 7 only recognize v4 signa- ^^ >> tures on key material. This option forces v3 signatures for >> signatures on data. Note that this option overrides --ask- >> sig-expire, as v3 signatures cannot have expiration dates. >> --no-force-v3-sigs disables this option. > > Where is the typo? -- Ivan Boldyrev Is 'morning' a gerund? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 480 bytes Desc: not available Url : /pipermail/attachments/20040410/cd228f01/attachment.bin From atom-gpg at suspicious.org Sat Apr 10 20:33:04 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 10 20:30:20 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature In-Reply-To: <20040410110016.GD5454@jabberwocky.com> References: <20040409233037.GB3931@jabberwocky.com> <20040410110016.GD5454@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > > PGP compatibility. No version of PGP before 8 can reliably handle v4 > > > signatures. > > ======================= > > > > then this looks like a typo in the man page... > > > > --force-v3-sigs > > --no-force-v3-sigs > > OpenPGP states that an implementation should generate v4 sig- > > natures but PGP versions 5 through 7 only recognize v4 signa- > > tures on key material. This option forces v3 signatures for > > signatures on data. Note that this option overrides --ask- > > sig-expire, as v3 signatures cannot have expiration dates. > > --no-force-v3-sigs disables this option. > > Where is the typo? ======================================== logic dictates that these statements can not both be correct: * "No version of PGP before 8 can reliably handle v4 signatures." -- dshaw * "PGP versions 5 through 7 only recognize v4 signatures on key material." -- gpg (1.2.4) man page assuming that you [dave] are correct, then the man page must be wrong. > > also, this seems like a deviation from the usual behavior, that if there's > > a "--force-xyz" and a "--no-force-xyz", the "--no-force-xyz" is usually > > the default, unless otherwise specified. > > The default is just the one that is most appropriate. There is no > standard behavior to have the "no" value be the default. Note > "--escape-from", "--mangle-dos-filenames", and "--ask-cert-level" are > also default-to-yes. ==================================== that seems like a slight (very slight!) deviation from the RFC (5.2): Implementations SHOULD generate V4 signatures. Implementations MAY generate a V3 signature that can be verified by PGP 2.6.x. obviously not a catastrophic deviation, but i didn't see it mentioned anywhere... not even doc/gnupg/OpenPGP. i would think that the same thing could be accomplished in a more RFC-compliant (and easier to figure out) way by making the default behavior to use v4 signatures, but including "force-v3-sigs" in the default config file. (IMHO) that would keep the application closer to the RFC ideal, but in practice it would still be compatible with older versions of PGP(tm). if there are only a few yes/no options that default "yes", then i'd like to request that the man page specify that those options (that you list above) default to "yes", while all of the other yes/no options default to "no". or... each yes/no option should explicitly state (in the man page) what it defaults to, and if it's overridden in the default config file. the default config file would also be a good place to explain why a default would be overridden, as with the case of why v3 signatures are generated. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "I spent 33 years and four months in active service in the country's most agile military force, the Marines. I served in all ranks from second lieutenant to major general. And during that period I spent most of my time being a high-class muscle man for Big Business, for Wall Street and the bankers. In short, I was a racketeer, a gangster for capitalism. "I suspected I was just part of a racket at the time. Now I am sure of it. Like all members of the military profession I never had an original thought until I left the service. My mental faculties remained in suspended animation while I obeyed the orders of the higher-ups. This is typical with everyone in the military service. "Thus I helped make Mexico, and especially Tampico, safe for American oil interests in 1914. I helped make Haiti and Cuba a decent place for the National City Bank boys to collect revenue in. I helped in the raping of half-a-dozen Central American republics for the benefit of Wall Street. The record of racketeering is long. I helped purify Nicaragua for the international banking house of Brown Brothers and Co. in 1909-1912. I brought light to the Dominican Republic for the sugar interests in 1916. I helped make Honduras "right" for American fruit companies in 1903. In China in 1927 I helped see to it that Standard Oil went its way unmolested. "During those years, I had, as the boys in the back room would say, a swell racket. I was rewarded with honors, medals, and promotion. Looking back on it, I feel that I might have given Al Capone a few hints. The best he could do was to operate a racket in three city districts. The Marines operated on three continents." -- Smedley D. Butler, (1881-1940) Major Gen U.S. Marines -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB4PeQACgkQnCgLvz19QeOregCfY6yD3hqjaP4t5DLeF3DP+Xjk uW0AoI8W7WacjEup/YUmOJEhrcBqomx+ =ZItX -----END PGP SIGNATURE----- From jam at athene.jamux.com Sat Apr 10 21:37:10 2004 From: jam at athene.jamux.com (John A. Martin) Date: Sat Apr 10 21:34:28 2004 Subject: gpg: can't put a policy URL into v3 (PGP 2.x style) signature References: <20040409233037.GB3931@jabberwocky.com> <20040410110016.GD5454@jabberwocky.com> Message-ID: <87fzbbsjyx.fsf@athene.jamux.com> >>>>> "Atom" == Atom >>>>> "Re: gpg: can't put a policy URL into v3 (PGP 2.x style) signature" >>>>> Sat, 10 Apr 2004 14:33:04 -0400 (EDT) >> > > PGP compatibility. No version of PGP before 8 can reliably >> > > handle v4 signatures. >> > ======================= >> > >> > then this looks like a typo in the man page... >> > >> > --force-v3-sigs --no-force-v3-sigs >> > OpenPGP states that an implementation should >> > generate v4 sig- natures but PGP versions 5 >> > through 7 only recognize v4 signa- tures on >> > key material. This option forces v3 >> > signatures for signatures on data. Note that >> > this option overrides --ask- sig-expire, as v3 >> > signatures cannot have expiration dates. >> > --no-force-v3-sigs disables this option. >> >> Where is the typo? Atom> ======================================== Atom> logic dictates that these statements can not both be Atom> correct: Atom> * "No version of PGP before 8 can reliably handle v4 Atom> signatures." Atom> -- dshaw Atom> * "PGP versions 5 through 7 only recognize v4 signatures on Atom> key material." Atom> -- gpg (1.2.4) man page Would ... PGP versions 5 through 7 recognize v4 signatures only on key material. be more readily understood? jam -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 154 bytes Desc: not available Url : /pipermail/attachments/20040410/81a87c3a/attachment.bin From atom-gpg at suspicious.org Sat Apr 10 23:29:14 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 10 23:26:31 2004 Subject: notation data & policy URL In-Reply-To: <20040410111300.GF5454@jabberwocky.com> References: <40760494.5080109@coldstonelabs.org> <20040409121227.GF25930@jabberwocky.com> <20040409195108.GC2867@jabberwocky.com> <20040409234826.GA5454@jabberwocky.com> <20040410111300.GF5454@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Once you have a @ character in the key, you are explicitly saying that > the format of the value is defined by you. You can thus put anything > you like in there. Since the IETF has not defined any "official" > keys, GnuPG requires you to have that @ character. ========================================== ok... i may be slow, but i'm catching on.... ;) so, there is an IETF name space and a user name space. any name that includes "@" is part of the user name space, and any name that does not include "@" is part of the IETF name space. how am i doing so far? now, since the IETF hasn't defined anything to be used in their name space, that leaves the user name space, which should (must?) be in the form of an email address (name@domain). am i still doing ok? so, what does IETF have planned for their notation data within openPGP keys? of what practical use is it for user data? are there any examples of some legitimate uses or planned uses? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "I don't know anything about music. In my line you don't have to." -- Elvis Presley -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB4ZzEACgkQnCgLvz19QeMb4ACdEDkHED+fXw8A/tEpmyd90riJ Y+AAn3+8HaPSUBHfNPuLeC99mD3KFMgB =jEpX -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Mon Apr 12 09:25:59 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Mon Apr 12 09:23:27 2004 Subject: stripping PGP data from cleartext Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 is there a way to pipe data through gpg and retain information that's above and/or below the pgp-header info? example: abc data --PGP HEADER-- signed data --PGP HEADER-- pgp signature data --END PGP DATA-- xyz data normally, if that message is piped into gpg, the only output (on STDOUT) would be the "signed data". i also know all of the reasons why it would normally be sensible to strip out the "abc data" and "xyz data". i do have an application in mind that would require data that's outside of the signed data area to be passed through gpg, but i'm not sure if gpg can do that. any ideas? thanks... ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "We have a saying in our company. Our competitors are our friends. Our customers are the enemy." -- James Randall, Archer Daniels Midland Corporation, ("ADM, Supermarket to the world") quoted in Fortune magazine 4/26/99 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB6RIwACgkQnCgLvz19QePv/gCeMlznzKSfI73GMl4oTq6U+zJY 5KUAnjkgxBCpG5Lx2Vk25AmH3aVhK4fX =2cak -----END PGP SIGNATURE----- From m at tthias.net Mon Apr 12 09:37:12 2004 From: m at tthias.net (Matthias Wimmer) Date: Mon Apr 12 09:34:22 2004 Subject: stripping PGP data from cleartext In-Reply-To: References: Message-ID: <20040412073712.GA7489@nbg2.amessage.info> Hi! Atom 'Smasher' schrieb am 2004-04-12 03:25:59: > i do have an application in mind that would require data that's outside of > the signed data area to be passed through gpg, but i'm not sure if gpg can > do that. any ideas? Stripe the unsigned data in your application yourself and you will also know which part of the data you are processing has been signed and which has not. If gpg stripes it you don't. Tot kijk Matthias -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : /pipermail/attachments/20040412/6d079b64/attachment.bin From atom-gpg at suspicious.org Mon Apr 12 10:04:29 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Mon Apr 12 10:01:44 2004 Subject: stripping PGP data from cleartext In-Reply-To: <20040412073712.GA7489@nbg2.amessage.info> References: <20040412073712.GA7489@nbg2.amessage.info> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Stripe the unsigned data in your application yourself and you will also > know which part of the data you are processing has been signed and which > has not. If gpg stripes it you don't. ===================== yeah... if i have to, i can do this with sed, but i'm wondering if it can be done just with gpg. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "If Jesus Christ were to come today, people would not even crucify him. They would ask him to dinner, and hear what he had to say, and make fun of it." -- Thomas Carlyle -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB6TZIACgkQnCgLvz19QeNncQCcC6RJZ5e7h11CssSF0U72L7jJ ISMAnR2b745F9tOcI37GU8/x5OE167/l =BA+C -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Tue Apr 13 02:00:09 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Tue Apr 13 01:57:26 2004 Subject: notation data & policy URL In-Reply-To: References: <20040409121227.GF25930@jabberwocky.com> <20040409195108.GC2867@jabberwocky.com> <20040409234826.GA5454@jabberwocky.com> <20040410111300.GF5454@jabberwocky.com> Message-ID: <20040413000008.GA17984@jabberwocky.com> On Sat, Apr 10, 2004 at 05:29:14PM -0400, Atom 'Smasher' wrote: > > Once you have a @ character in the key, you are explicitly saying that > > the format of the value is defined by you. You can thus put anything > > you like in there. Since the IETF has not defined any "official" > > keys, GnuPG requires you to have that @ character. > ========================================== > > ok... i may be slow, but i'm catching on.... ;) > > so, there is an IETF name space and a user name space. any name that > includes "@" is part of the user name space, and any name that does not > include "@" is part of the IETF name space. > > how am i doing so far? Yes. > now, since the IETF hasn't defined anything to be used in their name > space, that leaves the user name space, which should (must?) be in the > form of an email address (name@domain). Yes. > so, what does IETF have planned for their notation data within openPGP > keys? of what practical use is it for user data? are there any examples of > some legitimate uses or planned uses? There have been occasional suggestions, but none have been adopted. It does serve as a handy response to some of the more eccentric suggestions when people want to change OpenPGP in various ways: just use a notation. David From malsyned at cif.rochester.edu Tue Apr 13 03:56:10 2004 From: malsyned at cif.rochester.edu (Dennis Lambe Jr.) Date: Tue Apr 13 03:53:22 2004 Subject: Use of public key servers In-Reply-To: <40760494.5080109@coldstonelabs.org> References: <40760494.5080109@coldstonelabs.org> Message-ID: <1081480542.7647.20.camel@localhost> On Thu, 2004-04-08 at 22:04, Russell Valentine wrote: > 1) You can't prove a key from the public key server is really that > persons public key, you still have to validate it some way. So you can > get the key personally from that person in some way and at that time > validate the key. Instead of using the key even thought it might not be his. This isn't always true. If someone signs your key, what they're doing is asserting "This key belongs to the person named in the ID." From then on, as long as someone trusts one of your key's signers, they can trust your key that they downloaded from the public server. At some point, someone will have to have used an out-of-band method to verify someone else's key, but that person doesn't have to be you or I, just someone one of us knows and trusts, or someone that someone one of us trusts trusts, or... --D -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 279 bytes Desc: This is a digitally signed message part Url : /pipermail/attachments/20040412/531ae802/attachment.bin From atom-gpg at suspicious.org Tue Apr 13 08:43:05 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Tue Apr 13 08:40:22 2004 Subject: pgp/mime vs in-line pgp Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 what are the (dis)advantages of pgp/mime vs in-line pgp? of course, pgp/mime is an *official* standard, while in-line pgp is an *unofficial* standard. why isn't in-line pgp *officially* recognized as an email standard? two reasons i like in-line pgp: * i've only heard of one MUA that chokes on in-line pgp, but plenty of MUAs can't handle pgp/mime. * i've *never* had an in-line signature cause problems with someone's poorly configured virus filter. ok... a third reason that i like in-line pgp is because that's all i have with pine. disclaimer: yes, i've tried mutt a bunch of times... but i have pine set up to do things that mutt can't emulate (no, really!). ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Microsoft is a bad neighbour, whose allowed their yard to fill with filth and trash, subjecting the people around them to the vermin and roaches that breed within their unkempt property. It is on this day that the internet will begin to sputter and fail in places due to the tremendous burdon Microsofts incompetence has placed upon it. Microsoft's products spew pollution into the information space like a burning mountain of tires." -- Unknown -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB7i/0ACgkQnCgLvz19QePaLACeIHksx+Nb7iB5kHtIWsa12imv QYwAnjfwuilsdlY/sTwx0NezllNcL3/H =27GT -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Tue Apr 13 09:54:45 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Tue Apr 13 09:52:00 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: Message-ID: <200404130954.50900@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday 13 April 2004 08.43, Atom 'Smasher' wrote: > * i've only heard of one MUA that chokes on in-line pgp, but plenty > of MUAs can't handle pgp/mime. Plenty? I know only one MUA that chokes on PGP/MIME. Other's don't support PGP, but most still properly display the email. > > * i've *never* had an in-line signature cause problems with someone's > poorly configured virus filter. True. The big problem with inline is non-ascii text: when the senders charset differs from the receivers, the receivers MUA is likely to convert the message before displaying/storing - breaking the signature (not an issue when the receiver's MUA directly supports GPG/PGP.) I'm not sure, but I can imagine that there may be problems with line wrapping, too, on conversions to/ftom quoted-printable. Then there's the obvious advantage of PGP/MIME when you need to send attachments. cheers - -- vbi - -- Dejar de comer por haber comido, no hay nada perdido. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkB7nMpgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6BlwAn2u2738LH/mcFimVB6D+jnYj YodmAKCz+/0sxiViLyLBI47sMht4IvP3FA== =KIj6 -----END PGP SIGNATURE----- From pt at radvis.nu Tue Apr 13 12:39:33 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 13 12:54:53 2004 Subject: openPGP vs x509 In-Reply-To: <8765cb7i3d.fsf@vigenere.g10code.de> References: <4073BBEB.6070203@gmx.de> <200404071414.26114@fortytwo.ch> <8765cb7i3d.fsf@vigenere.g10code.de> Message-ID: <6.0.3.0.2.20040413123533.026d2198@localhost> At 20:34 2004-04-07, Werner Koch wrote: >For a casual user it is far easier to create a PGP key than an X.509 >key/certificate. You can instantly do this with OpenPGP tools (ssh is >similar) whereas you have to go into lengthy discussions with OpenSSL >to create even just a self-signed certificate containing a usable >email attribute. OT? I have been looking for tools to make self-signed X.509 but haven't found any. Do you have any hints? Links? Rational: some people refuse using PGP-encryption/signing. Per Tunedal From pt at radvis.nu Tue Apr 13 12:57:53 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 13 12:55:07 2004 Subject: secrets lying around on the HD Message-ID: <6.0.3.0.2.20040413124005.0264d2d0@localhost> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I happened to notice that WinPT left some unencrypted temp-files on the HD. Is this causes by WinPT-tray or GPG? I found a setting "don't use any temp-files" in WinPT. Is there a similar option on GPG? What will actually happen when you encrypt a file? A very large file? BTW I tested the encryption in WindowsXP Pro, downloaded a "recovery tool" and could "recover" the encrypted files in 0,3 seconds when logged in as a different user. How? Are the keys left unencrypted on the HD? Advanced EFS Data Recovery: http://www.regnow.com/softsell/visitor.cgi?affiliate=17066&action=site&vendor=1170&ref=http://www.elcomsoft.com/aefsdr.html V?nligen Per Tunedal Civ. ing. Civ. ek. S:t Mickelsgatan 148 129 44 H?gersten Telefon: 08-646 34 83 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) - GPGrelay v0.94 iD8DBQFAe8fAITLMlZFNlMoRAl4dAKDCzs3zfFmwkFp4dggdcEHZbpJhHwCdExkV oE7micWcgS/6qbe0X+a4sKs= =Nytg -----END PGP SIGNATURE----- From pt at radvis.nu Tue Apr 13 13:02:42 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 13 12:59:41 2004 Subject: TLS client authentication, Re: openPGP vs x509 Message-ID: <6.0.3.0.2.20040413130234.026d1ec0@localhost> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 14:07 2004-04-07, you wrote: > >Just adding a data-point: This is also in use for government<->citizen >communication in Denmark (e.g. I used it to file my tax 'papers'), and >it works pretty good. > >/dossen > The same in Sweden when using some services at bank sites and government sites. Per Tunedal -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) - GPGrelay v0.94 iD8DBQFAe8jgITLMlZFNlMoRAkTZAJ4/UBCcU+YY0bl9wMPIHOmGmSSCcACePtyZ SvC1iUOnL54tq2k2l3dC2/4= =Gcr1 -----END PGP SIGNATURE----- From wk at gnupg.org Tue Apr 13 13:43:52 2004 From: wk at gnupg.org (Werner Koch) Date: Tue Apr 13 13:26:33 2004 Subject: secrets lying around on the HD In-Reply-To: <6.0.3.0.2.20040413124005.0264d2d0@localhost> (Per Tunedal Casual's message of "Tue, 13 Apr 2004 12:57:53 +0200") References: <6.0.3.0.2.20040413124005.0264d2d0@localhost> Message-ID: <87isg4rtl3.fsf@vigenere.g10code.de> On Tue, 13 Apr 2004 12:57:53 +0200, Per Tunedal Casual said: > I happened to notice that WinPT left some unencrypted temp-files on the HD. > Is this causes by WinPT-tray or GPG? I found a setting "don't use any > temp-files" in WinPT. Is there a similar option on GPG? What will actually > happen when you encrypt a file? A very large file? A design goal of GnuPG is not to use temporary files. You migtht encounter temporary files with GnuPG on some platforms, but those are only used for communication between gpg and the keyserver modules - that data is public anyway. Don't known about WinPT. Werner From dshaw at jabberwocky.com Tue Apr 13 13:40:25 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Tue Apr 13 13:37:44 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: Message-ID: <20040413114024.GC4400@jabberwocky.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tue, Apr 13, 2004 at 02:43:05AM -0400, Atom 'Smasher' wrote: > what are the (dis)advantages of pgp/mime vs in-line pgp? > > of course, pgp/mime is an *official* standard, while in-line pgp is an > *unofficial* standard. why isn't in-line pgp *officially* recognized as an > email standard? Inline PGP is absolutely an official standard. See RFC-2440. It's just a different official standard than PGP/MIME (RFC-3156). Inline PGP can't handle all the situations that PGP/MIME does, and it's not nearly as transparent as PGP/MIME, but it is a genuine official standard. David -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.3.6-cvs (GNU/Linux) Comment: Key available at http://www.jabberwocky.com/david/keys.asc iHEEARECADEFAkB70agqGGh0dHA6Ly93d3cuamFiYmVyd29ja3kuY29tL2Rhdmlk L2tleXMuYXNjAAoJEOJmXIdJ4cvJRZwAoOAvXfXT/+vEuY/dhAcy7Iwd8RRvAJ9z qQ//7YzNWCP51QqByJDi2PaWmQ== =8d0Q -----END PGP SIGNATURE----- From m at tthias.net Tue Apr 13 14:26:21 2004 From: m at tthias.net (Matthias Wimmer) Date: Tue Apr 13 14:23:41 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404130954.50900@fortytwo.ch> References: <200404130954.50900@fortytwo.ch> Message-ID: <20040413122618.GA3337@hermes.muc.charente.de> Hi Adrian! Adrian 'Dagurashibanipal' von Bidder schrieb am 2004-04-13 09:54:45: > > * i've only heard of one MUA that chokes on in-line pgp, but plenty > > of MUAs can't handle pgp/mime. > Plenty? I know only one MUA that chokes on PGP/MIME. Other's don't > support PGP, but most still properly display the email. If a MUA's MIME implementation is broken and can't display PGP/MIME, it can be seen as a reason to use PGP/MIME as well ;) > Then there's the obvious advantage of PGP/MIME when you need to send > attachments. An advantage I like as well is that PGP/MIME signed messages look better in MUAs that do not support PGP at all (but support MIME). Tot kijk Matthias -- Fon: +49-(0)70 0770 07770 http://matthias.wimmer.name/ HAM: DB1MW xmpp:mawis@charente.de -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : /pipermail/attachments/20040413/fc90725d/attachment.bin From avbidder at fortytwo.ch Tue Apr 13 14:34:59 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Tue Apr 13 14:32:12 2004 Subject: secrets lying around on the HD In-Reply-To: <6.0.3.0.2.20040413124005.0264d2d0@localhost> References: <6.0.3.0.2.20040413124005.0264d2d0@localhost> Message-ID: <200404131435.00908@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday 13 April 2004 12.57, Per Tunedal Casual wrote: > BTW I tested the encryption in WindowsXP Pro, downloaded a "recovery > tool" and could "recover" the encrypted files in 0,3 seconds when > logged in as a different user. How? Are the keys left unencrypted on > the HD? You will almost always have unencrypted copies unencrypted file contents lying around after encryption and deleting the unencrypted file: it is almost impossible to force the file system to really overwrite blocks of a file - writing to an existing file is allowed to allocate new blocks on the filesystem instead of overwriting the currently allocated blocks. To ensure that unencrypted file contents is really overwwritten, you'll have to read the filesystem code to understand how block allocation works (so you may come to the conclusion that a certain way of overwriting a file will never allocate new blocks), or you'll have to write a filesystem yourself, offering control about overwriting blocks to the application. (I *think* that there was some version of the Linux file system ext2 offering the option of overwriting deallocated blocks automatically. Or perhaps this was just a rumour - not sure at all.) The next step is paging: either disable paging entirely, or use an encrypted swap file/swap partition (with quite high performance cost, of course.) Or, as a compromise, code something up to automatically overwrite the swap partition on system shutdown (users of swap files run into the same problems as above.) The next thing to worry about will then be block re-allocation within the disc: all modern discs may reallocate disc blocks internally on some errors - and some tools might get data from there. (Once you are at this level of paranoia, you've probably got the budget to run all your systems from solid state disks - yank the battery out and nobody will ever recover anything.) greetings - -- vbi - -- You are what you see. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkB73nNgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6ooAAnAgiu3Rmcua+gE18EPs4RKqo tAuqAJ9iR7VqxYpG6vc87VY2GUsuQC4mQA== =3gOi -----END PGP SIGNATURE----- From pt at radvis.nu Tue Apr 13 14:55:33 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 13 14:52:29 2004 Subject: secrets lying around on the HD In-Reply-To: <87isg4rtl3.fsf@vigenere.g10code.de> References: <6.0.3.0.2.20040413124005.0264d2d0@localhost> <87isg4rtl3.fsf@vigenere.g10code.de> Message-ID: <6.0.3.0.2.20040413145414.026c6eb0@localhost> At 13:43 2004-04-13, you wrote: >On Tue, 13 Apr 2004 12:57:53 +0200, Per Tunedal Casual said: > >> I happened to notice that WinPT left some unencrypted temp-files on the HD. >> Is this causes by WinPT-tray or GPG? I found a setting "don't use any >> temp-files" in WinPT. Is there a similar option on GPG? What will actually >> happen when you encrypt a file? A very large file? > >A design goal of GnuPG is not to use temporary files. You migtht >encounter temporary files with GnuPG on some platforms, but those are >only used for communication between gpg and the keyserver modules - >that data is public anyway. > >Don't known about WinPT. > > > Werner What about virtual memory (SWAP-files) in Windows? Per Tunedal From thomas at northernsecurity.net Tue Apr 13 15:29:22 2004 From: thomas at northernsecurity.net (Thomas =?iso-8859-1?Q?Sj=F6gren?=) Date: Tue Apr 13 15:27:02 2004 Subject: secrets lying around on the HD In-Reply-To: <200404131435.00908@fortytwo.ch> References: <6.0.3.0.2.20040413124005.0264d2d0@localhost> <200404131435.00908@fortytwo.ch> Message-ID: <20040413132921.GD5658@northernsecurity.net> On Tue, Apr 13, 2004 at 02:34:59PM +0200, Adrian 'Dagurashibanipal' von Bidder wrote: > The next step is paging: either disable paging entirely, or use an > encrypted swap file/swap partition (with quite high performance cost, > of course.) I run several computers with encrypted swap-partition and /tmp and the performance cost isn't noticable. The only thing is that booting takes a couple of seconds longer, since the partitions is trashed and then re-encrypted (an ugly form on wiping one might say) but thats about it. /Thomas -- == thomas@northernsecurity.net | thomas@se.linux.org == Encrypted e-mails preferred | GPG KeyID: 114AA85C -- -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 481 bytes Desc: Digital signature Url : /pipermail/attachments/20040413/2b140b62/attachment.bin From thomas at northernsecurity.net Tue Apr 13 15:34:40 2004 From: thomas at northernsecurity.net (Thomas =?iso-8859-1?Q?Sj=F6gren?=) Date: Tue Apr 13 15:32:21 2004 Subject: secrets lying around on the HD In-Reply-To: <6.0.3.0.2.20040413145414.026c6eb0@localhost> References: <6.0.3.0.2.20040413124005.0264d2d0@localhost> <87isg4rtl3.fsf@vigenere.g10code.de> <6.0.3.0.2.20040413145414.026c6eb0@localhost> Message-ID: <20040413133440.GE5658@northernsecurity.net> On Tue, Apr 13, 2004 at 02:55:33PM +0200, Per Tunedal Casual wrote: > What about virtual memory (SWAP-files) in Windows? You're able clean the swap in Windows by enabling "Shutdown: Clear virtual memory pagefile" in you local security policy. /Thomas -- == thomas@northernsecurity.net | thomas@se.linux.org == Encrypted e-mails preferred | GPG KeyID: 114AA85C -- -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 481 bytes Desc: Digital signature Url : /pipermail/attachments/20040413/8de24805/attachment-0001.bin From wk at gnupg.org Tue Apr 13 15:50:19 2004 From: wk at gnupg.org (Werner Koch) Date: Tue Apr 13 15:36:33 2004 Subject: secrets lying around on the HD In-Reply-To: <6.0.3.0.2.20040413145414.026c6eb0@localhost> (Per Tunedal Casual's message of "Tue, 13 Apr 2004 14:55:33 +0200") References: <6.0.3.0.2.20040413124005.0264d2d0@localhost> <87isg4rtl3.fsf@vigenere.g10code.de> <6.0.3.0.2.20040413145414.026c6eb0@localhost> Message-ID: <877jwkrnqc.fsf@vigenere.g10code.de> On Tue, 13 Apr 2004 14:55:33 +0200, Per Tunedal Casual said: > What about virtual memory (SWAP-files) in Windows? That may indeed happen and we can't do anything about it. The only possible solution is to write a device driver and let it allocate memory for secrets. This memory will then be managed via ioctls. There is one such driver available but it is not Free Software so we have not looked deeper into it. There are currently rumors that the W32 API VirtualLock does indeed lock pages against paging - if this is true we will ad this to the next GnuPG version. Werner From pt at radvis.nu Tue Apr 13 19:37:42 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 13 19:34:46 2004 Subject: secrets lying around on the HD Message-ID: <6.0.3.0.2.20040413193734.026d4840@localhost> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 15:50 2004-04-13, you wrote: >On Tue, 13 Apr 2004 14:55:33 +0200, Per Tunedal Casual said: > >> What about virtual memory (SWAP-files) in Windows? > >That may indeed happen and we can't do anything about it. The only >possible solution is to write a device driver and let it allocate >memory for secrets. This memory will then be managed via ioctls. >There is one such driver available but it is not Free Software so we >have not looked deeper into it. > >There are currently rumors that the W32 API VirtualLock does indeed >lock pages against paging - if this is true we will ad this to the >next GnuPG version. > > Werner Fine. I asked all these questions because I suddenly realised that things like e.g. choosing a safe algo is all in vane if someone easily can bypass the encryption. By e.g. finding unencrypted keys or the encrypted file in plaintext on the computer. Per Tunedal -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) - GPGrelay v0.94 iD8DBQFAfCV2ITLMlZFNlMoRAoa7AJ9MznZ1i0YOeNRl+8JgV9JM/uyF4gCgiJu4 nKZPeA7JOLnut+5ejGCE+iM= =F3ET -----END PGP SIGNATURE----- From graham.todd at dsl.pipex.com Tue Apr 13 14:53:10 2004 From: graham.todd at dsl.pipex.com (Graham) Date: Tue Apr 13 19:46:37 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: Message-ID: <200404131353.36917.graham.todd@dsl.pipex.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday 13 Apr 2004 7:43 am, Atom 'Smasher' wrote: > of course, pgp/mime is an *official* standard, while in-line pgp is > an *unofficial* standard. why isn't in-line pgp *officially* > recognized as an email standard? [snipped] It is. OpenPGP is an official standard (RFC 2440) and OpenPGP doesn't include PGP/MIME, which is a function of the email program you are using and NOT of PGP or GnuPG. The problem with PGP/MIME comes from the way it has developed. At first, there were NO standards, and different email programs supported the theory of PGP/MIME differently. The classic case often quoted is that of the Windows program Eudora. PGP/MIME created by Eudora in the early days could not be read by other email programs, and both sender and recipient had to use the same version of Eudora to be able to verify or decypt PGP/MIME. Slowly, a standard developed for PGP/MIME which was RFC2015. I quote from http://www.imc.org/smime-pgpmime.html which states that: "RFC 2015 is a Proposed Standard in the IETF, but it is not expected to move forwards because it relies on RFC 1991, which requires the use of RSA key exchange, and requires the use of IDEA encryption, both of which are encumbered by patents. Both of these patents would likely prevent the protocol from moving forwards as an IETF standard." Many email programs did, however, implement RFC 2015, and there were a number of plugins written for various programs complying with this standard. Versions of PGP had plugins with them that conformed to RFC2015 (for example PGP 6.5.0ckt). When implemented, these plugins changed the email program, not PGP or GnuPG, and if you are still using these plugins you are effectively using the email program which conforms to RFC2015. Now, further work went on and a new standard was proposed, which was RFC 3156, which further details MIME wrapping in OpenPGP. However, if you are not using an email program which conforms to RFC 3156 (and by using the old PGP 6.5.x plugins, you won't be) then there is no guarantee that email programs which conform to RFC3156 will produce PGP/MIME code that those conforming to RFC2015 alone, can verify or decrypt. This is where the problem lies. Inline PGP/GPG messages conform to RFC 2440 and can be verified or decrypted by any PGP/GPG compliant email program (if they support PGP or GPG then they will be compliant), but unless you KNOW which email program and even which version of that email program you recipient will use, its not always possible to send or receive messages by PGP/MIME that will be verified or decrypted properly. As I want to produce messages that can be verified or decrypted without having to check on the email program that might be used, then I will always use inline PGP. The situation is more prevalent in the Windows world as many programs are not standards compliant than in the Linux world, where they tend to be standards complant. Nevertheless, there are still anomalies in Linux, such as Evolution which only generates PGP/MIME which cannot always be decrypted or verified by other email programs. - -- Graham -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Please sign and encrypt for internet privacy iD8DBQFAe+LFjBIOiwu+FpURAquuAKCoWKrgKfJvPFUB8eIJM88ZkO/m7QCgiwSE JPG/DlQTpn7EaVJCK7eVsvI= =wnRu -----END PGP SIGNATURE----- From JPClizbe at comcast.net Tue Apr 13 19:59:33 2004 From: JPClizbe at comcast.net (John Clizbe) Date: Tue Apr 13 19:57:31 2004 Subject: openPGP vs x509 In-Reply-To: <6.0.3.0.2.20040413123533.026d2198@localhost> References: <4073BBEB.6070203@gmx.de> <200404071414.26114@fortytwo.ch> <8765cb7i3d.fsf@vigenere.g10code.de> <6.0.3.0.2.20040413123533.026d2198@localhost> Message-ID: <407C2A85.4080400@comcast.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Per Tunedal Casual wrote: > At 20:34 2004-04-07, Werner Koch wrote: > > >For a casual user it is far easier to create a PGP key than an X.509 > >key/certificate. You can instantly do this with OpenPGP tools (ssh is > >similar) whereas you have to go into lengthy discussions with OpenSSL > >to create even just a self-signed certificate containing a usable > >email attribute. > OT? I have been looking for tools to make self-signed X.509 but haven't > found any. Do you have any hints? Links? > Rational: some people refuse using PGP-encryption/signing. > Per Tunedal Why not just obtain a free cert from Thawte or CA Cert? That would certainly be less work than creating your own CA for a self-signed X.509 cert. - -- John P. Clizbe Inet: JPClizbe(a)comcast DOT nyet Golden Bear Networks PGP/GPG KeyID: 0x608D2A10 "Most men take the straight and narrow. A few take the road less traveled. I chose to cut through the woods." -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.3.5 (Windows 2000 SP4) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org Comment: Annoy John Asscraft -- Use Strong Encyption iD8DBQFAfCqDHQSsSmCNKhARAoJPAKDudOBhOrz3Vh+tteFYqXpdMAwTwwCffnsZ Y8ptgjNj7q6vLnCE6aYZdqI= =f/DQ -----END PGP SIGNATURE----- From SGates at olbh.com Tue Apr 13 20:19:51 2004 From: SGates at olbh.com (Gates, Scott) Date: Tue Apr 13 20:23:17 2004 Subject: secrets lying around on the HD Message-ID: There will be temp files, and pieces of files on your machine left as you create the plaintext document. The editing process is messy. Myself, I do a daily Disk Cleanup, and weekly Defrag followed by a PGP wipe*3 of the disk free space. On a 40GB HDD with a medium sized processor, it takes a while, otherwize I'd do it nightly. I use PGP to wipe files on my HDD that contain company secrets (yes, the company BOUGHT PGP for data security. Other than that, I trust GPG & PGP for TRANSFERING data securely--PGP when I'm able to use it interactively--GPG for batched processes. I know our network administrator and feel reasonably secure with his ability to keep the riff-raff out of the network. So, Don't write GPG/PGP off completely. Perhaps if you could get GPG to work on one of the various KNOPPIX distros? Burn the client s/w to the bootable CD, then boot with that CD with the HDD unplugged. Might work. Might not. Could make for an interesting few days of trying. Then you could encrypt to a virtual drive in memory, send the data to whomever, and the evidence is automagically permanently deleted when you power down the machine. A friend once quoted "Three people can keep a secret, if two of them are dead". I ran away before he had a chance to share any privledged information. I haven't seen the him or the other guy since. -----Original Message----- From: Per Tunedal Casual [mailto:pt@radvis.nu] Sent: Tuesday, April 13, 2004 1:38 PM To: gnupg-users@gnupg.org Subject: Re: secrets lying around on the HD -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 15:50 2004-04-13, you wrote: >On Tue, 13 Apr 2004 14:55:33 +0200, Per Tunedal Casual said: > >> What about virtual memory (SWAP-files) in Windows? > >That may indeed happen and we can't do anything about it. The only >possible solution is to write a device driver and let it allocate >memory for secrets. This memory will then be managed via ioctls. >There is one such driver available but it is not Free Software so we >have not looked deeper into it. > >There are currently rumors that the W32 API VirtualLock does indeed >lock pages against paging - if this is true we will ad this to the >next GnuPG version. > > Werner Fine. I asked all these questions because I suddenly realised that things like e.g. choosing a safe algo is all in vane if someone easily can bypass the encryption. By e.g. finding unencrypted keys or the encrypted file in plaintext on the computer. Per Tunedal -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) - GPGrelay v0.94 iD8DBQFAfCV2ITLMlZFNlMoRAoa7AJ9MznZ1i0YOeNRl+8JgV9JM/uyF4gCgiJu4 nKZPeA7JOLnut+5ejGCE+iM= =F3ET -----END PGP SIGNATURE----- _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users From atom-gpg at suspicious.org Tue Apr 13 21:32:51 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Tue Apr 13 21:30:17 2004 Subject: openPGP vs x509 In-Reply-To: <6.0.3.0.2.20040413123533.026d2198@localhost> References: <4073BBEB.6070203@gmx.de> <200404071414.26114@fortytwo.ch> <8765cb7i3d.fsf@vigenere.g10code.de> <6.0.3.0.2.20040413123533.026d2198@localhost> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Rational: some people refuse using PGP-encryption/signing. ============================= huh?? why would someone prefer to use an x509 based system for email?? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Nothing will benefit human health and increase chances for survival of life on Earth as much as the evolution to a vegetarian diet." -- Albert Einstein -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB8QGcACgkQnCgLvz19QeOKWwCePdS1+ewjO4Op6JU6ZICyNrnq BzgAn0ggkXH/XuUAWw7J4NsFDkS4W97E =wOYA -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Tue Apr 13 21:52:51 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Tue Apr 13 21:50:04 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <20040413114024.GC4400@jabberwocky.com> References: <20040413114024.GC4400@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > of course, pgp/mime is an *official* standard, while in-line pgp is an > > *unofficial* standard. why isn't in-line pgp *officially* recognized as an > > email standard? > > Inline PGP is absolutely an official standard. See RFC-2440. It's > just a different official standard than PGP/MIME (RFC-3156). > > Inline PGP can't handle all the situations that PGP/MIME does, and > it's not nearly as transparent as PGP/MIME, but it is a genuine > official standard. ============================== i didn't realize that RFC 2440 made in-line pgp officially accepted as an *email* standard... something about the body of an email must be plain text... (what RFC is that...?) as an official standard then, does mutt support it? i've heard that it doesn't, which is one of the reasons that i was under the impression that it's not an official standard. it would be nice to sign/encrypt attachments all at once, but i can always sign/encrypt them individually, before attaching them. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "As soon as men decide that all means are permitted to fight an evil, then their good becomes indistinguishable from the evil that they set out to destroy." -- Christopher Dawson, The Judgment of Nations, 1942 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB8RRcACgkQnCgLvz19QeMlTwCfRmpFeJS3qGvTECgZfiHq74TF z/wAn33baY/PeHPEcW5xVnTjNv6e8foj =lQtY -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Tue Apr 13 22:04:03 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Tue Apr 13 22:01:12 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404130954.50900@fortytwo.ch> References: <200404130954.50900@fortytwo.ch> Message-ID: > > * i've *never* had an in-line signature cause problems with someone's > > poorly configured virus filter. > > True. > > The big problem with inline is non-ascii text: when the senders charset > differs from the receivers, the receivers MUA is likely to convert the > message before displaying/storing - breaking the signature (not an > issue when the receiver's MUA directly supports GPG/PGP.) > > I'm not sure, but I can imagine that there may be problems with line > wrapping, too, on conversions to/ftom quoted-printable. ============================== these seem like implementation problems of the MUA, more than problems with the format... if the raw message was piped into gpg, it would work fine, right? the problem seem to come from a receiving MUA tries to be smart about interpreting a message.... it sound slike the same problem of a message that's converted from text to html before verification. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "I believe that banking institutions are more dangerous to our liberties than standing armies." -- Thomas Jefferson From pt at radvis.nu Tue Apr 13 22:05:27 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 13 22:02:51 2004 Subject: openPGP vs x509 In-Reply-To: References: <4073BBEB.6070203@gmx.de> <200404071414.26114@fortytwo.ch> <8765cb7i3d.fsf@vigenere.g10code.de> <6.0.3.0.2.20040413123533.026d2198@localhost> Message-ID: <6.0.3.0.2.20040413215940.026a6f90@localhost> At 21:32 2004-04-13, you wrote: >> Rational: some people refuse using PGP-encryption/signing. >============================= > >huh?? > >why would someone prefer to use an x509 based system for email?? > > > ...atom > Ask them! I guess it has two advantages: 1) easier to use (e.g. built into Outlook and Outlook Express and key distribution is automatic.) 2) some people have easier to understand the hierarchical trust model. And they do trust Thawte :-) - I might be able to convince them to trust me too! Per Tunedal From pt at radvis.nu Tue Apr 13 22:05:52 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 13 22:03:01 2004 Subject: openPGP vs x509 Message-ID: <6.0.3.0.2.20040413220545.026b0368@localhost> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 19:59 2004-04-13, you wrote: >Per Tunedal Casual wrote: > >> At 20:34 2004-04-07, Werner Koch wrote: >> >> >For a casual user it is far easier to create a PGP key than an X.509 >> >key/certificate. You can instantly do this with OpenPGP tools (ssh is >> >similar) whereas you have to go into lengthy discussions with OpenSSL >> >to create even just a self-signed certificate containing a usable >> >email attribute. >> OT? I have been looking for tools to make self-signed X.509 but haven't >> found any. Do you have any hints? Links? >> Rational: some people refuse using PGP-encryption/signing. >> Per Tunedal > >Why not just obtain a free cert from Thawte or CA Cert? > >That would certainly be less work than creating your own CA for a >self-signed X.509 cert. > >- -- >John P. Clizbe Inet: JPClizbe(a)comcast DOT nyet It would be far better to create my own certificates for my company. I would be more reliable than any free Thawte certs. Per Tunedal -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) - GPGrelay v0.94 iD8DBQFAfEgxITLMlZFNlMoRArAIAKCzYvnMsHD7NI3jxxCsjrbZtVY/8gCfQhg3 Qk6Rx3uko5F/goc/jHNnTiA= =I1DU -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Tue Apr 13 22:12:07 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Tue Apr 13 22:09:23 2004 Subject: openPGP vs x509 In-Reply-To: <6.0.3.0.2.20040413215940.026a6f90@localhost> References: <4073BBEB.6070203@gmx.de> <200404071414.26114@fortytwo.ch> <8765cb7i3d.fsf@vigenere.g10code.de> <6.0.3.0.2.20040413123533.026d2198@localhost> <6.0.3.0.2.20040413215940.026a6f90@localhost> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Ask them! I guess it has two advantages: > 1) easier to use (e.g. built into Outlook and Outlook Express and key > distribution is automatic.) =============================== is that another case of M$ trying to break a standard so they can fill their pockets? if trust has to come from a commercial entity.... it just seems icky... paying for trust appeals to me about as much as paying for sex. > 2) some people have easier to understand the hierarchical trust model. And > they do trust Thawte :-) - I might be able to convince them to trust me too! =============================== trust thawte? ha! i've never even met the guy ;) ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarrely inexplicable. There is another theory which states that this has already happened." -- Douglas Adams -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB8SZsACgkQnCgLvz19QeOEdwCeNYDdxRPRjALrs2kE8nwHGPTy KeUAn1az/cuN2ETf1zYGqlyntHJI8KHo =U6Ep -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Tue Apr 13 22:22:59 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Tue Apr 13 22:20:20 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404130954.50900@fortytwo.ch> References: <200404130954.50900@fortytwo.ch> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > The big problem with inline is non-ascii text: when the senders charset > differs from the receivers, the receivers MUA is likely to convert the > message before displaying/storing - breaking the signature (not an > issue when the receiver's MUA directly supports GPG/PGP.) > > I'm not sure, but I can imagine that there may be problems with line > wrapping, too, on conversions to/ftom quoted-printable. ========================== please disregard my previous response to this... i was "thinking out loud" and accidently hit send. anyway, what if gpg supported quoted-printable input? how much of the problem would that solve? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Generally, news reporting and punditry are respectful of the rich and disdainful of the poor." -- Syndicated columnist Norman Solomon -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB8TC8ACgkQnCgLvz19QeN3RACePlsdoEk4VCxZQ4oSoJpclMGf jogAoJD9h/CVFbqycPV6DJyF8mDZQ5T1 =3EO6 -----END PGP SIGNATURE----- From wk at gnupg.org Tue Apr 13 22:57:07 2004 From: wk at gnupg.org (Werner Koch) Date: Tue Apr 13 22:41:30 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: (atom-gpg@suspicious.org's message of "Tue, 13 Apr 2004 15:52:51 -0400 (EDT)") References: <20040413114024.GC4400@jabberwocky.com> Message-ID: <87vfk3r3z0.fsf@vigenere.g10code.de> On Tue, 13 Apr 2004 15:52:51 -0400 (EDT), Atom 'Smasher' said: > it would be nice to sign/encrypt attachments all at once, but i can always > sign/encrypt them individually, before attaching them. An attacker can delete one of those attached and signed documents and insert an old (previously catched) version of that document which carries an old price tag for the now entirely changed specification document in the other attachment ... Oh yes, you can add another attchment with checksums over all the attached documents. Why on earth going into such troubles if a well working and good protocol exists for nearly a decade: rfc-1847 (MOSS) on which S/MIME as well as PGP/MIME are based upon. BTW, the changes in rfc-3156 compared to rfc-2015 are minimal and actually a clarification of some points in 2015. Rfc-2015 is nearly 8 years old - that's the Windows-95 area! So please don't say, there was not enough time to implement such a simple thing. Shalom-Salam, Werner From atom-gpg at suspicious.org Tue Apr 13 22:53:02 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Tue Apr 13 22:50:16 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <87vfk3r3z0.fsf@vigenere.g10code.de> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > it would be nice to sign/encrypt attachments all at once, but i can always > > sign/encrypt them individually, before attaching them. > > An attacker can delete one of those attached and signed documents and > insert an old (previously catched) version of that document which > carries an old price tag for the now entirely changed specification > document in the other attachment ... Oh yes, you can add another > attchment with checksums over all the attached documents. =================== usually, if there's a need to sign (not encrypt) an attachment, i can (while limited by pine) include a hash of the attachment(s) in the body of the email and sign that. you point out a need to do that also for encrypted attachments. > Why on earth going into such troubles if a well working and good > protocol exists for nearly a decade: rfc-1847 (MOSS) on which S/MIME > as well as PGP/MIME are based upon. BTW, the changes in rfc-3156 > compared to rfc-2015 are minimal and actually a clarification of some > points in 2015. Rfc-2015 is nearly 8 years old - that's the > Windows-95 area! So please don't say, there was not enough time to > implement such a simple thing. =================== now to convince the pine team at UW to implement such a simple thing! ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "We must learn to live together as brothers or perish together as fools." -- Martin Luther King, Jr. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB8UzIACgkQnCgLvz19QeNYpgCghd8R2nusiAeSrA5sb0Tqt7PO t5sAmQHl57Krp1O3ybG60bdqqilmsoUQ =1+YW -----END PGP SIGNATURE----- From dlc at sevenroot.org Tue Apr 13 22:58:16 2004 From: dlc at sevenroot.org (darren chamberlain) Date: Tue Apr 13 23:11:16 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> Message-ID: * Atom 'Smasher' [2004/04/13 15:52]: > as an official standard then, does mutt support it? i've heard that it > doesn't, which is one of the reasons that i was under the impression > that it's not an official standard. Yes, it does. > it would be nice to sign/encrypt attachments all at once, but i can > always sign/encrypt them individually, before attaching them. With PGP/MIME you can sign the whole message, instead of individual parts. Verifying an entire message is different than individually verifying each individual component. (darren) -- How you look depends on where you go. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : /pipermail/attachments/20040413/aad241ce/attachment.bin From dlc at sevenroot.org Tue Apr 13 22:56:03 2004 From: dlc at sevenroot.org (darren chamberlain) Date: Tue Apr 13 23:11:40 2004 Subject: openPGP vs x509 In-Reply-To: References: <4073BBEB.6070203@gmx.de> <200404071414.26114@fortytwo.ch> <8765cb7i3d.fsf@vigenere.g10code.de> <6.0.3.0.2.20040413123533.026d2198@localhost> <6.0.3.0.2.20040413215940.026a6f90@localhost> Message-ID: <92180a90-262d-49ba-9496-f70eb0dd4ed2@tumbleweed.boston.com> * Atom 'Smasher' [2004/04/13 16:12]: > > Ask them! I guess it has two advantages: > > 1) easier to use (e.g. built into Outlook and Outlook Express and key > > distribution is automatic.) > =============================== > > is that another case of M$ trying to break a standard so they can fill > their pockets? if trust has to come from a commercial entity.... No, there are many mail clients that can natively use x509 certs, like Lotus Notes and Netscape Messenger. The standards exist and are implemented independently of Microsoft. (darren) -- It is impossible to travel faster than the speed of light, and certainly not desirable, as one's hat keeps blowing off. -- Woody Allen -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : /pipermail/attachments/20040413/49507dcb/attachment.bin From dshaw at jabberwocky.com Tue Apr 13 23:24:27 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Tue Apr 13 23:21:41 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> Message-ID: <20040413212426.GA15752@jabberwocky.com> On Tue, Apr 13, 2004 at 03:52:51PM -0400, Atom 'Smasher' wrote: > > > of course, pgp/mime is an *official* standard, while in-line pgp is an > > > *unofficial* standard. why isn't in-line pgp *officially* recognized as an > > > email standard? > > > > Inline PGP is absolutely an official standard. See RFC-2440. It's > > just a different official standard than PGP/MIME (RFC-3156). > > > > Inline PGP can't handle all the situations that PGP/MIME does, and > > it's not nearly as transparent as PGP/MIME, but it is a genuine > > official standard. > ============================== > > i didn't realize that RFC 2440 made in-line pgp officially accepted as an > *email* standard... What is a standard? A standard is a bunch of people agreeing to do a particular task in a particular way. RFC-2440 makes inline OpenPGP standards track. If you choose to use it for email, that's great, but the standard says nothing one way or the other. OpenPGP is not an email standard any more than TCP is an email standard, even though it is commonly used in/for email. > as an official standard then, does mutt support it? i've heard that it > doesn't, which is one of the reasons that i was under the impression that > it's not an official standard. If mutt supports it (it does), that's great, but it has nothing to do with whether it is "official" or a "standard". David From dshaw at jabberwocky.com Tue Apr 13 23:35:34 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Tue Apr 13 23:32:48 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <200404130954.50900@fortytwo.ch> Message-ID: <20040413213534.GB15752@jabberwocky.com> On Tue, Apr 13, 2004 at 04:22:59PM -0400, Atom 'Smasher' wrote: > > The big problem with inline is non-ascii text: when the senders charset > > differs from the receivers, the receivers MUA is likely to convert the > > message before displaying/storing - breaking the signature (not an > > issue when the receiver's MUA directly supports GPG/PGP.) > > > > I'm not sure, but I can imagine that there may be problems with line > > wrapping, too, on conversions to/ftom quoted-printable. > ========================== > > please disregard my previous response to this... i was "thinking out loud" > and accidently hit send. > > anyway, what if gpg supported quoted-printable input? how much of the > problem would that solve? For every one problem that it would solve, it would create at least one hundred more. The answer to mail problems is to fix mail programs. David From graham.todd at dsl.pipex.com Tue Apr 13 23:46:23 2004 From: graham.todd at dsl.pipex.com (Graham) Date: Tue Apr 13 23:43:26 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <200404130954.50900@fortytwo.ch> Message-ID: <200404132246.23133.graham.todd@dsl.pipex.com> On Tuesday 13 Apr 2004 9:22 pm, Atom 'Smasher' wrote: [snipped] > anyway, what if gpg supported quoted-printable input? how much of the > problem would that solve? GPG is not required to support quoted-printable (as in RFC2045), nor PGP/MIME (as in RFC3156): its a function of the MTA to support these. Some MTAs do support quoted-printable, some don't, just as some MTAs support PGP/MIME and some don't. Remember you can always use GPG externally from your MTA to sign or encrypt with GPG and paste the result in, if you want to use inline coding. Its just convenience to use your MTA. However, if you want to use PGP/MIME or quoted printable you MUST have an MTA that supports these and be sending to a recipient which uses an MTA that supports these to the same standards. For more details on these and other RFCs look at: http://www.faqs.org/rfcs/rfcsearch.html -- Graham From b.buerger at penguin.de Wed Apr 14 00:26:26 2004 From: b.buerger at penguin.de (Bjoern Buerger) Date: Wed Apr 14 00:12:12 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> Message-ID: <20040413222626.GC1422@susie.penguin.de> [ 13 Apr 2004 - Atom 'Smasher' ]: > > Windows-95 area! So please don't say, there was not enough time to > > implement such a simple thing. > > now to convince the pine team at UW to implement such a simple thing! Since _you_ want to use pine, it would be up to _you_ to do that. If they don't plan to implement it themselves, best thing would be to send them patches. I wouldn't want to miss pgp/mime any second. Just my 2? Bj?rn -- OpenPGP Keyserver +-----------+ --------------------------------- |\ O---m /| http://pgpkeys.tu-bs.de |/`-------'\| http://sks.keyserver.penguin.de +-----------+ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 253 bytes Desc: not available Url : /pipermail/attachments/20040414/442da3fe/attachment.bin From gnichols at tpg.com.au Wed Apr 14 08:15:22 2004 From: gnichols at tpg.com.au (Graeme Nichols) Date: Wed Apr 14 08:12:33 2004 Subject: Problems importing public key Message-ID: <1081923324.1671.17.camel@localhost.localdomain> Hello Folks, I was sent a public key by a fellow so that I could encrypt a file to him. The name of the file, for what it is worth is: 0xF94BBB03.asc which happens to be the DSA key ID of his key used to sign his emails. It imports OK into my gnupg V1.2.3 but when I click on the lock icon in Evolution (so I can test the public key) I get the following error: gpg: armor header: Version: GnuPG v1.2.4 (GNU/Linux) gpg: Signature made Sat 03 Apr 2004 09:54:31 EST using DSA key ID F94BBB03 gpg: BAD signature from "Benoit Gr?goire (Serveur t?l?matique des ?tudiants de Polytechnique) " gpg: textmode signature, digest algorithm SHA1 What have I done wrong? I have another public key that I imported from a keyserver that works OK if I click the lock icon on Evolution yet if I fire up seahorse and click on the Key Manager icon to list the keys I have under the 'trust' column both the public keys I have imported show 'ERROR'. Mine naturally shows 'ultimate' I really need to find out if I have done something wrong so I can rectify the problem as Benoit is waiting for the file he needs me to send him and I am not going to send it if something is wrong as the file has some very private data in it and I don't want it to fall into the wrong hands. -- Kind regards, Graeme Nichols ---------------------------------------------------------------------- AMAZING BUT TRUE ... There is so much sand in Northern Africa that if it were spread out it would completely cover the Sahara Desert. ---------------------------------------------------------------------- - IMPORTANT. - - The contents of this email and any attachments, which may be con- - - fidential, are sent for the personal attention of the addressee/s - - only. If you receive this email and are not the intended addressee - - please inform the sender and delete this email immediately. Use, - - copying, disclosure or forwarding of the contents of this email - - and/or any attachment/s is not authourised. - ---------------------------------------------------------------------- -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : /pipermail/attachments/20040414/3d8daf16/attachment.bin From anonymous at remailer.metacolo.com Wed Apr 14 08:18:55 2004 From: anonymous at remailer.metacolo.com (Anonymous Sender) Date: Wed Apr 14 08:16:04 2004 Subject: openPGP vs x509 Message-ID: Werner Koch: > TLS provides such a mechanism but sane users don't hand their money > over to Verisign for a user certificate. I don't know whether > browsers support this at all. 'curl' and 'libcurl' do support TLS client authentication: $ curl --help|grep cert -E/--cert Specifies your certificate file and password (HTTPS) --cert-type Specifies certificate file type (DER/PEM/ENG) (HTTPS) --cacert CA certifciate to verify peer against (SSL) -k/--insecure Allow curl to connect to SSL sites without certs (H) Stunnel, which also supports client authentication, can be used with browsers that do not support it natively... From pt at radvis.nu Wed Apr 14 08:32:05 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Wed Apr 14 08:34:56 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> Message-ID: <6.0.3.0.2.20040414082340.026d4b60@localhost> At 21:52 2004-04-13, you wrote: > >it would be nice to sign/encrypt attachments all at once, but i can always >sign/encrypt them individually, before attaching them. > > > ...atom You can set GPGrelay to sign/encrypt the attachments inlined, but it is only recommended if the receiver cannot handle PGP-MIME. PGP-MIME is far better but unfortunately PGP-MIME signed mail is not transparent to non-pgp users. They will be confused by the empty mail with the two strange attachments (though readable in a texteditor). Thus I cannot use PGP-MIME widely. (S/MIME has an advantage as such signatures doesn't confuse users: the mail appears as usual in e.g. Outlook Express, only a fancy attachment is added.) Per Tunedal From pt at radvis.nu Wed Apr 14 08:37:54 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Wed Apr 14 08:35:09 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404130954.50900@fortytwo.ch> References: <200404130954.50900@fortytwo.ch> Message-ID: <6.0.3.0.2.20040414083404.0268b970@localhost> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 09:54 2004-04-13, Adrian 'Dagurashibanipal' von Bidder wrote: > >The big problem with inline is non-ascii text: when the senders charset >differs from the receivers, the receivers MUA is likely to convert the >message before displaying/storing - breaking the signature (not an >issue when the receiver's MUA directly supports GPG/PGP.) > >I'm not sure, but I can imagine that there may be problems with line >wrapping, too, on conversions to/ftom quoted-printable. > I have noticed many non-valid inlined signatures but I have never ran into any problems with inlined encrypted (and signed) mail. Why? What's the difference? Does the encryption somehow protect the original mail (so the signature always will verify)? Can this knowledge be used to somehow improve the inlined signatures? Per Tunedal -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) - GPGrelay v0.94 iD8DBQFAfNxUITLMlZFNlMoRAqJBAJ4uzp5xOGm9sKyQIsGbLNDRul+KLACfXtx9 8KtLnOBdntOf7sGwvEiA7zA= =OHLL -----END PGP SIGNATURE----- From wk at gnupg.org Wed Apr 14 09:24:46 2004 From: wk at gnupg.org (Werner Koch) Date: Wed Apr 14 09:06:29 2004 Subject: openPGP vs x509 In-Reply-To: (Anonymous Sender's message of "Wed, 14 Apr 2004 06:18:55 +0000 (UTC)") References: Message-ID: <87llkzqawx.fsf@vigenere.g10code.de> On Wed, 14 Apr 2004 06:18:55 +0000 (UTC), Anonymous Sender said: > 'curl' and 'libcurl' do support TLS client authentication: Sure, but I was not talking about hackers but average computer users. Salam-Shalom, Werner From avbidder at fortytwo.ch Wed Apr 14 09:38:09 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 14 09:35:20 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404132246.23133.graham.todd@dsl.pipex.com> References: <200404132246.23133.graham.todd@dsl.pipex.com> Message-ID: <200404140938.10167@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday 13 April 2004 23.46, Graham wrote: > On Tuesday 13 Apr 2004 9:22 pm, Atom 'Smasher' wrote: > > [snipped] > > > anyway, what if gpg supported quoted-printable input? how much of > > the problem would that solve? > > GPG is not required to support quoted-printable (as in RFC2045), nor > PGP/MIME (as in RFC3156): its a function of the MTA to support these. You mean MUA. The only requirement on a MTA by PGP is that it should be able to transport byte-for-byte copies of ascii messages. Unfortunately, this is an area where many MTA have failed in the past (and some still do), and especially when we leave ascii and enter 8-bit land this is sometimes still a problem. greetings - -- vbi - -- The content of this message may or may not reflect the opinion of me, my employer, my girlfriend, my cat or anybody else, regardless of the fact whether such an employer, girlfriend, cat, or anybody else exists. I (or my employer, girlfriend, cat or whoever) disclaim any legal obligations resulting from the above message. You, as the reader of this message, may or may not have the permission to redistribute this message as a whole or in parts, verbatim or in modified form, or to distribute any message at all. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkB86mFgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6IIgAnjUXWr32k4udk3pr8nwFB4nn 3EGxAJ4hntkpUzSf03s7cqfISNd2GaPn9A== =ZP2p -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Wed Apr 14 09:45:12 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 14 09:42:27 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <6.0.3.0.2.20040414083404.0268b970@localhost> References: <200404130954.50900@fortytwo.ch> <6.0.3.0.2.20040414083404.0268b970@localhost> Message-ID: <200404140945.17637@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 People, why do you all insist on cc:ing me? Plese do not. On Wednesday 14 April 2004 08.37, Per Tunedal Casual wrote: > I have noticed many non-valid inlined signatures but I have never ran > into any problems with inlined encrypted (and signed) mail. Why? > What's the difference? Does the encryption somehow protect the > original mail (so the signature always will verify)? Can this > knowledge be used to somehow improve the inlined signatures? > Per Tunedal simple: encrypted mail is encoded in base64, all changes in whitespace are irrelevant (and there are very few MTA/MUA/MDA problems that really corrupt non-whitespace ascii characters in email. And those usually don't survive long...) with signed mail, whitespace becomes significant. There are some MTA and many MDA and MUA-related issues regarding whitespace. Also, not all gpg/pgp versions behave the same regarding whitespace at end of line. Additionally: encoding issues. Encrypted mail again is protected by the base64 encoding and by having the signature inside the encrypted part. So gpg will always be able to look at the signed data directly, because no other code in the mailer can interpret the base64 encoded data. Signed mail is open to be modified by the MUA because it is just text, and contains the 8bit-characters directly (raw or in qp form). So many MUA/MDA and even some MTAs play with the encoding... PGP/MIME can go around the encoding issues: it is properly standardized how a PGP/MIME message must be generated, so all implementations should know how to verify it (not all get it right - evo had some nasty bugs there.) The whitespace issues are *in principle* also possible to work around for inline PGP: strip whitespace at end of line, and you're mostly safe. greetings - -- vbi - -- Today is Prickle-Prickle, the 31st day of Discord in the YOLD 3170 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkB87A1gGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l65N8An3Px8mWaQvzo/hWcEh5wlueu n5EJAKCfI0xVekBgzwfP+xInBljBN/02Bg== =/cHQ -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Wed Apr 14 09:49:47 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 14 09:46:58 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> Message-ID: <200404140949.49143@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday 13 April 2004 21.52, Atom 'Smasher' wrote: > i didn't realize that RFC 2440 made in-line pgp officially accepted > as an *email* standard... something about the body of an email must > be plain text... (what RFC is that...?) If the output of gpg --clearsign < 'some text file' is not plain text, I don't know what is. Email is a tool to transport plain text bodies around. gpg --clearsign produces plain text, given plain text input. The output of gpg - --clearsign follows rfc2440. That's all there is. > it would be nice to sign/encrypt attachments all at once, but i can > always sign/encrypt them individually, before attaching them. Having to send attachments to Outlook-Expressivistically challenged people, I just include the sha1 of the attached files at the bottom of the clearsigned body. This protects the attachments well enough, and the recipient doesn't have to know anything if he doesn't want to verify the attachments. Of course, none of the OE users I know ill ever have the faintest idea what I'm doing - but it's a matter of principle... greets - -- vbi (Please don't cc: me) - -- "The chain which can be yanked is not the eternal chain." -- G. Fitch -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkB87RtgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6Y4YAoIUvTKN6p3+jyFQJzLSZ/uoM f61cAJ9I8wffz3FUZ5GTKbBTvqPpulXAzQ== =vzWW -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Wed Apr 14 09:49:44 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 09:47:07 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <6.0.3.0.2.20040414083404.0268b970@localhost> References: <200404130954.50900@fortytwo.ch> <6.0.3.0.2.20040414083404.0268b970@localhost> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > >I'm not sure, but I can imagine that there may be problems with line > >wrapping, too, on conversions to/ftom quoted-printable. > > > I have noticed many non-valid inlined signatures but I have never ran into > any problems with inlined encrypted (and signed) mail. Why? What's the > difference? Does the encryption somehow protect the original mail (so the > signature always will verify)? Can this knowledge be used to somehow > improve the inlined signatures? ========================================== in a clearsigned email, lines can be longer than they should be... but an inline encrypted email would have lines of interest no longer then 64 characters (excluding CR/NL) within the body of the email. during decryption, the signature is checked as the message is being decrypted, before the plain-text is output from gpg... so if an MUA wraps text and tries to verify that (incorrectly) wrapped text, the signature will fail.... but if an MUA decrypts a message, the signature is verified before the text has a chance to wrap in the MUA. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Freedom without opportunity is a devil's gift, and the refusal to provide such opportunities is criminal" -- Market Democracy in a Neoliberal Order: Doctrines and Reality, Noam Chomsky -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB87R0ACgkQnCgLvz19QeObiACeOs6l6TJIsZ6KbRonhaBTrGSI BpoAoI/RhRI5zshM0vM31Fd5HS6Ya6zU =5yZp -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Wed Apr 14 09:56:55 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 14 09:54:04 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <6.0.3.0.2.20040414082340.026d4b60@localhost> References: <6.0.3.0.2.20040414082340.026d4b60@localhost> Message-ID: <200404140956.55981@fortytwo.ch> (don't cc: me please) On Wednesday 14 April 2004 08.32, Per Tunedal Casual wrote: > PGP-MIME is far better but unfortunately PGP-MIME signed mail is not > transparent to non-pgp users. They will be confused by the empty mail > with the two strange attachments (though readable in a texteditor). > Thus I cannot use PGP-MIME widely. Name one mailer except MSOE which also shows this behaviour. All other mailers I've come across properly implement MIME and can sensibly display MIME multipart messages with an unknown subtype. I have by now just given up on MSOE users: they want to live that way, so be it (There are very few exceptions to this rule - job applications is one example where it is vital for me that my messages are read :-/. The confusion for the user is exactly the same: it is a strange attachment in one case, and it is '----- BEGIN PGP SIGNED MESSAGE -----' at the top and a strange blurb at the bottom in the other case. (Arrgh! And now I can't find the magic checkbox to enable PGP/MIME as default in kmail. It's the default at home - but inline is the default on this machine. Anybody?) cheers -- vbi -- Some people pray for more than they are willing to work for. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040414/6c6d50d3/attachment.bin From atom-gpg at suspicious.org Wed Apr 14 09:58:32 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 09:55:47 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404140945.17637@fortytwo.ch> References: <200404130954.50900@fortytwo.ch> <6.0.3.0.2.20040414083404.0268b970@localhost> <200404140945.17637@fortytwo.ch> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > The whitespace issues are *in principle* also possible to work > around for inline PGP: strip whitespace at end of line, and you're > mostly safe. ========================= you can add (or delete) as much white-space as you want at the ends of lines, and that should NOT affect clearsigned signature creation or verification. you can test this by running: $ date | gpg --clearsign > test-file then edit "test-file" to add spaces at the end of lines within the signed data.... then: $ gpg --verify < test-file should still say the signature is good. RFC 2440, 7.1. Dash-Escaped Text: "Also, any trailing whitespace (spaces, and tabs, 0x09) at the end of any line is ignored when the cleartext signature is calculated." ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "I am a former Los Angeles Police narcotics detective. I worked South Central Los Angeles. And I can tell you, Director Deutch, emphatically and without equivocation, that the Agency has dealt drugs in this country for a long time." -- Mike Ruppert to DCI John Deutch, 15 Nov 1996 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB87ywACgkQnCgLvz19QeNbNgCeL5+rVhhaj6kysJrDmj8HExaD qmoAoI5OL0YjZp30AQYdS3nsCPDnNegN =iVtK -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Wed Apr 14 10:12:43 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 14 10:09:54 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <200404140945.17637@fortytwo.ch> Message-ID: <200404141012.44793@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'd be grateful if you respected proper attribution in email. On Wednesday 14 April 2004 09.58, Atom 'Smasher' wrote: > Dagurashibanipal: > > The whitespace issues are *in principle* also possible to work > > around for inline PGP: strip whitespace at end of line, and you're > > mostly safe. > > ========================= > > you can add (or delete) as much white-space as you want at the ends > of lines, and that should NOT affect clearsigned signature creation > or verification. Please read what I wrote. | Also, not all | gpg/pgp versions behave the same regarding whitespace at end of line. Read the archives of the ietf-openpgp at imc.org mailing list, there was much discussion about whitespace handling. cheers - -- vbi -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkB88ntgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l650oAn1z5V6H9/NVkTQiOxoePP0yn JSemAJ9sMhvYKmX6uev/NioIFLqggbroBg== =UAk+ -----END PGP SIGNATURE----- From pt at radvis.nu Wed Apr 14 10:31:30 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Wed Apr 14 10:28:41 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404140956.55981@fortytwo.ch> References: <6.0.3.0.2.20040414082340.026d4b60@localhost> <200404140956.55981@fortytwo.ch> Message-ID: <6.0.3.0.2.20040414102508.026d95e0@localhost> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 09:56 2004-04-14, you wrote: > >On Wednesday 14 April 2004 08.32, Per Tunedal Casual wrote: > >> PGP-MIME is far better but unfortunately PGP-MIME signed mail is not >> transparent to non-pgp users. They will be confused by the empty mail >> with the two strange attachments (though readable in a texteditor). >> Thus I cannot use PGP-MIME widely. > >Name one mailer except MSOE which also shows this behaviour. All other >mailers I've come across properly implement MIME and can sensibly >display MIME multipart messages with an unknown subtype. I have by now >just given up on MSOE users: they want to live that way, so be it >(There are very few exceptions to this rule - job applications is one >example where it is vital for me that my messages are read :-/. > >The confusion for the user is exactly the same: it is a strange >attachment in one case, and it is '----- BEGIN PGP SIGNED MESSAGE >-----' at the top and a strange blurb at the bottom in the other case. > NO, PGP-MIME looks really strange in Outlook Express. You'd better try it! I cannot give up the OE-users as they are the mainpart of my customers. Thus I have to stick to inlined pgp-signing (doesn't always verify OK) or switch to S-MIME (rather not!). Per Tunedal -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) - GPGrelay v0.94 iD8DBQFAfPb+ITLMlZFNlMoRAm2vAJ0Z/Vs/6X/WMeE8xrkHPfmCHmMHUACfd7+g 4Y50jPm58r7hcOyGoThk+zU= =nyeV -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Wed Apr 14 10:54:51 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 14 10:52:07 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <6.0.3.0.2.20040414102508.026d95e0@localhost> References: <200404140956.55981@fortytwo.ch> <6.0.3.0.2.20040414102508.026d95e0@localhost> Message-ID: <200404141054.56093@fortytwo.ch> On Wednesday 14 April 2004 10.31, Per Tunedal Casual wrote: > At 09:56 2004-04-14, you wrote: > >On Wednesday 14 April 2004 08.32, Per Tunedal Casual wrote: > >> PGP-MIME is far better but unfortunately PGP-MIME signed mail is > >> not transparent to non-pgp users. They will be confused by the > >> empty mail with the two strange attachments (though readable in a > >> texteditor). Thus I cannot use PGP-MIME widely. > > > >Name one mailer except MSOE which also shows this behaviour. All > > other mailers I've come across Please, please, please, pretty please, do NOT cc: me on replies. > NO, PGP-MIME looks really strange in Outlook Express. You'd better > try it! Also, please read what I wrote. I challenge you to name another mailer which *also* shows that broken behaviour. I know MSOE's behaviour. But your claim | PGP-MIME signed mail is not transparent to non-pgp users. They will be | confused by the empty mail with the two strange attachments makes it sound like a general problem. It is *not* a general problem, it is a problem with one specific mail application from one vendor widely known for the outstanding quality of its software products. From hearsay, it's even only MSOE and not Outlook. > I cannot give up the OE-users as they are the mainpart of my > customers. Thus I have to stick to inlined pgp-signing (doesn't > always verify OK) or switch to S-MIME (rather not!). I can fully understand that situation. I tend to use PGP/MIME wherever possible, especially on mailing lists. As I've said in my previous email, I do make exceptions for the few MSOE users I have to communicate with. (and I also do not play the begin game with MSOE users.) To my enjoyment, even many non-tech-savvy computer users and die-hard Windows victims I know are slowly switching from MSOE to Mozilla (in most cases without me urging me to), so that problem should disappear within a few years. greetings -- vbi -- Available for key signing in Z?rich and Basel, Switzerland (what's this? Look at http://fortytwo.ch/gpg/intro) -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040414/8831da77/attachment.bin From avbidder at fortytwo.ch Wed Apr 14 13:22:02 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 14 13:19:28 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <883640782.20040414130643@zedat.fu-berlin.de> References: <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643@zedat.fu-berlin.de> Message-ID: <200404141322.17445@fortytwo.ch> Skipped content of type multipart/mixed-------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040414/b7e7eaa0/attachment.bin From svwright+lists at amtp.liv.ac.uk Wed Apr 14 13:42:14 2004 From: svwright+lists at amtp.liv.ac.uk (Stewart V. Wright) Date: Wed Apr 14 13:39:23 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404141322.17445@fortytwo.ch> References: <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643@zedat.fu-berlin.de> <200404141322.17445@fortytwo.ch> Message-ID: <20040414114214.GA5860@amtp.liv.ac.uk> G'day Adrian, * Adrian 'Dagurashibanipal' von Bidder [040414 12:33]: > On Wednesday 14 April 2004 13.06, Johannes Posel wrote: > > > Well, one thing that tends to be forgotten is that PGP/MIME > > absolutely reliably breaks list footers ;-) > > Not on this list (and any other recent mailman installation - dunno > about other mlms.) It's not even that simple. I'm on both this list (obviously) and the GnuCash lists. What surprises me is that both use Mailman v2.1.4, I sign my messages the same way on both lists and yet my messages to the GnuCash list don't seem to be verified in an acceptable way (to Mutt). The _signed_ part of the message is verified OK, it is just that the signed message plus footer returns a failed check. Where as on this list everything works fine! > Or is this a display problem with some mailers? kmail just displays the > mailman additional part like it is supposed to be. Mutt shows the whole message so there is no problem there, it is something screwy in the way Mailman is set up (I guess...). Cheers, S. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 274 bytes Desc: Digital signature Url : /pipermail/attachments/20040414/b93fc22c/attachment.bin From jam at athene.jamux.com Wed Apr 14 16:16:40 2004 From: jam at athene.jamux.com (John A. Martin) Date: Wed Apr 14 16:14:00 2004 Subject: pgp/mime vs in-line pgp References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643__20007.8024332926$1081941345@zedat.fu-berlin.de> Message-ID: <87wu4iod9z.fsf@athene.jamux.com> >>>>> "Johannes" == Johannes Posel >>>>> "Re: pgp/mime vs in-line pgp" >>>>> Wed, 14 Apr 2004 13:06:43 +0200 Johannes> Well, one thing that tends to be forgotten is that Johannes> PGP/MIME absolutely reliably breaks list footers ;-) Is that true for this here mail on this list? jam -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 154 bytes Desc: not available Url : /pipermail/attachments/20040414/0f12aa63/attachment.bin From ipalette at hotmail.com Tue Apr 13 21:23:50 2004 From: ipalette at hotmail.com (Kevin McNally) Date: Wed Apr 14 16:25:54 2004 Subject: Windows 2003 installation Message-ID: I am really new to GNUPG and I had a encryption email form working correctly on Windows 2K server. I put it on the windows 2003 server and now nothing seems to work. I have looked all over for help on this with no luck, has anyone successfully gotten GNUPG working on windows 2003 or maybe someone saw a help file somewhere. Any help would be greatly appreciated. Thanks Kevin _________________________________________________________________ Tax headache? MSN Money provides relief with tax tips, tools, IRS forms and more! http://moneycentral.msn.com/tax/workshop/welcome.asp From rabbi at abditum.com Wed Apr 14 04:41:20 2004 From: rabbi at abditum.com (Len Sassaman) Date: Wed Apr 14 16:26:01 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <20040413222626.GC1422@susie.penguin.de> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> Message-ID: On Wed, 14 Apr 2004, Bjoern Buerger wrote: > > now to convince the pine team at UW to implement such a simple thing! > > Since _you_ want to use pine, it would be up to > _you_ to do that. If they don't plan to implement > it themselves, best thing would be to send them > patches. Unfortunately, it's not that easy. Pine is under a "non-free" license, and the Pine authors have repeatedly rejected such patches. They have no interest in OpenPGP support, period. Maintaining a patch-set in parallel with pine would be a painful task. From dlc at sevenroot.org Wed Apr 14 16:51:51 2004 From: dlc at sevenroot.org (darren chamberlain) Date: Wed Apr 14 16:49:34 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> Message-ID: * Len Sassaman [2004/04/13 19:41]: > On Wed, 14 Apr 2004, Bjoern Buerger wrote: > > > > now to convince the pine team at UW to implement such a simple > > > thing! > > > > Since _you_ want to use pine, it would be up to _you_ to do that. If > > they don't plan to implement it themselves, best thing would be to > > send them patches. > > Unfortunately, it's not that easy. Pine is under a "non-free" license, > and the Pine authors have repeatedly rejected such patches. They have > no interest in OpenPGP support, period. Sounds like a pretty good reason for a security- and privacy-conscious user to switch to something else... (darren) -- What a strange illusion it is to suppose that beauty is goodness. -- Leo Tolstoy -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : /pipermail/attachments/20040414/e1224d5e/attachment.bin From dlc at sevenroot.org Wed Apr 14 16:56:47 2004 From: dlc at sevenroot.org (darren chamberlain) Date: Wed Apr 14 16:54:31 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <87wu4iod9z.fsf@athene.jamux.com> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643__20007.8024332926$1081941345@zedat.fu-berlin.de> <87wu4iod9z.fsf@athene.jamux.com> Message-ID: <68c288a3-37d0-4d26-b3ae-88d68af0cf63@tumbleweed.boston.com> * John A. Martin [2004/04/14 10:16]: > >>>>> "Johannes" == Johannes Posel > >>>>> "Re: pgp/mime vs in-line pgp" > >>>>> Wed, 14 Apr 2004 13:06:43 +0200 > > Johannes> Well, one thing that tends to be forgotten is that > Johannes> PGP/MIME absolutely reliably breaks list footers ;-) > > Is that true for this here mail on this list? No. (darren) -- In the fight between you and the world, back the world. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : /pipermail/attachments/20040414/b14b331c/attachment-0001.bin From Freedom_Lover at pobox.com Wed Apr 14 17:23:10 2004 From: Freedom_Lover at pobox.com (Todd) Date: Wed Apr 14 17:20:56 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <20040414114214.GA5860@amtp.liv.ac.uk> References: <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643@zedat.fu-berlin.de> <200404141322.17445@fortytwo.ch> <20040414114214.GA5860@amtp.liv.ac.uk> Message-ID: <20040414152310.GH5727@psilocybe.teonanacatl.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Stewart V. Wright wrote: > It's not even that simple. I'm on both this list (obviously) and > the GnuCash lists. What surprises me is that both use Mailman > v2.1.4, I sign my messages the same way on both lists and yet my > messages to the GnuCash list don't seem to be verified in an > acceptable way (to Mutt). The _signed_ part of the message is > verified OK, it is just that the signed message plus footer returns > a failed check. Where as on this list everything works fine! Perhaps you could give mutt 1.5.6 a try? There was a change since 1.5.4 in the output when verifying signatures with a multipart/mixed content-type. I don't know why you get different results from two different lists running the same version of Mailman without comparing the headers and mime-parts from the two lists. If mutt 1.5.6 doesn't change the results, I'll be more curious about it. I'd like to see mutt work as well as possibly in these case as I'm on a lot of mailman lists as well. - -- Todd OpenPGP -> KeyID: 0xD654075A | URL: www.pobox.com/~tmz/pgp ====================================================================== Gradualism is perpetuity in practice. -- William Lloyd Garrison -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: When crypto is outlawed bayl bhgynjf jvyy unir cevinpl. iD8DBQFAfVdduv+09NZUB1oRAmR7AJ9QePPJwhRB2AVw6vr1Z/agTRVsBQCfZWMp tPPEyP9F8cmXOuaGuKbIgj8= =1+Yk -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Wed Apr 14 18:41:16 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 18:38:33 2004 Subject: pine vs mutt - was: Re: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 14 Apr 2004, darren chamberlain wrote: > * Len Sassaman [2004/04/13 19:41]: > > On Wed, 14 Apr 2004, Bjoern Buerger wrote: > > > > > > now to convince the pine team at UW to implement such a simple > > > > thing! > > > > > > Since _you_ want to use pine, it would be up to _you_ to do that. If > > > they don't plan to implement it themselves, best thing would be to > > > send them patches. > > > > Unfortunately, it's not that easy. Pine is under a "non-free" license, > > and the Pine authors have repeatedly rejected such patches. They have > > no interest in OpenPGP support, period. ====================== exactly. i've been using pine for about 10 years... it's not a case of "this is what i'm used to", it's really a case of "i've pushed this to the limits of what it can do and nothing else can do it." i emailed them to report my observations about the way pine handles ssl/tls imap connections, and the response made it clear that they just don't have a clue about security. not even a little bit. > Sounds like a pretty good reason for a security- and privacy-conscious > user to switch to something else... ====================== i'd like to switch to a free (as in speech) MUA for ideological reasons. and of course, everyone knows that mutt is ~better~. i've tried mutt several times, and i'm willing to learn a new interface, but it just can't do what i need. i've asked in the mutt mailing list & usenet, but i'm told the problem (with colors) is in ncurses/slang, not mutt. either way, mutt can't do what i need, and in the meantime pine can, and for me the color support is necessary. i keep checking back with mutt every few years, and when they get the color support as good as pine has, i'm confident that i can either configure everything else the way i need it or write my own hooks. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Like many big companies, Microsoft wins by dominating distribution channels, not by having better products. Having a technical edge over competitors is not critical to their business." -- Paul Graham -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB9abEACgkQnCgLvz19QeNxHwCfeh02JB4VzvzUCKu9QTIzZSSb NSIAoKgX95IKywqLEDnhymxxIujtVOlG =dEo8 -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Wed Apr 14 18:46:54 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 18:44:08 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <883640782.20040414130643@zedat.fu-berlin.de> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643@zedat.fu-berlin.de> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 14 Apr 2004, Johannes Posel wrote: > Well, one thing that tends to be forgotten is that PGP/MIME absolutely > reliably breaks list footers ;-) =========================== inline signatures don't ~break~ list footers, but since the footers exist outside of the pgp headers, they tend to not be displayed after the message is piped through gpg. at least that's the way pine handles it with ez-pine-gpg and the other gpg filters i've tried. i consider this a good thing, at least for lists where adverts are added to the bottom of every email ;) ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Anyone who doubts that terrorists could smuggle a nuclear warhead into New York City should note that they could always wrap it in a bale of marijuana." -- Graham Allison, The Boston Globe 27 October 1999 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB9awIACgkQnCgLvz19QeNCPgCgnTU918L/n5Mhef346eYL8g0c hY4An1/QicvJP/TTCP3z4N3nK2kMcKaR =1ujF -----END PGP SIGNATURE----- From torduninja at netcourrier.com Wed Apr 14 09:53:11 2004 From: torduninja at netcourrier.com (Maxine Brandt) Date: Wed Apr 14 19:18:46 2004 Subject: openPGP vs x509 Message-ID: <407CEDE7.5020503@netcourrier.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Per Tunedal wrote: > OT? I have been looking for tools to make self-signed X.509 but > haven't > found any. Do you have any hints? Links? > Rational: some people refuse using PGP-encryption/signing. On Linux there's TinyCA to set up your own CA and issue e-mail or server certcificates. Very simple to use. http://tinyca.sm-zone.net Salut Maxine - -- PGP/GPG keys: http://www.torduninja.tk -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) iD8DBQFAfO3gKBY/R6nbCcARAghnAJ9eMIBAI/V2Z/KVXckUYzg47yX1RQCeOXgE w1h81x/VS6tS/bE3N1pp+Ok= =s+CI -----END PGP SIGNATURE----- From svwright+lists at amtp.liv.ac.uk Wed Apr 14 19:34:41 2004 From: svwright+lists at amtp.liv.ac.uk (Stewart V. Wright) Date: Wed Apr 14 19:31:50 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643@zedat.fu-berlin.de> Message-ID: <20040414173441.GD6699@amtp.liv.ac.uk> G'day Atom, * Atom 'Smasher' [040414 18:09]: > inline signatures don't ~break~ list footers, but since the footers exist > outside of the pgp headers, they tend to not be displayed after the > message is piped through gpg. at least that's the way pine handles it with > ez-pine-gpg and the other gpg filters i've tried. That is a _seriously_ broken tool that you're using. How are you to know whether something important (but not signed) is included after a signed message? What if someone (who doesn't use GPG) forwarded a signed message to you and then made comments after the included message? Ugh! I'd try to get your system fixed and use mutt, or some other working MUA... > i consider this a good thing, at least for lists where adverts are added > to the bottom of every email ;) I would like someone (*Hint*) to use the '-- ' standard for their "advertising", oops, sorry, signature. Try Googling on sig.etiquette... :) (Now I'm off to have another cup of coffee to stop being so grumpy...) S. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 274 bytes Desc: Digital signature Url : /pipermail/attachments/20040414/d3876fa0/attachment.bin From svwright+lists at amtp.liv.ac.uk Wed Apr 14 19:45:44 2004 From: svwright+lists at amtp.liv.ac.uk (Stewart V. Wright) Date: Wed Apr 14 19:42:55 2004 Subject: pine vs mutt - was: Re: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> Message-ID: <20040414174544.GE6699@amtp.liv.ac.uk> G'day Atom, * Atom 'Smasher' [040414 18:04]: > i'd like to switch to a free (as in speech) MUA for ideological reasons. > and of course, everyone knows that mutt is ~better~. *Grin* Mutt is better - before you even look at the freedom issue, but that is a side issue. > i've tried mutt several times, and i'm willing to learn a new interface, > but it just can't do what i need. i've asked in the mutt mailing list & > usenet, but i'm told the problem (with colors) is in ncurses/slang, not > mutt. either way, mutt can't do what i need, and in the meantime pine can, > and for me the color support is necessary. > > i keep checking back with mutt every few years, and when they get the > color support as good as pine has, i'm confident that i can either > configure everything else the way i need it or write my own hooks. Do you want to take this off-list for a while to see if I can make any suggestions to help your transition? S. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 274 bytes Desc: Digital signature Url : /pipermail/attachments/20040414/3185c4b0/attachment.bin From pschott at drivefinancial.com Wed Apr 14 19:26:20 2004 From: pschott at drivefinancial.com (Peter Schott) Date: Wed Apr 14 19:46:27 2004 Subject: Gnupg-users Digest, Vol 7, Issue 19 Message-ID: <4E28ECEE2E06784AA8921F82878C889E9D14C8@DFSTXEXCH3.dfs.com> Kevin, Not sure what exactly you're trying to do. I've been using GPG on a Win2K3 cluster with no issues, but I use it in batch jobs, not an e-mail form. Perhaps if you posted more information about the e-mail form you're trying to use? -Pete Schott ====================================================== Date: Tue, 13 Apr 2004 15:23:50 -0400 From: "Kevin McNally" Subject: Windows 2003 installation I am really new to GNUPG and I had a encryption email form working correctly on Windows 2K server. I put it on the windows 2003 server and now nothing seems to work. I have looked all over for help on this with no luck, has anyone successfully gotten GNUPG working on windows 2003 or maybe someone saw a help file somewhere. Any help would be greatly appreciated. Thanks Kevin ___________________________________________________________________________________ This e-mail is covered by the Electronic Communications Privacy Act, 18 U.S.C. Sections 2510-2521. The information contained in this e-mail is confidential and intended only for use of the individual or entity named above. If the reader of this message is not the intended recipient, or the employee or agent responsible to deliver it to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this message in error or there are any problems please notify the originator immediately. The unauthorized use, disclosure, copying or alteration of this message is strictly forbidden. This mail and any attachments have been scanned for viruses prior to leaving the Drive Financial Services network. Drive Financial Services will not be liable for direct, special, indirect or consequential damages arising from alteration of the contents of this message by a third party or as a result of any virus being passed on. ___________________________________________________________________________________ From atom-gpg at suspicious.org Wed Apr 14 20:22:30 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 20:19:50 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <20040414173441.GD6699@amtp.liv.ac.uk> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643@zedat.fu-berlin.de> <20040414173441.GD6699@amtp.liv.ac.uk> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 14 Apr 2004, Stewart V. Wright wrote: > * Atom 'Smasher' [040414 18:09]: > > inline signatures don't ~break~ list footers, but since the footers exist > > outside of the pgp headers, they tend to not be displayed after the > > message is piped through gpg. at least that's the way pine handles it with > > ez-pine-gpg and the other gpg filters i've tried. > > That is a _seriously_ broken tool that you're using. How are you to > know whether something important (but not signed) is included after a > signed message? What if someone (who doesn't use GPG) forwarded a > signed message to you and then made comments after the included > message? Ugh! I'd try to get your system fixed and use mutt, or some > other working MUA... ============================== well, if that text is included then how would a recipient tell that it ISN'T signed? i consider it a security feature that all any non-signed text is suppressed. and i don't know of ANY list where all of the messages are signed, so a user would still be able to see such information from other users' posts... and... if anything that important is happening on the list, then it probably warrants a message, rather than a footer. footers below the pgp headers can still be viewed (in pine) using the [H]eader command, set up to display the raw message. > > i consider this a good thing, at least for lists where adverts are added > > to the bottom of every email ;) > > I would like someone (*Hint*) to use the '-- ' standard for their > "advertising", oops, sorry, signature. Try Googling on > sig.etiquette... :) ============================ what RFC is that? ;p ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Since trade ignores national boundaries and the manufacturer insists on having the world as a market, the flag of his nation must follow him, and the doors of the nations which are closed against him must be battered down. Concessions obtained by financiers must be safeguarded by ministers of state, even if the sovereignty of unwilling nations be outraged in the process. Colonies must be obtained or planted, in order that no useful corner of the world may be overlooked or left unused." -- Woodrow Wilson, President of the United States, 1919 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB9gWoACgkQnCgLvz19QeNhGwCcD3b5AGXHIYtd610tD4iNine8 vncAn1amkIq1NZ/lzkUIBxBYZMM8esAN =VMV2 -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Wed Apr 14 20:31:10 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 20:28:24 2004 Subject: pine vs mutt - was: Re: pgp/mime vs in-line pgp In-Reply-To: <20040414174544.GE6699@amtp.liv.ac.uk> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <20040414174544.GE6699@amtp.liv.ac.uk> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 14 Apr 2004, Stewart V. Wright wrote: > * Atom 'Smasher' [040414 18:04]: > > i'd like to switch to a free (as in speech) MUA for ideological reasons. > > and of course, everyone knows that mutt is ~better~. > > *Grin* Mutt is better - before you even look at the freedom issue, > but that is a side issue. ======================== i knew ~someone~ here would agree with me ;) > > i've tried mutt several times, and i'm willing to learn a new interface, > > but it just can't do what i need. i've asked in the mutt mailing list & > > usenet, but i'm told the problem (with colors) is in ncurses/slang, not > > mutt. either way, mutt can't do what i need, and in the meantime pine can, > > and for me the color support is necessary. > > > > i keep checking back with mutt every few years, and when they get the > > color support as good as pine has, i'm confident that i can either > > configure everything else the way i need it or write my own hooks. > > Do you want to take this off-list for a while to see if I can make any > suggestions to help your transition? ========================= i know there are plenty of people here who use/advocate mutt, and probably some hard-core power users... why not see if anyone can help me migrate to a ~better~ MUA than pine... here's what i'm running into... http://groups.google.com/groups?threadm=103omg7t0uqgm78%40corp.supernews.com since this is veering off topic, feel free to send replies off list. thanks...! ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "This society began with stolen labor on stolen land, which makes all such claims to a tradition of freedom, null and void. It must also be said, that democracy and slavery cannot exist side by side. Nor can democracy and capitalism for that matter. No class society, based on exploitation of the many to enrich a greedy few, can properly claim to be a democracy." -- Ron Wilkins, KPFK (Pacifica), Los Angeles -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB9g3MACgkQnCgLvz19QeOEmgCfXdE82AS8NDIxfyYhiRe/9d1b jssAoKPJLugmiB/TRhgZzRDTzjVPDwco =JCO8 -----END PGP SIGNATURE----- From svwright+lists at amtp.liv.ac.uk Wed Apr 14 20:43:44 2004 From: svwright+lists at amtp.liv.ac.uk (Stewart V. Wright) Date: Wed Apr 14 20:40:54 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643@zedat.fu-berlin.de> <20040414173441.GD6699@amtp.liv.ac.uk> Message-ID: <20040414184344.GF6699@amtp.liv.ac.uk> G'day Atom, * Atom 'Smasher' [040414 19:25]: > well, if that text is included then how would a recipient tell that it > ISN'T signed? I've put a (temporary) screenshot up showing how mutt handles this with your last message: http://www.liv.ac.uk/~svwright/Screenshot-Gnome-terminal.png The '[-- END PGP SIGNED MESSAGE --]' is a dead give away! :-) > i consider it a security feature that all any non-signed > text is suppressed. Each to their own... I'm going to let others smack you about for this position. > and i don't know of ANY list where all of the messages > are signed, so a user would still be able to see such information from > other users' posts... and... if anything that important is happening on > the list, then it probably warrants a message, rather than a footer. You're missing the point. Say I included the WHOLE of your signed message and then made some important comments below it - you'd miss them. > > I would like someone (*Hint*) to use the '-- ' standard for their > > "advertising", oops, sorry, signature. Try Googling on > > sig.etiquette... :) > ============================ > > what RFC is that? ;p A quick google turned up this... RFC 2646 ("The Text/Plain Format Parameter") references the "-- " signature separator (with the space) as being a Usenet convention: "There is a convention in Usenet news of using '-- ' as the separator line between the body and the signature of a message." http://www.rfc-editor.org/rfc/rfc2646.txt It's not exactly for email, but......... As a side issue it is just nicer for people to be able to set up their editor to strip the .sig automatically. Cheers, S. P.S. You just included that bloody long quote to get on my nerves!!! :-P -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 274 bytes Desc: Digital signature Url : /pipermail/attachments/20040414/1fa0dfe4/attachment.bin From atom-gpg at suspicious.org Wed Apr 14 21:04:33 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 21:01:45 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <20040414184344.GF6699@amtp.liv.ac.uk> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <883640782.20040414130643@zedat.fu-berlin.de> <20040414173441.GD6699@amtp.liv.ac.uk> <20040414184344.GF6699@amtp.liv.ac.uk> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 14 Apr 2004, Stewart V. Wright wrote: > * Atom 'Smasher' [040414 19:25]: > > i consider it a security feature that all any non-signed > > text is suppressed. > > Each to their own... I'm going to let others smack you about for this > position. =================== i would be interested in hearing reasons to include footers in the display... maybe if someone can "show me the light" i'll update the ez-pine-gpg. > > and i don't know of ANY list where all of the messages > > are signed, so a user would still be able to see such information from > > other users' posts... and... if anything that important is happening on > > the list, then it probably warrants a message, rather than a footer. > > You're missing the point. Say I included the WHOLE of your signed > message and then made some important comments below it - you'd miss > them. ===================== would you include an extra line above what you're quoting? or append '> ' to each line that's being quoted? that would keep everything behaving properly. nonetheless, i *will* count that as the first argument towards changing the behavior of ez-pine-gpg... which AFAIK would then be the only gpg filter for pine that behaves that way that. > > > I would like someone (*Hint*) to use the '-- ' standard for their > > > "advertising", oops, sorry, signature. Try Googling on > > > sig.etiquette... :) > > ============================ > > > > what RFC is that? ;p > > A quick google turned up this... > > RFC 2646 ("The Text/Plain Format Parameter") references the "-- " > signature separator (with the space) as being a Usenet convention: > > "There is a convention in Usenet news of using '-- ' as the separator > line between the body and the signature of a message." ====================== seriously though, the reason that i don't use that convention is because i send signed mail to lists and people who don't use pgp/gpg... and i don't want my email to be formatted differently depending on whether or not it is signed... so, to the extent that i can, i avoid using lines that start with a dash. > P.S. You just included that bloody long quote to get on my nerves!!! ================= it wasn't me! it's the computer! i'll yell at it, for you ;) here's a short one... ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Facts are stupid things." -- Ronald Reagan (1988) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB9i0QACgkQnCgLvz19QePdmgCeLPtHyVqHCBdaMqg2kwpRlSaJ 6FMAn3IkDLEHao16LU1UfsJekSNEUgac =M8Zj -----END PGP SIGNATURE----- From anonymous at remailer.metacolo.com Wed Apr 14 21:56:14 2004 From: anonymous at remailer.metacolo.com (Anonymous Sender) Date: Wed Apr 14 21:53:25 2004 Subject: openPGP vs x509 Message-ID: <446a1151348bb39fcc52293b505c80bf@remailer.metacolo.com> Werner Koch: > On Wed, 14 Apr 2004 06:18:55 +0000 (UTC), Anonymous Sender said: > > 'curl' and 'libcurl' do support TLS client authentication: > Sure, but I was not talking about hackers but average computer users. Another data point: 'lftp' (when compiled with the OpenSSL library) does support client authentication for both ftps and https: $ lftp --version Lftp | Version 3.0.1 | Copyright (c) 1996-2004 Alexander V. Lukyanov [ ... ] $ lftp lftp :~> set ssl:key-file /path/to/certs/user1.key lftp :~> set ssl:cert-file /path/to/certs/user1.cert lftp :~> set ssl:verify-certificate true lftp :~> set ssl:ca-file /path/to/certs/CAs.cert lftp :~> open https://localhost lftp localhost:/> dir -rw-r--r-- -- a.html -rw-r--r-- -- b.html -rw-r--r-- -- c.html -rw-r--r-- -- d.txt -rw-r--r-- -- e.html lftp localhost:/> and the logs clearly show that client authentication did occur: www connected from 127.0.0.1:2264 VERIFY OK: depth=1, /C=XX/CN=xxxxxxxxx Certification Authority VERIFY OK: depth=0, /C=XX/CN=user1 Negotiated ciphers: AES256-SHA SSLv3 Kx=RSA Au=RSA Enc=AES(256) Mac=SHA1 Connection closed: 1487 bytes sent to SSL, 96 bytes sent to socket From linux at codehelp.co.uk Wed Apr 14 22:16:42 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Wed Apr 14 22:13:58 2004 Subject: Problems importing public key In-Reply-To: <1081923324.1671.17.camel@localhost.localdomain> References: <1081923324.1671.17.camel@localhost.localdomain> Message-ID: <200404142116.47755.linux@codehelp.co.uk> On Wednesday 14 Apr 2004 7:15, Graeme Nichols wrote: > Hello Folks, > > I was sent a public key by a fellow so that I could encrypt a file to > him. The name of the file, for what it is worth is: 0xF94BBB03.asc which You might be able to trust this file, but why should GnuPG? You've just imported a public key that has no relation to your own key, there is nothing for GnuPG to use to work out whether to trust the key. You comment that this is for sensitive data yet you seem prepared to take the key at face value. > happens to be the DSA key ID of his key used to sign his emails. It So all you really know about this key is that the email address matches the keyID. Is that enough? Can't be particularly sensitive data for encryption! > imports OK into my gnupg V1.2.3 but when I click on the lock icon in > Evolution (so I can test the public key) I get the following error: > > gpg: armor header: Version: GnuPG v1.2.4 (GNU/Linux) > gpg: Signature made Sat 03 Apr 2004 09:54:31 EST using DSA key ID > F94BBB03 > gpg: BAD signature from "Benoit Gr?goire (Serveur t?l?matique des Oops. You've got the right key but something is wrong with the email. > ?tudiants de Polytechnique) " > gpg: textmode signature, digest algorithm SHA1 > > What have I done wrong? Nothing, necessarily. There may be something wrong with the signed email though. > I have another public key that I imported from a keyserver that works OK If you import a new copy of the same key, it'll simply overwrite the old one. The only differences that will make any odds here are things like extra subkeys, extra signatures, extra UID's. If GnuPG accepts these two keys as the same, the public key in both copies is the same. > if I click the lock icon on Evolution yet if I fire up seahorse and > click on the Key Manager icon to list the keys I have under the 'trust' > column both the public keys I have imported show 'ERROR'. Mine naturally > shows 'ultimate' This is the separate trust issue. GnuPG cannot trust this key because you haven't verified the key. You would need to follow the keysigning procedure and then sign the key for GnuPG to be able to trust this key. > I really need to find out if I have done something wrong so I can No, just that there is something that you haven't yet done. > rectify the problem as Benoit is waiting for the file he needs me to > send him and I am not going to send it if something is wrong as the file > has some very private data in it and I don't want it to fall into the > wrong hands. Then you MUST verify the key properly. Usually, this involves meeting face-to-face to exchange GnuPG fingerprints and verify proof of photo ID AS WELL as verifying that the email address in the key is the right destination. GnuPG cannot be expected to encrypt sensitive data if there no way of knowing if you are encrypting to the right person. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040414/f7a54b06/attachment.bin From DougB at DougBarton.net Wed Apr 14 22:16:18 2004 From: DougB at DougBarton.net (Doug Barton) Date: Wed Apr 14 22:14:09 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> Message-ID: <20040414130024.U8942@qbhto.arg> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tue, 13 Apr 2004, Len Sassaman wrote: > Unfortunately, it's not that easy. Pine is under a "non-free" license, and > the Pine authors have repeatedly rejected such patches. One could argue that large parts of this thread have been off topic for this list, however I'm firmly of the opinion that the type of political rhetoric about licenses that you've stated here is definitely off topic, and personally I'd appreciate it if it was just skipped altogether. > They have no interest in OpenPGP support, period. I'm not sure this is accurate either, or at least I think you're being too specific by mentioning OpenPGP. It would be safer to say that PGP integration in general has not been a goal for the Pine development team, and that it's handled fairly well by a variety of third party plugins. In regards to the MIME issue, Pine doesn't give the user the ability to manipulate attachments at all, so you can't create PGP MIME messages. (You can of course attach files to e-mail, you just can't script the attachment process like you can other things, like message creation.) It's not specifically a PGP issue. I do have a filter for Pine that will allow you to verify properly formatted PGP MIME messages in Pine though. As for the "mutt is better, so you should use it" argument, I think that "better" is a very relative term, and that MUA choices in general are a topic of great emotion. :) The one perspective that I don't have any sympathy for though is that of, "Well, those people who use $MUA are losers, so you shouldn't care if you can't communicate with them securely." Some of us actually NEED to be able to communicate with those people, have absolutely no chance of convincing them to change, and therefore need to find valid solutions. If you are not in that group, lucky you, but telling those of us who are that our concerns are pointless isn't really useful. Speaking only for myself, Doug - -- If you're never wrong, you're not trying hard enough. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFAfZwXyIakK9Wy8PsRAqdaAKDttFdszGHfiW0rIXlap+0+CVaZnACghVtT bFffS9DZ3Fb054wDHKDOTl4= =7ug0 -----END PGP SIGNATURE----- From rabbi at abditum.com Wed Apr 14 22:26:23 2004 From: rabbi at abditum.com (Len Sassaman) Date: Wed Apr 14 22:23:38 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <20040414130024.U8942@qbhto.arg> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <20040414130024.U8942@qbhto.arg> Message-ID: On Wed, 14 Apr 2004, Doug Barton wrote: > > Unfortunately, it's not that easy. Pine is under a "non-free" license, and > > the Pine authors have repeatedly rejected such patches. > > One could argue that large parts of this thread have been off topic for > this list, however I'm firmly of the opinion that the type of political > rhetoric about licenses that you've stated here is definitely off topic, > and personally I'd appreciate it if it was just skipped altogether. It is fortunate then that your opinion is irrelevant to this discussion. > > They have no interest in OpenPGP support, period. > > I'm not sure this is accurate either, or at least I think you're being > too specific by mentioning OpenPGP. It would be safer to say that PGP > integration in general has not been a goal for the Pine development > team, and that it's handled fairly well by a variety of third party > plugins. It is handled horribly. Show me one good plugin that supports PGP well. PGP4Pine? pgpenvelope? I've been involved in the development/testing of both of those for years, and they both are horrible. It is the latter that I use, since it is the best that Pine can get us, but that doesn't mean it is good. This is a fault not of the authors of the plugins, but of the Pine plugin API. > In regards to the MIME issue, Pine doesn't give the user the ability to > manipulate attachments at all, so you can't create PGP MIME messages. A major flaw in Pine. > therefore need to find valid solutions. If you are not in that group, > lucky you, but telling those of us who are that our concerns are > pointless isn't really useful. When did I ever say anything about pointless? I am a Pine user. I believe Mutt to be greatly at fault for much of the PGP/MIME issues, by their repeated propaganda attempts to convince PGP users that inline-signatures are "old-style" or "deprecated" in favor of the PGP/MIME format (originally devised by the author of Mutt, surprisingly enough.) But none of this changes the fact that the answer to Pine's problems is *not* "go implement it yourself and it will all be better." In order for there to be good OpenPGP/MIME support in Pine, the Pine authors must decide there will be. There is no possibility of a fork. This is something that is fact, and you will need to deal with if you are a Pine user. From ipalette at hotmail.com Wed Apr 14 22:31:35 2004 From: ipalette at hotmail.com (Kevin McNally) Date: Wed Apr 14 22:29:14 2004 Subject: Gnupg-users Digest, Vol 7, Issue 19 Message-ID: Thanks Pete, >From the little information I have found it may be a permissions issue on the server, but I can not be sure. Basically this is what happens: 1. Someone fills out a form on the website and it is emailed to someone I created a key for. 2. The recipient gets an email with an encrypted message in the body. 3. Once the the email is received, she goes to a seperate webpage on the server to decrypt it. 4. She copies the encrypted body of the message into a text box on the page and types the pass phrase into a sperate field. 5. The form data is formatted and dispalyed for her to use. Does that makes sense? When I try to use a sample to try and decode a message I get this error: PHP Warning: exec(): Unable to fork [echo Life is Motivational|C:\GnuPG\gpg --passphrase-fd 0 -d C:\GnuPG\temp.gpg] in C:\Domains\domain.org\wwwroot\cgi-bin\decode.php on line 59 Now if I copy the file "cmd.exe" into the php folder I get this error (I read somewhere this might help): CGI Error The specified CGI application misbehaved by not returning a complete set of HTTP headers. Thanks Kevin >From: "Peter Schott" >To: >Subject: RE: Gnupg-users Digest, Vol 7, Issue 19 >Date: Wed, 14 Apr 2004 12:26:20 -0500 > >Kevin, > Not sure what exactly you're trying to do. I've been using GPG >on a Win2K3 cluster with no issues, but I use it in batch jobs, not an >e-mail form. Perhaps if you posted more information about the e-mail >form you're trying to use? > > -Pete Schott > >====================================================== >Date: Tue, 13 Apr 2004 15:23:50 -0400 >From: "Kevin McNally" >Subject: Windows 2003 installation > >I am really new to GNUPG and I had a encryption email form working >correctly on Windows 2K server. I put it on the windows 2003 server and >now nothing seems to work. I have looked all over for help on this with >no luck, has anyone successfully gotten GNUPG working on windows 2003 or >maybe someone saw a help file somewhere. Any help would be greatly >appreciated. > >Thanks >Kevin > > >___________________________________________________________________________________ >This e-mail is covered by the Electronic Communications Privacy Act, 18 >U.S.C. >Sections 2510-2521. The information contained in this e-mail is >confidential >and intended only for use of the individual or entity named above. If the >reader >of this message is not the intended recipient, or the employee or agent >responsible to deliver it to the intended recipient, you are hereby >notified >that any dissemination, distribution or copying of this communication is >strictly >prohibited. If you have received this message in error or there are any >problems >please notify the originator immediately. > >The unauthorized use, disclosure, copying or alteration of this message is >strictly forbidden. This mail and any attachments have been scanned for >viruses >prior to leaving the Drive Financial Services network. Drive Financial >Services >will not be liable for direct, special, indirect or consequential damages >arising >from alteration of the contents of this message by a third party or as a >result of >any virus being passed on. >___________________________________________________________________________________ > > >_______________________________________________ >Gnupg-users mailing list >Gnupg-users@gnupg.org >http://lists.gnupg.org/mailman/listinfo/gnupg-users _________________________________________________________________ Persistent heartburn? Check out Digestive Health & Wellness for information and advice. http://gerd.msn.com/default.asp From hmujtaba at forumsys.com Wed Apr 14 22:44:27 2004 From: hmujtaba at forumsys.com (Hasnain Mujtaba) Date: Wed Apr 14 22:42:13 2004 Subject: Most commonly used OpenPGP symmetric ciphers Message-ID: <4DCE15B9C4E66F4CA967EBF64C53D64D67B31D@bstn-exch1.forumsys.com> Hi, The OpenPGP standard supports many symmetric key algorithms (IDEA, 3-DES, CAST5, AES, Blowfish, SAFER). I was wondering which of these algorithms are most commonly used in the industry. Is there a most commonly used algorithm from this list? Thanks Hasnain. ---- The information contained in this electronic mail and any attached document is the confidential and proprietary business information of Forum Systems, Inc. It is intended solely for the addressed recipient listed above. It may not be distributed in any manner without the express written consent of Forum Systems, Inc. From atom-gpg at suspicious.org Wed Apr 14 23:00:36 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 22:57:55 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <20040414130024.U8942@qbhto.arg> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 14 Apr 2004, Len Sassaman wrote: > On Wed, 14 Apr 2004, Doug Barton wrote: > > > Unfortunately, it's not that easy. Pine is under a "non-free" license, and > > > the Pine authors have repeatedly rejected such patches. > > > > One could argue that large parts of this thread have been off topic for > > this list, however I'm firmly of the opinion that the type of political > > rhetoric about licenses that you've stated here is definitely off topic, > > and personally I'd appreciate it if it was just skipped altogether. ========================= that's not "political rhetoric", that's a point of fact regarding pine. as long as it involves the USE of gpg, i'd say it's on topic here. > > > They have no interest in OpenPGP support, period. > > > > I'm not sure this is accurate either, or at least I think you're being > > too specific by mentioning OpenPGP. It would be safer to say that PGP > > integration in general has not been a goal for the Pine development > > team, and that it's handled fairly well by a variety of third party > > plugins. > > It is handled horribly. Show me one good plugin that supports PGP well. > PGP4Pine? pgpenvelope? I've been involved in the development/testing of > both of those for years, and they both are horrible. It is the latter that > I use, since it is the best that Pine can get us, but that doesn't mean it > is good. This is a fault not of the authors of the plugins, but of the > Pine plugin API. ==================== they don't seem to have an interest in security, either (i say this after raising security issues with the pine devel group, and receiving an asinine response that demonstrated a serious lack of understanding about security protocols). despite my own use of pine, this is evident in the pine changelog. as i become more aware of these issues, my motivation to migrate away from pine grows stronger. len, have you tried ez-pine-gpg? it was written after i was disappointed with the other gpg/pine filters. http://business-php.com/opensource/ez-pine-gpg/ > > In regards to the MIME issue, Pine doesn't give the user the ability to > > manipulate attachments at all, so you can't create PGP MIME messages. > > A major flaw in Pine. ======================= and a violation of the pgp/mime and s/mime RFCs. <> > But none of this changes the fact that the answer to Pine's problems is > *not* "go implement it yourself and it will all be better." ====================== *if* the pine API even allows it... ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Since trade ignores national boundaries and the manufacturer insists on having the world as a market, the flag of his nation must follow him, and the doors of the nations which are closed against him must be battered down. Concessions obtained by financiers must be safeguarded by ministers of state, even if the sovereignty of unwilling nations be outraged in the process. Colonies must be obtained or planted, in order that no useful corner of the world may be overlooked or left unused." -- Woodrow Wilson, President of the United States, 1919 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB9pngACgkQnCgLvz19QeN/GACgh4f4GwkI4uNKVTpzcbDL4Rgh n+MAnihTTRospSO9r23sUYf2ulu9LiPo =ZaV9 -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Wed Apr 14 23:01:08 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Wed Apr 14 22:58:30 2004 Subject: Most commonly used OpenPGP symmetric ciphers In-Reply-To: <4DCE15B9C4E66F4CA967EBF64C53D64D67B31D@bstn-exch1.forumsys.com> References: <4DCE15B9C4E66F4CA967EBF64C53D64D67B31D@bstn-exch1.forumsys.com> Message-ID: <20040414210108.GB28820@jabberwocky.com> On Wed, Apr 14, 2004 at 04:44:27PM -0400, Hasnain Mujtaba wrote: > Hi, > > The OpenPGP standard supports many symmetric key algorithms (IDEA, > 3-DES, CAST5, AES, Blowfish, SAFER). I was wondering which of these > algorithms are most commonly used in the industry. Is there a most > commonly used algorithm from this list? Most commonly used is a hard question. Most commonly used for *what*? Historically, banking systems used 3DES (common for financial stuff). PGP users used IDEA (because that is all that PGP 2 had), etc. David From linux at codehelp.co.uk Wed Apr 14 23:12:36 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Wed Apr 14 23:09:49 2004 Subject: Gnupg-users Digest, Vol 7, Issue 19 In-Reply-To: References: Message-ID: <200404142212.37427.linux@codehelp.co.uk> On Wednesday 14 Apr 2004 9:31, Kevin McNally wrote: > Thanks Pete, > > >>From the little information I have found it may be a permissions issue on > > the server, but I can not be sure. Basically this is what happens: > 1. Someone fills out a form on the website and it is emailed to someone I > created a key for. > 2. The recipient gets an email with an encrypted message in the body. > 3. Once the the email is received, she goes to a seperate webpage on the > server to decrypt it. > 4. She copies the encrypted body of the message into a text box on the page > and types the pass phrase into a sperate field. > 5. The form data is formatted and dispalyed for her to use. At which point, all your effort in encrypting the transmitted data is lost! Everything that the server displays in the browser is sent in plain text - including the decrypted block!! Unless the block is decrypted locally, it will be pointless encrypting it in the first place! If you never encrypted it in the first place, the data would still be sent once in clear text, just like in your system. The second problem is that the PASSPHRASE is sent in clear text too, so now the key is compromised too, or are you doing this over https:// ? > > Does that makes sense? Umm, No. Sorry. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040414/8dde32be/attachment.bin From atom-gpg at suspicious.org Wed Apr 14 23:18:07 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 14 23:15:19 2004 Subject: Most commonly used OpenPGP symmetric ciphers In-Reply-To: <4DCE15B9C4E66F4CA967EBF64C53D64D67B31D@bstn-exch1.forumsys.com> References: <4DCE15B9C4E66F4CA967EBF64C53D64D67B31D@bstn-exch1.forumsys.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > The OpenPGP standard supports many symmetric key algorithms (IDEA, > 3-DES, CAST5, AES, Blowfish, SAFER). I was wondering which of these > algorithms are most commonly used in the industry. Is there a most > commonly used algorithm from this list? ============================ most people never change the defaults, so AES-256 seems to be the most widely used first preference with newer keys. for a better answer of what preferences are really out there, maybe someone who maintains a key-server can run that report. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Proprietary software seeks to maximize its value solely in monetary terms by achieving a monopoly. Open Source software maximizes its value by assuring that a monopoly cannot be achieved." -- Mark Webbink, Senior Vice President and General Counsel of Red Hat -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB9qpMACgkQnCgLvz19QeOvxwCfWukug/uhjSgdXOjvuy2eUMqt /jAAnjalliXF5TtKtcF55dIeLU2d4x7J =rpBm -----END PGP SIGNATURE----- From jas at extundo.com Wed Apr 14 23:16:48 2004 From: jas at extundo.com (Simon Josefsson) Date: Wed Apr 14 23:17:56 2004 Subject: pgp/mime vs in-line pgp References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <20040414130024.U8942@qbhto.arg> Message-ID: Len Sassaman writes: > I am a Pine user. I believe Mutt to be greatly at fault for much of the > PGP/MIME issues, by their repeated propaganda attempts to convince PGP > users that inline-signatures are "old-style" or "deprecated" in favor of > the PGP/MIME format (originally devised by the author of Mutt, > surprisingly enough.) For messaging applications, I don't think it is unreasonable to consider plain OpenPGP is "old-style" and "deprecated". Quoting RFC 2440: Note that many applications, particularly messaging applications, will want more advanced features as described in the OpenPGP-MIME document, RFC 2015. An application that implements OpenPGP for messaging SHOULD implement OpenPGP-MIME. For plain OpenPGP to work in e-mail, you pretty much need to only use ASCII. If you want to be compatible with PGP 2.x/5.x, there are even more restrictions (no dash-escaped text, allegedly you also need to remove trailing whitespace which would break format=flowed, and probably more that I forget). Regards, Simon From graham.todd at dsl.pipex.com Wed Apr 14 23:42:48 2004 From: graham.todd at dsl.pipex.com (Graham) Date: Wed Apr 14 23:39:55 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: <200404140938.10167@fortytwo.ch> References: <200404132246.23133.graham.todd@dsl.pipex.com> <200404140938.10167@fortytwo.ch> Message-ID: <200404142242.48501.graham.todd@dsl.pipex.com> On Wednesday 14 Apr 2004 8:38 am, Adrian 'Dagurashibanipal' von Bidder wrote: > You mean MUA. The only requirement on a MTA by PGP is that it should > be able to transport byte-for-byte copies of ascii messages. My apologies. You are correct (my bad typing) -- Graham From DougB at DougBarton.net Thu Apr 15 00:45:24 2004 From: DougB at DougBarton.net (Doug Barton) Date: Thu Apr 15 00:43:09 2004 Subject: pgp/mime vs in-line pgp In-Reply-To: References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <20040414130024.U8942@qbhto.arg> Message-ID: <20040414153957.U8942@qbhto.arg> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 14 Apr 2004, Len Sassaman wrote: > On Wed, 14 Apr 2004, Doug Barton wrote: > > > > Unfortunately, it's not that easy. Pine is under a "non-free" license, and > > > the Pine authors have repeatedly rejected such patches. > > > > One could argue that large parts of this thread have been off topic for > > this list, however I'm firmly of the opinion that the type of political > > rhetoric about licenses that you've stated here is definitely off topic, > > and personally I'd appreciate it if it was just skipped altogether. > > It is fortunate then that your opinion is irrelevant to this discussion. Um, ok, thanks for that. My point is simply that arguing about whether Pine's license is "Free" or not, in my opinion, is not on topic here. As a member of the list I'm simply making a request, which you are of course free to disregard. > > > They have no interest in OpenPGP support, period. > > > > I'm not sure this is accurate either, or at least I think you're being > > too specific by mentioning OpenPGP. It would be safer to say that PGP > > integration in general has not been a goal for the Pine development > > team, and that it's handled fairly well by a variety of third party > > plugins. > > It is handled horribly. Show me one good plugin that supports PGP well. Well I'm biased, but you can find mine at http://dougbarton.net/FreeBSD/Downloads/. I think that "well" is a relative term here, and I'm sure that you will find things in that package that you won't like. :) > > In regards to the MIME issue, Pine doesn't give the user the ability to > > manipulate attachments at all, so you can't create PGP MIME messages. > > A major flaw in Pine. Personally I don't miss it, but YMMV. > > therefore need to find valid solutions. If you are not in that group, > > lucky you, but telling those of us who are that our concerns are > > pointless isn't really useful. > > When did I ever say anything about pointless? Sorry I wasn't clear. I was referring here to the thread in general, not replying directly to you. > I am a Pine user. I believe Mutt to be greatly at fault for much of the > PGP/MIME issues, by their repeated propaganda attempts to convince PGP > users that inline-signatures are "old-style" or "deprecated" in favor of > the PGP/MIME format (originally devised by the author of Mutt, > surprisingly enough.) I agree, I find that rhetoric inappropriate and pretty much non-useful. The topic of in line signatures was discussed on the list recently, so I won't rehash that here. Doug - -- If you're never wrong, you're not trying hard enough. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFAfb8GyIakK9Wy8PsRAmriAJ42xB7aZtOqC0RJd9EIMxzMPN4OggCg/OE+ 7WjZaoiGMGzMCAVstmrOxKA= =Dj4S -----END PGP SIGNATURE----- From anonymous at remailer.metacolo.com Thu Apr 15 01:44:05 2004 From: anonymous at remailer.metacolo.com (Anonymous Sender) Date: Thu Apr 15 01:41:15 2004 Subject: openPGP vs x509 Message-ID: <446a1151348bb39fcc52293b505c80bf@remailer.metacolo.com> Werner Koch: > On Wed, 14 Apr 2004 06:18:55 +0000 (UTC), Anonymous Sender said: > > 'curl' and 'libcurl' do support TLS client authentication: > Sure, but I was not talking about hackers but average computer users. Another data point: 'lftp' (when compiled with the OpenSSL library) does support client authentication for both ftps and https: $ lftp --version Lftp | Version 3.0.1 | Copyright (c) 1996-2004 Alexander V. Lukyanov [ ... ] $ lftp lftp :~> set ssl:key-file /path/to/certs/user1.key lftp :~> set ssl:cert-file /path/to/certs/user1.cert lftp :~> set ssl:verify-certificate true lftp :~> set ssl:ca-file /path/to/certs/CAs.cert lftp :~> open https://localhost lftp localhost:/> dir -rw-r--r-- -- a.html -rw-r--r-- -- b.html -rw-r--r-- -- c.html -rw-r--r-- -- d.txt -rw-r--r-- -- e.html lftp localhost:/> and the logs clearly show that client authentication did occur: www connected from 127.0.0.1:2264 VERIFY OK: depth=1, /C=XX/CN=xxxxxxxxx Certification Authority VERIFY OK: depth=0, /C=XX/CN=user1 Negotiated ciphers: AES256-SHA SSLv3 Kx=RSA Au=RSA Enc=AES(256) Mac=SHA1 Connection closed: 1487 bytes sent to SSL, 96 bytes sent to socket From ipalette at hotmail.com Thu Apr 15 01:36:02 2004 From: ipalette at hotmail.com (Kevin McNally) Date: Thu Apr 15 01:44:47 2004 Subject: Gnupg-users Digest, Vol 7, Issue 19 Message-ID: It's sent over https:// When the body of the email is built the message is encrypted then emailed, nothing is displayed in the browser. The page that decodes the message is secured and only known by one person. Kevin >From: Neil Williams >To: gnupg-users@gnupg.org >Subject: Re: Gnupg-users Digest, Vol 7, Issue 19 >Date: Wed, 14 Apr 2004 22:12:36 +0100 > >On Wednesday 14 Apr 2004 9:31, Kevin McNally wrote: > > Thanks Pete, > > > > >>From the little information I have found it may be a permissions issue >on > > > > the server, but I can not be sure. Basically this is what happens: > > 1. Someone fills out a form on the website and it is emailed to someone >I > > created a key for. > > 2. The recipient gets an email with an encrypted message in the body. > > 3. Once the the email is received, she goes to a seperate webpage on the > > server to decrypt it. > > 4. She copies the encrypted body of the message into a text box on the >page > > and types the pass phrase into a sperate field. > > 5. The form data is formatted and dispalyed for her to use. > >At which point, all your effort in encrypting the transmitted data is lost! >Everything that the server displays in the browser is sent in plain text - >including the decrypted block!! > >Unless the block is decrypted locally, it will be pointless encrypting it >in >the first place! If you never encrypted it in the first place, the data >would >still be sent once in clear text, just like in your system. > >The second problem is that the PASSPHRASE is sent in clear text too, so now >the key is compromised too, or are you doing this over https:// ? > > > > > Does that makes sense? > >Umm, No. Sorry. > >-- > >Neil Williams >============= >http://www.codehelp.co.uk/ >http://www.dclug.org.uk/ >http://www.isbn.org.uk/ >http://sourceforge.net/projects/isbnsearch/ > >http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 ><< attach4 >> >_______________________________________________ >Gnupg-users mailing list >Gnupg-users@gnupg.org >http://lists.gnupg.org/mailman/listinfo/gnupg-users _________________________________________________________________ Is your PC infected? Get a FREE online computer virus scan from McAfee® Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 From marcos at fc.unesp.br Thu Apr 15 21:29:01 2004 From: marcos at fc.unesp.br (=?ISO-8859-1?Q?Marcos_Ant=F4nio_Cavenaghi?=) Date: Thu Apr 15 21:26:45 2004 Subject: GnuPG and OpenSSL Message-ID: <407EE27D.3030907@fc.unesp.br> Hello, I'm not sure if this is the apropriate list (should it be gnupg-devel?), but are the algorithms source codes (X.c) in GnuPG the same as the ones used in OpenSSL? thanks, Marcos From mctylr at privacy.nb.ca Fri Apr 16 00:00:14 2004 From: mctylr at privacy.nb.ca (M Taylor) Date: Thu Apr 15 23:57:41 2004 Subject: GnuPG and OpenSSL In-Reply-To: <407EE27D.3030907@fc.unesp.br>; from marcos@fc.unesp.br on Thu, Apr 15, 2004 at 04:29:01PM -0300 References: <407EE27D.3030907@fc.unesp.br> Message-ID: <20040415230014.A31848@pull.privacy.nb.ca> On Thu, Apr 15, 2004 at 04:29:01PM -0300, Marcos Ant?nio Cavenaghi wrote: > > are the algorithms source codes (X.c) in GnuPG the same as the ones used > in OpenSSL? If you mean, are the source code files (rsa.c) in GnuPG the same as OpenSSL? No. Different source code files, different licenses. (OpenSSL) (GnuPG) If you mean, are there same algorithms in both GnuPG and OpenSSL? Then yes, they both implement a number of common algorithms, and as well there is a fair bit of non-overlap (not all algorithms are implemented in both). E.g. both have RSA, DSA/DSS, AES, MD-5, SHA-1, and Triple-DES. From ian_bruce at fastmail.fm Fri Apr 16 04:33:02 2004 From: ian_bruce at fastmail.fm (Ian Bruce) Date: Fri Apr 16 04:28:14 2004 Subject: exporting secret keys Message-ID: <20040415193302.2cecf328.ian_bruce@fastmail.fm> I recently wanted to move an ElGamal/DSA keypair to another machine. I used the "--export-secret-keys" option to write a keyfile. After importing the key on the other machine, it appeared that only the secret key had been transferred, but not the public key. I have a few questions about this. -- under what circumstances would you want to transfer ONLY a secret key? Shouldn't the public key always go with it? Why doesn't GnuPG do this by default? -- in any case, isn't the public key easily computable from the secret key? Or is that only true for RSA? -- this seems like a fairly common thing to want to do. How is it supposed to be accomplished? Can you use the "--export" and "--export-secret-keys" options simultaneously? I couldn't find any mention of this in the documentation. Thanks for any explanation of what you're supposed to do in this situation. Ian Bruce ian_bruce@fastmail.fm From atom-gpg at suspicious.org Fri Apr 16 05:39:44 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Fri Apr 16 05:37:10 2004 Subject: exporting secret keys In-Reply-To: <20040415193302.2cecf328.ian_bruce@fastmail.fm> References: <20040415193302.2cecf328.ian_bruce@fastmail.fm> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > I recently wanted to move an ElGamal/DSA keypair to another machine. I > used the "--export-secret-keys" option to write a keyfile. After > importing the key on the other machine, it appeared that only the secret > key had been transferred, but not the public key. I have a few questions > about this. > > -- under what circumstances would you want to transfer ONLY a secret > key? Shouldn't the public key always go with it? Why doesn't GnuPG do > this by default? > > -- in any case, isn't the public key easily computable from the secret > key? Or is that only true for RSA? ======================= you can recover the public key from a private key using gpgsplit, but it's much easier to just export/import both keys. > -- this seems like a fairly common thing to want to do. How is it > supposed to be accomplished? Can you use the "--export" and > "--export-secret-keys" options simultaneously? I couldn't find any > mention of this in the documentation. ======================= i'm sure there are almost as many ways to do this as there are people on the list ;) here's ~one~ of the ways i'd do it: $ gpg --export-secret-key 0x12345678 > key-file $ gpg --export-key 0x12345678 >> key-file copy the "key-file" to the other machine, and run: $ gpg --import < key-file ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "TELEVISION IS DRUGS" -- Bumper Sticker -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkB/VYQACgkQnCgLvz19QePDXACfRLZXRPmI0rRT61HUtHLf8vWD K5wAoKp8jUkMPKAMll8OefTBytBjl/S+ =M/eO -----END PGP SIGNATURE----- From gnichols at tpg.com.au Fri Apr 16 08:59:23 2004 From: gnichols at tpg.com.au (Graeme Nichols) Date: Fri Apr 16 08:56:28 2004 Subject: Problems importing public key In-Reply-To: <200404142116.47755.linux@codehelp.co.uk> References: <1081923324.1671.17.camel@localhost.localdomain> <200404142116.47755.linux@codehelp.co.uk> Message-ID: <1082098767.1649.44.camel@localhost.localdomain> On Thu, 2004-04-15 at 06:16, Neil Williams wrote: > On Wednesday 14 Apr 2004 7:15, Graeme Nichols wrote: > > Hello Folks, > > > > I was sent a public key by a fellow so that I could encrypt a file to > > him. The name of the file, for what it is worth is: 0xF94BBB03.asc which > > You might be able to trust this file, but why should GnuPG? You've just > imported a public key that has no relation to your own key, there is nothing > for GnuPG to use to work out whether to trust the key. You comment that this > is for sensitive data yet you seem prepared to take the key at face value. > > > happens to be the DSA key ID of his key used to sign his emails. It > > So all you really know about this key is that the email address matches the > keyID. Is that enough? Can't be particularly sensitive data for encryption! > > > imports OK into my gnupg V1.2.3 but when I click on the lock icon in > > Evolution (so I can test the public key) I get the following error: > > > > gpg: armor header: Version: GnuPG v1.2.4 (GNU/Linux) > > gpg: Signature made Sat 03 Apr 2004 09:54:31 EST using DSA key ID > > F94BBB03 > > gpg: BAD signature from "Benoit Gr?goire (Serveur t?l?matique des > > Oops. You've got the right key but something is wrong with the email. > > > ?tudiants de Polytechnique) " > > gpg: textmode signature, digest algorithm SHA1 > > > > What have I done wrong? > > Nothing, necessarily. There may be something wrong with the signed email > though. > > > I have another public key that I imported from a keyserver that works OK > > If you import a new copy of the same key, it'll simply overwrite the old one. > The only differences that will make any odds here are things like extra > subkeys, extra signatures, extra UID's. If GnuPG accepts these two keys as > the same, the public key in both copies is the same. > > > if I click the lock icon on Evolution yet if I fire up seahorse and > > click on the Key Manager icon to list the keys I have under the 'trust' > > column both the public keys I have imported show 'ERROR'. Mine naturally > > shows 'ultimate' > > This is the separate trust issue. GnuPG cannot trust this key because you > haven't verified the key. You would need to follow the keysigning procedure > and then sign the key for GnuPG to be able to trust this key. > > > I really need to find out if I have done something wrong so I can > > No, just that there is something that you haven't yet done. > > > rectify the problem as Benoit is waiting for the file he needs me to > > send him and I am not going to send it if something is wrong as the file > > has some very private data in it and I don't want it to fall into the > > wrong hands. > > Then you MUST verify the key properly. Usually, this involves meeting > face-to-face to exchange GnuPG fingerprints and verify proof of photo ID AS > WELL as verifying that the email address in the key is the right destination. > > GnuPG cannot be expected to encrypt sensitive data if there no way of knowing > if you are encrypting to the right person. Thanks for the info Neil. I can encrypt the sensitive data file OK using the public key provided, but, as you say, I have no idea whether I can trust this key. I have never met the gentleman in question (who is as honest as the day is long, I'm sure) but I have no way of knowing if someone is impersonating him thus my concern when ALL his signed emails fail to authenticate properly with the key he provided. As I am no expert on this topic can you tell me how I could attempt to download his public key from a keyserver? The only public key I have downloaded so far had foolproof instructions in the form of a url in the comment field and then foolproof instructions when one connected to that url. It was the key for Dennis Patrick Lamb Jr., very cleverly done. -- Kind regards, Graeme Nichols ---------------------------------------------------------------------- Politics, as a practice, whatever its professions, has always been the systematic organisation of hatreds. -- Henry Adams, "The Education of Henry Adams" ---------------------------------------------------------------------- - IMPORTANT. - - The contents of this email and any attachments, which may be con- - - fidential, are sent for the personal attention of the addressee/s - - only. If you receive this email and are not the intended addressee - - please inform the sender and delete this email immediately. Use, - - copying, disclosure or forwarding of the contents of this email - - and/or any attachment/s is not authourised. - ---------------------------------------------------------------------- From marcos at cavenaghi.net Thu Apr 15 21:15:13 2004 From: marcos at cavenaghi.net (=?ISO-8859-1?Q?Marcos_Ant=F4nio_Cavenaghi?=) Date: Fri Apr 16 09:26:21 2004 Subject: GnuPG and OpenSSL Message-ID: <407EDF41.8030701@cavenaghi.net> Hello, I'm not sure if this is the apropriate list (should it be gnupg-devel?), but are the algorithms source codes (X.c) in GnuPG the same as the ones used in OpenSSL? thanks, Marcos From ekot at protek.ru Fri Apr 16 14:18:33 2004 From: ekot at protek.ru (Eugene Kotlyarov) Date: Fri Apr 16 14:17:20 2004 Subject: non ascii characters in key name Message-ID: <407FCF19.2000204@protek.ru> Hello I am using GnuPG v1.2.4 on Windows 2000. Some users send me keys where name contains non ascii characters. It is ok in PGP, but in GnuPG, when I use --list-keys I have something like this: pub 1024D/308A940E 2003-07-08 \xce\xc8\xd2\x20\xcf\xf0\xee\xf2\xe5\xea-33 Is it possible to view such key names in readable form? From hmujtaba at forumsys.com Fri Apr 16 19:14:10 2004 From: hmujtaba at forumsys.com (Hasnain Mujtaba) Date: Fri Apr 16 19:11:59 2004 Subject: Checking expiration dates Message-ID: <4DCE15B9C4E66F4CA967EBF64C53D64D67B321@bstn-exch1.forumsys.com> Hi, If I understood it correctly, V4 OpenPGP Master keys never expire and are only used for signing. So, if the subkey expires, does that render the master key useless as well? Or can the user continue to use the master key for signing and verifying? Thanks Hasnain. ---- The information contained in this electronic mail and any attached document is the confidential and proprietary business information of Forum Systems, Inc. It is intended solely for the addressed recipient listed above. It may not be distributed in any manner without the express written consent of Forum Systems, Inc. From jluehr at gmx.net Fri Apr 16 20:04:13 2004 From: jluehr at gmx.net (Jan =?utf-8?q?L=C3=BChr?=) Date: Fri Apr 16 20:01:48 2004 Subject: twofish keysize Message-ID: <200404162004.13385.jluehr@gmx.net> Greetings, refering to the source code, the twofish implementation is able to handle 128 and 256 Bit keys. But how do I specify the size? Keep smilig yanosz From dshaw at jabberwocky.com Fri Apr 16 20:19:11 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Fri Apr 16 20:16:54 2004 Subject: twofish keysize In-Reply-To: <200404162004.13385.jluehr@gmx.net> References: <200404162004.13385.jluehr@gmx.net> Message-ID: <20040416181910.GA26399@jabberwocky.com> On Fri, Apr 16, 2004 at 08:04:13PM +0200, Jan L?hr wrote: > Greetings, > > refering to the source code, the twofish implementation is able to > handle 128 and 256 Bit keys. But how do I specify the size? You cannot without modifying the code. While the actual code can handle 128 bit keys, that is not a valid key size in OpenPGP and is not used. David From linux at codehelp.co.uk Fri Apr 16 21:13:17 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Fri Apr 16 21:10:34 2004 Subject: Problems importing public key In-Reply-To: <1082098767.1649.44.camel@localhost.localdomain> References: <1081923324.1671.17.camel@localhost.localdomain> <200404142116.47755.linux@codehelp.co.uk> <1082098767.1649.44.camel@localhost.localdomain> Message-ID: <200404162013.24406.linux@codehelp.co.uk> On Friday 16 Apr 2004 7:59, Graeme Nichols wrote: > trust this key. I have never met the gentleman in question (who is as > honest as the day is long, I'm sure) but I have no way of knowing if > someone is impersonating him thus my concern when ALL his signed emails > fail to authenticate properly with the key he provided. If his signed emails fail to verify, it's probably not the key that's at fault. Can you view the same messages in mutt, Mozilla+enigmail or KMail? Dennis' message on Tuesday 2:56:10 (Re: Use of public key servers) validates fine with key 0xF53BA904, he uses PGP/MIME to sign, like this one. It should validate in Evolution. (The signature is valid, but the key's validity is unknown - so I can't trust Dennis' key either.) > url. It was the key for Dennis Patrick Lamb Jr., very cleverly done. All messages on this gnupg-users mailing list in the last few months have validated in KMail - have any of mine not validated in your copy of Evolution? -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040416/632df653/attachment.bin From jluehr at gmx.net Fri Apr 16 21:14:57 2004 From: jluehr at gmx.net (Jan =?iso-8859-1?q?L=FChr?=) Date: Fri Apr 16 21:12:31 2004 Subject: twofish keysize In-Reply-To: <20040416181910.GA26399@jabberwocky.com> References: <200404162004.13385.jluehr@gmx.net> <20040416181910.GA26399@jabberwocky.com> Message-ID: <200404162114.57964.jluehr@gmx.net> Greetings,... Am Freitag, 16. April 2004 20:19 schrieb David Shaw: > On Fri, Apr 16, 2004 at 08:04:13PM +0200, Jan L?hr wrote: > > Greetings, > > > > refering to the source code, the twofish implementation is able to > > handle 128 and 256 Bit keys. But how do I specify the size? > > You cannot without modifying the code. While the actual code can > handle 128 bit keys, that is not a valid key size in OpenPGP and is > not used. is there a (simple) patch available? I fact, I don't have time to dig into the sources, I just read the comment... Keep smiling yanoz From malsyned at cif.rochester.edu Fri Apr 16 22:12:03 2004 From: malsyned at cif.rochester.edu (Dennis Lambe Jr.) Date: Fri Apr 16 22:09:25 2004 Subject: Problems importing public key In-Reply-To: <200404162013.24406.linux@codehelp.co.uk> References: <1081923324.1671.17.camel@localhost.localdomain> <200404142116.47755.linux@codehelp.co.uk> <1082098767.1649.44.camel@localhost.localdomain> <200404162013.24406.linux@codehelp.co.uk> Message-ID: <1082146323.28053.3.camel@localhost> On Fri, 2004-04-16 at 15:13, Neil Williams wrote: > It should > validate in Evolution. It had better, Evolution generated it! --D P.S. What exactly are the issues with Evolution and PGP/MIME? People keep saying "it doesn't work quite right" but I've never had a problem, and nobody's ever said how it fails when it does, or why. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 279 bytes Desc: This is a digitally signed message part Url : /pipermail/attachments/20040416/f35b27fe/attachment.bin From linux at codehelp.co.uk Fri Apr 16 23:58:41 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Fri Apr 16 23:55:57 2004 Subject: Problems importing public key In-Reply-To: <1082146323.28053.3.camel@localhost> References: <1081923324.1671.17.camel@localhost.localdomain> <200404162013.24406.linux@codehelp.co.uk> <1082146323.28053.3.camel@localhost> Message-ID: <200404162258.43588.linux@codehelp.co.uk> On Friday 16 Apr 2004 9:12, Dennis Lambe Jr. wrote: > P.S. What exactly are the issues with Evolution and PGP/MIME? People I thought the problems were with Evolution and INLINE - Evolution developers apparently said that inline sigs would never be supported by Evolution. > keep saying "it doesn't work quite right" but I've never had a problem, > and nobody's ever said how it fails when it does, or why. If PGP/MIME doesn't quite work and inline is unsupported, it doesn't look good! Your signatures have always verified perfectly in KMail and Mutt, like almost everyone else on this list. I've only had a handful of bad signatures and each of these was down to me fiddling with the maildrop features etc. because others were able to verify the same message OK. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040416/60a165d5/attachment.bin From atom-gpg at suspicious.org Sat Apr 17 00:19:50 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 17 00:17:04 2004 Subject: update to ez-pine-gpg Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wed, 14 Apr 2004, Stewart V. Wright wrote: > * Atom 'Smasher' [040414 19:25]: > I've put a (temporary) screenshot up showing how mutt handles this > with your last message: > http://www.liv.ac.uk/~svwright/Screenshot-Gnome-terminal.png > The '[-- END PGP SIGNED MESSAGE --]' is a dead give away! :-) > > > i consider it a security feature that all any non-signed > > text is suppressed. > > Each to their own... I'm going to let others smack you about for this > position. ================================ thanks to Stewart for pointing this out. even though no one else presented any arguments supporting this behavior, it started to seem like more of a good idea, the more i thought about it. so, ez-pine-gpg v0.3g is now out and any text outside of the pgp headers is displayed in a way that will clearly show what text is signed and what text isn't. http://business-php.com/opensource/ez-pine-gpg/ ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "It is estimated that we spend $322,000 for each enemy we kill, while we spend in the so-called war on poverty in America only about $53 for each person classified as 'poor'. And much of that $53 goes for salaries of people who are not poor. We have escalated the war in Viet Nam and de-escalated the skirmish against poverty. It challenges the imagination to contemplate what lives we could transform if we were to cease killing." -- Martin Luther King, Jr. The Casualties of the War in Vietnam, 25 February 1967 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCAW70ACgkQnCgLvz19QeODdgCfWjFErW57VEW+c8nLz21IaFGq Y+UAn1ZSG0R2SBvRc35TUPNnLESbtwCV =D+gJ -----END PGP SIGNATURE----- From hmujtaba at forumsys.com Sat Apr 17 03:07:32 2004 From: hmujtaba at forumsys.com (Hasnain Mujtaba) Date: Sat Apr 17 03:05:18 2004 Subject: encrypting files in a script Message-ID: <4DCE15B9C4E66F4CA967EBF64C53D64D1902C6@bstn-exch1.forumsys.com> Hi, I am encrypting some files using a shell script. GPG asks the question: "It is NOT certain that the key belongs to the person named in the user ID. If you *really* know what you are doing, you may answer the next question with yes" How can I supress this question so the script can run through the end? Thanks Hasnain. ---- The information contained in this electronic mail and any attached document is the confidential and proprietary business information of Forum Systems, Inc. It is intended solely for the addressed recipient listed above. It may not be distributed in any manner without the express written consent of Forum Systems, Inc. From cwsiv at keepandbeararms.com Sat Apr 17 03:55:04 2004 From: cwsiv at keepandbeararms.com (Carl William Spitzer IV) Date: Sat Apr 17 03:34:09 2004 Subject: OT discussions In-Reply-To: <20040414130024.U8942@qbhto.arg> References: <20040413114024.GC4400@jabberwocky.com> <87vfk3r3z0.fsf@vigenere.g10code.de> <20040413222626.GC1422@susie.penguin.de> <20040414130024.U8942@qbhto.arg> Message-ID: <1082140855.2804.4.camel@linux.local> For those wishing to discuss issues of Open Source vs Redmond and the politics of same I suggest the Suse OT list hosted by them. www.suse.com/lists Even windows people can join and learn. Enjoy :) CWSIV From dshaw at jabberwocky.com Sat Apr 17 04:19:29 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Sat Apr 17 04:16:44 2004 Subject: twofish keysize In-Reply-To: <200404162114.57964.jluehr@gmx.net> References: <200404162004.13385.jluehr@gmx.net> <20040416181910.GA26399@jabberwocky.com> <200404162114.57964.jluehr@gmx.net> Message-ID: <20040417021929.GC30100@jabberwocky.com> On Fri, Apr 16, 2004 at 09:14:57PM +0200, Jan L?hr wrote: > Greetings,... > > Am Freitag, 16. April 2004 20:19 schrieb David Shaw: > > On Fri, Apr 16, 2004 at 08:04:13PM +0200, Jan L?hr wrote: > > > Greetings, > > > > > > refering to the source code, the twofish implementation is able to > > > handle 128 and 256 Bit keys. But how do I specify the size? > > > > You cannot without modifying the code. While the actual code can > > handle 128 bit keys, that is not a valid key size in OpenPGP and is > > not used. > > is there a (simple) patch available? I fact, I don't have time to > dig into the sources, I just read the comment... There isn't one available that I know of. Why do you want to use 128 bit keys? The default is the larger 256 bits. David -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 330 bytes Desc: not available Url : /pipermail/attachments/20040416/ee1d0b88/attachment.bin From atom-gpg at suspicious.org Sat Apr 17 04:56:34 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sat Apr 17 04:53:50 2004 Subject: encrypting files in a script In-Reply-To: <4DCE15B9C4E66F4CA967EBF64C53D64D1902C6@bstn-exch1.forumsys.com> References: <4DCE15B9C4E66F4CA967EBF64C53D64D1902C6@bstn-exch1.forumsys.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 16 Apr 2004, Hasnain Mujtaba wrote: > I am encrypting some files using a shell script. GPG asks the question: > > "It is NOT certain that the key belongs to the person named > in the user ID. If you *really* know what you are doing, > you may answer the next question with yes" > > How can I supress this question so the script can run through the end? =============================== this will do the trick: --trust-model always alternatively, you could mark the key as trusted. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Only when the last tree has died, And the last river poisoned, And the last fish been caught, Will the white man realize that he cannot eat money" --19th Century Cree Indian Proverb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCAnOcACgkQnCgLvz19QeN0AgCdE5/NJHByHxhJ6vYf0b6zaakC jR8AnArTpTX7DuW4lUQzgs3oD1dg57h0 =C1HR -----END PGP SIGNATURE----- From pt at radvis.nu Sat Apr 17 09:45:32 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Sat Apr 17 09:42:28 2004 Subject: twofish keysize In-Reply-To: <200404162114.57964.jluehr@gmx.net> References: <200404162004.13385.jluehr@gmx.net> <20040416181910.GA26399@jabberwocky.com> <200404162114.57964.jluehr@gmx.net> Message-ID: <6.0.3.0.2.20040417094113.027e8ab8@localhost> At 21:14 2004-04-16, you wrote: >Greetings,... > >Am Freitag, 16. April 2004 20:19 schrieb David Shaw: >> On Fri, Apr 16, 2004 at 08:04:13PM +0200, Jan L?hr wrote: >> > Greetings, >> > >> > refering to the source code, the twofish implementation is able to >> > handle 128 and 256 Bit keys. But how do I specify the size? >> >> You cannot without modifying the code. While the actual code can >> handle 128 bit keys, that is not a valid key size in OpenPGP and is >> not used. > >is there a (simple) patch available? I fact, I don't have time to dig into the >sources, I just read the comment... > >Keep smiling >yanoz > Don't change or patch the code in that way! You'd better not use any encryption that isn't OpenPGP compliant and further not compatible with GnuPG. If you do, you cannot expect anyone to be able to decrypt your files/messsages. Besides, there isn't much need for a weaker key anyway. So, stick to the default settings in GnuPG! Per Tunedal From linux at codehelp.co.uk Sun Apr 18 00:49:10 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Sun Apr 18 00:46:32 2004 Subject: gpgme-smime.so not initialising in KMail 1.6.1 Message-ID: <200404172349.20005.linux@codehelp.co.uk> I've got openssl and accessory packages installed and working on Debian sid (unstable) to allow other email clients to verify Thawte style messages, but II can't get gpgme-smime.so to initialise. The message is signed, but the validity of the signature can't be verified. Reason: Crypto plug-in "/usr/lib/cryptplug/gpgme-smime.so" is not initialised. what might I be missing? I had to create .gnupg/gpgme.conf as it didn't exist before. http://www.gnupg.org/aegypten/development.en.html It now only contains: agent-program /usr/bin/gpg-agent It doesn't seem like a KMail config problem, although it's only KMail that I've set to use this plugin. Should I transfer this query to the KMail lists? -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040417/9a04e445/attachment.bin From bigm1980 at yahoo.de Fri Apr 16 10:50:08 2004 From: bigm1980 at yahoo.de (=?iso-8859-1?q?martin=20stojko?=) Date: Sun Apr 18 18:59:09 2004 Subject: gnuPG: no valid OpenPGP data found Message-ID: <20040416085008.89564.qmail@web25207.mail.ukl.yahoo.com> hi, i use gnupg to generate a key and send this key to a local keyserver. this work. then i delete the key from the local keyring. after that i use: gpg --keyserver --recv-keys but the answer is: gpg: no valid OpenPGP data found. gpg: Total number processed: 0 the same result when i use --search-keys, gpg show the keys and wait that i enter the number of a key, but after this i get the same error message. with "pgp keys" from www.pgp.com it works without problems. have somebody a solution for my problem. sorry for my bad english but i hope you can understand what i mean. thx Mit sch?nen Gr??en von Yahoo! Mail - http://mail.yahoo.de From linux at codehelp.co.uk Sun Apr 18 20:36:36 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Sun Apr 18 20:33:08 2004 Subject: gnuPG: no valid OpenPGP data found In-Reply-To: <20040416085008.89564.qmail@web25207.mail.ukl.yahoo.com> References: <20040416085008.89564.qmail@web25207.mail.ukl.yahoo.com> Message-ID: <200404181936.43479.linux@codehelp.co.uk> On Friday 16 April 2004 9:50, martin stojko wrote: > hi, > > i use gnupg to generate a key and send this key to a > local keyserver. this work. You could check that the keyserver has actually accepted the key by using a web interface - that would prevent you losing the key. What do you mean by local keyserver - does it synchronise with the other keyservers? Doesn't it have a web interface? > then i delete the key from the local keyring. why? You've probably now lost the secret key too. Yet another useless key on the keyservers that cannot be deleted or revoked! > after that i use: > gpg --keyserver --recv-keys This would only import the public key anyway - if you've generated this key, people searching the keyserver would be justified in hoping that you had the secret key for it! > with "pgp keys" from www.pgp.com it works without > problems. So have you got the secret key stored somewhere because any keyserver will only ever return the public key. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040418/299c34ad/attachment.bin From jharris at widomaker.com Mon Apr 19 00:41:26 2004 From: jharris at widomaker.com (Jason Harris) Date: Mon Apr 19 00:39:02 2004 Subject: new (2004-04-18) keyanalyze results (+sigcheck) Message-ID: <20040418224126.GO10980@pm1.ric-41.lft.widomaker.com> New keyanalyze results are available at: http://keyserver.kjsl.com/~jharris/ka/2004-04-18/ Signatures are now being checked using keyanalyze+sigcheck: http://dtype.org/~aaronl/ but preprocess.keys.no-sigcheck.bz2 is available with today's report to facilitate running a non-sigcheck analysis, for comparison. Earlier reports are also available, for comparison: http://keyserver.kjsl.com/~jharris/ka/ Even earlier monthly reports are at: http://dtype.org/keyanalyze/ SHA-1 hashes and sizes for all the "permanent" files: 8d5108f4353256313dbbf8520a1bb98dac4a9847 9459270 preprocess.keys 4b14c75c48d9adf66e5a77007f440c6c2a676e5b 13731606 preprocess.keys.no-sigcheck 843e73d45ac50e907a327faf87f5c86b7c1f0ef7 6481175 othersets.txt 2dce43671bc21352e7a6e13d58ef28c8c1072737 2420494 msd-sorted.txt b0f152cbac2bff77aeed70a933fec6d7ac3e7b71 1484 index.html e6b8bb2dcac213b742bca8ac6e2fb52adf823fef 2289 keyring_stats 2e4671ee97c08acf26ed0e1a67027c8b0432b361 955639 msd-sorted.txt.bz2 97e5cc7a7fa12d2eda9176c345bdb8137f7e5224 26 other.txt 4b000064a56d524253db30e9b17f2fc772de6575 1380246 othersets.txt.bz2 79602d6bb1020c966816285414627daff303558b 3874213 preprocess.keys.bz2 ed6c52e69e508e66b76a38b70ca142ba0eb3b24f 6138075 preprocess.keys.no-sigcheck.bz2 e7e76dbc1b1e6e3e750c2bf56e4770cd6d3f4fe3 9276 status.txt 4d37419b6519643c089581d2025896abc9afca02 211999 top1000table.html 3af55d7c254e182ee90b7bcfe7d97d016eb127fe 30555 top1000table.html.gz feca6d7252e45672c32767ab3e275f1fb07b6865 11037 top50table.html 7281c645978e9574dd0d0b66589f637eae509e12 2234 D3/D39DA0E3 -- Jason Harris | NIC: JH329, PGP: This _is_ PGP-signed, isn't it? jharris@widomaker.com _|_ web: http://keyserver.kjsl.com/~jharris/ Got photons? (TM), (C) 2004 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: not available Url : /pipermail/attachments/20040418/805ba9c6/attachment.bin From malte.gell at gmx.de Mon Apr 19 00:19:58 2004 From: malte.gell at gmx.de (Malte Gell) Date: Mon Apr 19 00:48:18 2004 Subject: twofish keysize In-Reply-To: <20040417021929.GC30100@jabberwocky.com> References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> Message-ID: <200404190020.15022.malte.gell@gmx.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > There isn't one available that I know of. Why do you want to use 128 > bit keys? The default is the larger 256 bits. Maybe he expects a performance advantage when using 128 bit keys instead of 256 bit. By the way, are there estimations how much slower a 256 bit Twofish key is? I guess one will only notice it whith huge amount of data. Malte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAgv8aGzg12gD8wBYRAkc0AJ9nR6QVq2PWlBBh+3jGoJs5nu6//QCgjL19 1o5W9Xyc6IFvZcfgOn6Ptvk= =0W7f -----END PGP SIGNATURE----- From jharris at widomaker.com Mon Apr 19 05:30:23 2004 From: jharris at widomaker.com (Jason Harris) Date: Mon Apr 19 05:28:04 2004 Subject: [Biglumber-news] new (2004-04-18) keyanalyze results (+sigcheck) In-Reply-To: <20040419005645.GR18329@parcelfarce.linux.theplanet.co.uk> References: <20040418224126.GO10980@pm1.ric-41.lft.widomaker.com> <20040419005645.GR18329@parcelfarce.linux.theplanet.co.uk> Message-ID: <20040419033023.GR10980@pm1.ric-41.lft.widomaker.com> On Mon, Apr 19, 2004 at 01:56:45AM +0100, Matthew Wilcox wrote: > On Sun, Apr 18, 2004 at 06:41:26PM -0400, Jason Harris wrote: > > New keyanalyze results are available at: > > > > http://keyserver.kjsl.com/~jharris/ka/2004-04-18/ > > > > Signatures are now being checked using keyanalyze+sigcheck: > > > > http://dtype.org/~aaronl/ > > > > but preprocess.keys.no-sigcheck.bz2 is available with today's report > > to facilitate running a non-sigcheck analysis, for comparison. > > Seems a bit broken to me. The following keys are no longer listed in > the report: > > 191FCD8A Kyle McMartin NB: Everyone, please followup only to keyanalyze-discuss and/or follow this thread there: http://www.dtype.org/pipermail/keyanalyze-discuss/ Reply-To: has been set. Thanks. Some of the selfsigs expired this key on 2003-03-18, but newer selfsigs don't expire the key. > 4975F947 Curtis Ireland (Personal Key) 2003-07-04, then removed. > E2B11082 Jody McIntyre 2003-11-27 for the selfsig and many of the cert. sigs. GPG won't encrypt to it. > 7D2ED788 Richard Guy Briggs (gpg-lap) 2001-08-30, then removed. > 12F506C8 Peter Sjoberg 2003-04-27, then removed. > 1B9200FC Bruce R. Miller (personal use) The userid with all the external signatures was revoked. > 2D9179B0 Scott Moynes 2002-01-21 and 2004-02-01, both of which have passed. GPG won't encrypt to it. > ... and that's just from the set of 79 keys in my LUG. I can probably > find other no-long-appearing keys if necessary. I've signed many of > these keys myself, so I know them to be valid. By design, expired keys aren't considered. But, it looks like older selfsigs with expiration dates aren't being overridden by newer selfsigs, which is a definite error. Aaron? -- Jason Harris | NIC: JH329, PGP: This _is_ PGP-signed, isn't it? jharris@widomaker.com _|_ web: http://keyserver.kjsl.com/~jharris/ Got photons? (TM), (C) 2004 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: not available Url : /pipermail/attachments/20040418/ab626100/attachment-0001.bin From nickjohnson at virginia.edu Mon Apr 19 04:51:07 2004 From: nickjohnson at virginia.edu (Nicholas Paul Johnson) Date: Mon Apr 19 05:53:31 2004 Subject: General Crypto Question Message-ID: Hey, Not directly related to gpg, but I think some of you may enjoy answering this. I'm reading thru _Applied Cryptography_, and I want to make sure I understand it: Suppose two people, A and B, know a shared secret S; for simplicity, pretend it is a long string of bits. Suppose A wants to send a message M (another long string of bits) to B, and A wants to sign it. Would it be sufficient and secure to construct a signature like this: Signature = One-Way-Hash-Function( Concatenate(M, S) ) In other words, one creates a new string of bits which is the concatenation of the shared secret and the message, and then hashes that new string (assume perfect hash function) to yield the signature. A then sends both M and the signature to B. B could then verify the signature by generating the signature from the M and S in the same manner as above, and then comparing that signature to the one sent by A. As the one-way hash function is "perfect", nobody could determine S if he has a (message, signature) pair. Am I thinking this out right, or am I missing something critical? Thanks, -- Nicholas Paul Johnson | Pianos are cool, | nickjohnson \at\ virginia \dot\ edu | o _______________ | 3ebf10a7 subkeys.pgp.net | /\_ _| | | http://manjac.ath.cx/nick | _\__`[_______________| | "When all you've got is a hammer, | ] [ \, ][ ][ | everything looks like a nail." | \___ . . . | | Ping pong is cooler. "The ways of God are impenetrable... | o_, . __o The ways of LINUX are non understandable... | | | /| Marc Breton | -- From wk at gnupg.org Mon Apr 19 09:26:13 2004 From: wk at gnupg.org (Werner Koch) Date: Mon Apr 19 09:11:19 2004 Subject: twofish keysize In-Reply-To: <200404190020.15022.malte.gell@gmx.de> (Malte Gell's message of "Mon, 19 Apr 2004 00:19:58 +0200") References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> <200404190020.15022.malte.gell@gmx.de> Message-ID: <87fzb0h1ii.fsf@vigenere.g10code.de> On Mon, 19 Apr 2004 00:19:58 +0200, Malte Gell said: > Maybe he expects a performance advantage when using 128 bit keys instead > of 256 bit. By the way, are there estimations how much slower a 256 bit > Twofish key is? I guess one will only notice it whith huge amount of here are encrypt/decrypt result from libgcrypt. ECB CBC CFB CTR --------------- --------------- --------------- ---------------- 3DES 140ms 150ms 170ms 190ms 150ms 160ms 180ms 180ms CAST5 50ms 50ms 70ms 80ms 60ms 60ms 90ms 80ms BLOWFISH 90ms 100ms 110ms 120ms 110ms 110ms 130ms 130ms AES 40ms 40ms 50ms 60ms 40ms 50ms 80ms 80ms AES192 40ms 40ms 60ms 60ms 50ms 50ms 80ms 80ms AES256 40ms 50ms 60ms 70ms 50ms 60ms 90ms 80ms TWOFISH 50ms 50ms 60ms 60ms 60ms 60ms 100ms 100ms TWOFISH128 50ms 50ms 60ms 60ms 60ms 60ms 100ms 100ms As you can see, there is no difference between twofish (256 bit) and twofish 128. The disadvantage of a 256 bit key is that is requires double as much entropy to create a session key. Werner From avbidder at fortytwo.ch Mon Apr 19 09:48:37 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Mon Apr 19 09:45:49 2004 Subject: Checking expiration dates In-Reply-To: <4DCE15B9C4E66F4CA967EBF64C53D64D67B321@bstn-exch1.forumsys.com> References: <4DCE15B9C4E66F4CA967EBF64C53D64D67B321@bstn-exch1.forumsys.com> Message-ID: <200404190948.41757@fortytwo.ch> On Friday 16 April 2004 19.14, Hasnain Mujtaba wrote: > Hi, > > If I understood it correctly, V4 OpenPGP Master keys never expire and > are only used for signing. Sorry, you did not undestand entirely correctly. You *may* use v4 OpenPGP keys in that way, but you don't have to. By default, gnupg creates a primary key (what you call master key) which is used for signing (keys and data both) and a subkey for encryption. The change wrt expiry is that v4 does not have an expiration date in the key, but only in the self signature. So a key expires when all userids have expired. By renewing the self-signatures on the userids you can change the expiration date of the userid (and, hence, the key) - IIRC all implementations should only look at the newest self-signature of the userid under consideration when verifying a key. > So, if the subkey expires, does that > render the master key useless as well? Or can the user continue to > use the master key for signing and verifying? The primary key is completely independent from its subkeys. You can add subkeys to a primary, and you can have subkeys expire, and you can revoke subkeys, all this doesn't affect the primary key. Validity of the primary is determined by the validity of the userids - as long as there is a valid userid, the key can be considered valid. You could even have a window of time where the key was not valid at all, but it would become valid again when a valid selfsignature is added to a userid, or a valid userid is added to the key. I'm not entirely certain of all facts, for details you'll have to figure out exactly which self-signatures are interpreted by the various OpenPGP implementations on key verification. cheers -- vbi -- Today is Prickle-Prickle, the 36th day of Discord in the YOLD 3170 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040419/efbca9d6/attachment.bin From rmalayter at bai.org Mon Apr 19 14:12:29 2004 From: rmalayter at bai.org (Ryan Malayter) Date: Mon Apr 19 14:10:09 2004 Subject: General Crypto Question Message-ID: <792DE28E91F6EA42B4663AE761C41C2A021C90A1@cliff.bai.org> [Nicholas Paul Johnson] ... >B could then verify the signature by generating the signature >from the M and S in the same manner as above, and then >comparing that signature to the one sent by A. > >As the one-way hash function is "perfect", nobody could >determine S if he has a (message, signature) pair. > >Am I thinking this out right, or am I missing something critical? You have it basically right. In fact, this is the way most secure password authentication systems work. Both the user and the server have a shared secret, which is the password. The server sends a random message over the wire to the client, which then concatenates the secret password, and hashes the result. This hash is then sent over the wire back to the server. Presuming the hash function is secure, an eavesdropper cannot figure out the password, nor can they "replay" the attack, since the random message used only once. Sometimes the details change; in most modern systems only a hash of the user password is stored on the server, so an "extra" hashing step is required before concatenation at the client. Also, in systems like Kerberos, a timestamp is part of the hashed authentication string, so that a particular authentication session is "valid" for only a certain amount of time. But the basic underlying idea of authentication as the validation of a "signature" of a random message as you describe remains. Of course, finding a secure way to share the secret (password) in the first place can be difficult. But _Applied Cryptography_ illustrates several protocols for this as well. Regards, Ryan Malayter Bank Administration Institute Chicago, Illinois, USA PGP Key: http://www.malayter.com/pgp-public.txt ========================= All problems can be solved by diplomacy, but violence and treachery are equally effective, and more fun. From dasymond at it.usyd.edu.au Tue Apr 20 01:47:57 2004 From: dasymond at it.usyd.edu.au (Dave Symonds) Date: Tue Apr 20 01:45:10 2004 Subject: Remote signing Message-ID: <20040419234757.GA23117@it.usyd.edu.au> Hi All, I have a slightly unusual setup that I would like to use GnuPG in, and wanted to ask for some guidance. At my Uni we have a Sun machine that runs all the mail stuff, and where I mostly prefer to do all my mail reading/composing from. However, I don't want to entrust my GPG private key(s) to that system, and would prefer to keep them on my laptop (or a USB key). What my ideal setup would be is for my mailer (mutt) running on the mail server to call out to a little script that would connect securely (via ssh) to my laptop, on which would pop up a window showing the message and prompting for the passphrase to sign that message (encryption isn't so important at the moment). The signed message would be sent back, and then emailed out. This seems reasonably secure to me, and shouldn't be too difficult for me to code up. Does this sound like a reasonable approach? Has it been done before? Am I completely daft? Thanks, Dave. -- David Symonds USyd::SITRG::PhD Student http://www.it.usyd.edu.au/~dasymond/ From boldyrev+nospam at cgitftp.uiggm.nsc.ru Tue Apr 20 06:17:10 2004 From: boldyrev+nospam at cgitftp.uiggm.nsc.ru (Ivan Boldyrev) Date: Tue Apr 20 06:18:26 2004 Subject: Why GPG doesn't use filename inside encrypted file? Message-ID: I encrypted some-file. some-file.gpg was produced. $ gpg -e some-file When I list packets I see: :literal data packet: mode b, created 1082433628, name="some-file", raw data: 9371 bytes I renamed some-file.gpg to data.gpg. Now when I decrypt data.gpg with --decrypt-file or without options output file goes to ./data. (I do not use --decrypt because output goes to stdout if I do not specify output file explicitely). How can I decrypt file so that original name of file is restored automatically? [ib@ ~]$ gpg --version gpg (GnuPG) 1.3.5 ... -- Ivan Boldyrev Outlook has performed an illegal operation and will be shut down. If the problem persists, contact the program vendor. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 480 bytes Desc: not available Url : /pipermail/attachments/20040420/256e8cf7/attachment.bin From Freedom_Lover at pobox.com Tue Apr 20 06:30:55 2004 From: Freedom_Lover at pobox.com (Todd) Date: Tue Apr 20 06:28:38 2004 Subject: Why GPG doesn't use filename inside encrypted file? In-Reply-To: References: Message-ID: <20040420043055.GC5727@psilocybe.teonanacatl.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ivan Boldyrev wrote: > I renamed some-file.gpg to data.gpg. Now when I decrypt data.gpg with > --decrypt-file or without options output file goes to ./data. (I do > not use --decrypt because output goes to stdout if I do not specify > output file explicitely). > > How can I decrypt file so that original name of file is restored > automatically? Does --use-embedded-filename do what you want? - -- Todd OpenPGP -> KeyID: 0xD654075A | URL: www.pobox.com/~tmz/pgp ====================================================================== Hell hath no fury like a bureaucrat scorned. -- Dr. Milton Friedman -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: When crypto is outlawed bayl bhgynjf jvyy unir cevinpl. iD8DBQFAhKd/uv+09NZUB1oRAsZFAKDjfcxGtkaqq4TIa7qiPBK7xkJvVQCeNN9G 0D40MKssYDNv2281y24J9Wo= =vyO/ -----END PGP SIGNATURE----- From willy at debian.org Mon Apr 19 02:56:45 2004 From: willy at debian.org (Matthew Wilcox) Date: Tue Apr 20 06:34:31 2004 Subject: [Biglumber-news] new (2004-04-18) keyanalyze results (+sigcheck) In-Reply-To: <20040418224126.GO10980@pm1.ric-41.lft.widomaker.com> References: <20040418224126.GO10980@pm1.ric-41.lft.widomaker.com> Message-ID: <20040419005645.GR18329@parcelfarce.linux.theplanet.co.uk> On Sun, Apr 18, 2004 at 06:41:26PM -0400, Jason Harris wrote: > New keyanalyze results are available at: > > http://keyserver.kjsl.com/~jharris/ka/2004-04-18/ > > Signatures are now being checked using keyanalyze+sigcheck: > > http://dtype.org/~aaronl/ > > but preprocess.keys.no-sigcheck.bz2 is available with today's report > to facilitate running a non-sigcheck analysis, for comparison. Seems a bit broken to me. The following keys are no longer listed in the report: 191FCD8A Kyle McMartin 4975F947 Curtis Ireland (Personal Key) E2B11082 Jody McIntyre 7D2ED788 Richard Guy Briggs (gpg-lap) 12F506C8 Peter Sjoberg 1B9200FC Bruce R. Miller (personal use) 2D9179B0 Scott Moynes ... and that's just from the set of 79 keys in my LUG. I can probably find other no-long-appearing keys if necessary. I've signed many of these keys myself, so I know them to be valid. -- "Next the statesmen will invent cheap lies, putting the blame upon the nation that is attacked, and every man will be glad of those conscience-soothing falsities, and will diligently study them, and refuse to examine any refutations of them; and thus he will by and by convince himself that the war is just, and will thank God for the better sleep he enjoys after this process of grotesque self-deception." -- Mark Twain From stuart.yeates at computing-services.oxford.ac.uk Tue Apr 20 09:55:10 2004 From: stuart.yeates at computing-services.oxford.ac.uk (Stuart A Yeates) Date: Tue Apr 20 09:54:13 2004 Subject: Remote signing In-Reply-To: <20040419234757.GA23117@it.usyd.edu.au> References: <20040419234757.GA23117@it.usyd.edu.au> Message-ID: <4084D75E.8070503@computing-services.oxford.ac.uk> Dave Symonds wrote: > I have a slightly unusual setup that I would like to use GnuPG in, and wanted > to ask for some guidance. At my Uni we have a Sun machine that runs all the > mail stuff, and where I mostly prefer to do all my mail reading/composing from. > However, I don't want to entrust my GPG private key(s) to that system, and > would prefer to keep them on my laptop (or a USB key). What my ideal setup > would be is for my mailer (mutt) running on the mail server to call out to > a little script that would connect securely (via ssh) to my laptop, on which > would pop up a window showing the message and prompting for the passphrase to > sign that message (encryption isn't so important at the moment). The signed > message would be sent back, and then emailed out. If the Sun is hacked, an attacker can (potentially) see every password you type. You almost certainly want two passwords, first for the ssh connection and second for the gpg. What I've seen other people do is use two keys one for home and one for work/school. It seems to work pretty well, you can sign them with each other and take both to keysegning parties. cheers stuart -- Stuart Yeates stuart.yeates@computing-services.oxford.ac.uk OSS Watch http://www.oss-watch.ac.uk/ Oxford Text Archive http://ota.ahds.ac.uk/ Humbul Humanities Hub http://www.humbul.ac.uk/ From samuel at Update.UU.SE Tue Apr 20 10:41:11 2004 From: samuel at Update.UU.SE (Samuel ]slund) Date: Tue Apr 20 10:38:48 2004 Subject: Remote signing In-Reply-To: <4084D75E.8070503@computing-services.oxford.ac.uk> References: <20040419234757.GA23117@it.usyd.edu.au> <4084D75E.8070503@computing-services.oxford.ac.uk> Message-ID: <20040420084111.GA28365@Update.UU.SE> On Tue, Apr 20, 2004 at 08:55:10AM +0100, Stuart A Yeates wrote: > Dave Symonds wrote: > >I have a slightly unusual setup that I would like to use GnuPG in, and > >wanted > >to ask for some guidance. At my Uni we have a Sun machine that runs all the > >mail stuff, and where I mostly prefer to do all my mail reading/composing > >from. > >However, I don't want to entrust my GPG private key(s) to that system, and > >would prefer to keep them on my laptop (or a USB key). What my ideal setup > >would be is for my mailer (mutt) running on the mail server to call out to > >a little script that would connect securely (via ssh) to my laptop, on > >which > >would pop up a window showing the message and prompting for the passphrase > >to > >sign that message (encryption isn't so important at the moment). The signed > >message would be sent back, and then emailed out. > > If the Sun is hacked, an attacker can (potentially) see every password > you type. I do not see this, could you elaborate? Any password/secret for the ssh connection is of course lost but how would a localy running script that only have two text streams as connection to the compromised machine be threatened? Assuming the signing is done localy and the recieving ssh session only takes the message and no commands as input from the mail server. I would worry a little about the possibility of missing a change in the text shown before signing. Since you wrote the text you "know" what is writen in it and might not read it carfully enough to catch a change. If the mail server is compromised such a change is possible. //Samuel From stuart.yeates at computing-services.oxford.ac.uk Tue Apr 20 11:56:04 2004 From: stuart.yeates at computing-services.oxford.ac.uk (Stuart A Yeates) Date: Tue Apr 20 11:55:03 2004 Subject: Remote signing In-Reply-To: <20040420084111.GA28365@Update.UU.SE> References: <20040419234757.GA23117@it.usyd.edu.au> <4084D75E.8070503@computing-services.oxford.ac.uk> <20040420084111.GA28365@Update.UU.SE> Message-ID: <4084F3B4.2030507@computing-services.oxford.ac.uk> Samuel ]slund wrote: > On Tue, Apr 20, 2004 at 08:55:10AM +0100, Stuart A Yeates wrote: > >>Dave Symonds wrote: >> >>>I have a slightly unusual setup that I would like to use GnuPG in, and >>>wanted >>>to ask for some guidance. At my Uni we have a Sun machine that runs all the >>>mail stuff, and where I mostly prefer to do all my mail reading/composing >>>from. >>>However, I don't want to entrust my GPG private key(s) to that system, and >>>would prefer to keep them on my laptop (or a USB key). What my ideal setup >>>would be is for my mailer (mutt) running on the mail server to call out to >>>a little script that would connect securely (via ssh) to my laptop, on >>>which >>>would pop up a window showing the message and prompting for the passphrase >>>to >>>sign that message (encryption isn't so important at the moment). The signed >>>message would be sent back, and then emailed out. >> >>If the Sun is hacked, an attacker can (potentially) see every password >>you type. > > > I do not see this, could you elaborate? > Any password/secret for the ssh connection is of course lost but how > would a localy running script that only have two text streams as > connection to the compromised machine be threatened? > Assuming the signing is done localy and the recieving ssh session only > takes the message and no commands as input from the mail server. > > I would worry a little about the possibility of missing a change in the > text shown before signing. Since you wrote the text you "know" what is > writen in it and might not read it carfully enough to catch a change. > If the mail server is compromised such a change is possible. Let me see if I've understood you correctly: You have a Sun which does all your you normal mail handling, and at whose console you do do your computing. You have another laptop connected to the Sun via a public network which has your secret key. When you wish to use your public keys you use ssh to start a bash shell (or similar) on the laptop, transfer your data, perform your operation and transfer your data back. Possibly you have a script to automate some of these connection/transfer/opration/transfer steps. If the Sun is compromised, then your ssh connection and password is compromised. If your ssh password is compromised then an attacker can use it to connect to the laptop and get your secret key. If the secret key has no passphrase, then the attacker has all they need. If you have a passphrase on your secret key, then to perform the operation the passphrase must come from somewhere. If it comes across the ssh connection and the ssh connection has been compromised then your passphrase can be known to the attacker. If it comes off a disk local to the laptop and the attacker has you ssh password, then the attacker can connect to the laptop and collect the passphrase. If the attacker has access to the sun to install a keylogger (phyicsal or software) they can access any passwords and phrases you type at the keyboard. Another approach is to have a restricted account whose login shell isn't bash but a script which allows only certain operations to be performed. If you're very good at writing secure code this might be an option. I'm not sure you could use any out-of-the-box security scripting solutions because most of them prioritise the integrety of the system over the protection of some secret which the script has access to anyway. cheers stuart -- Stuart Yeates stuart.yeates@computing-services.oxford.ac.uk OSS Watch http://www.oss-watch.ac.uk/ Oxford Text Archive http://ota.ahds.ac.uk/ Humbul Humanities Hub http://www.humbul.ac.uk/ From dasymond at it.usyd.edu.au Tue Apr 20 14:10:05 2004 From: dasymond at it.usyd.edu.au (Dave Symonds) Date: Tue Apr 20 14:07:21 2004 Subject: Remote signing In-Reply-To: <4084F3B4.2030507@computing-services.oxford.ac.uk> References: <20040419234757.GA23117@it.usyd.edu.au> <4084D75E.8070503@computing-services.oxford.ac.uk> <20040420084111.GA28365@Update.UU.SE> <4084F3B4.2030507@computing-services.oxford.ac.uk> Message-ID: <20040420121005.GB10970@it.usyd.edu.au> On Tue, Apr 20, 2004 at 10:56:04AM +0100, Stuart A Yeates wrote: > Let me see if I've understood you correctly: I think you have it slightly backwards... > You have a Sun which does all your you normal mail handling, and at > whose console you do do your computing. You have another laptop > connected to the Sun via a public network which has your secret key. > When you wish to use your public keys you use ssh to start a bash shell > (or similar) on the laptop, transfer your data, perform your operation > and transfer your data back. Possibly you have a script to automate some > of these connection/transfer/opration/transfer steps. The Sun does all the normal mail handling, but there is no console that I use to access it directly (only via ssh). The ssh connection from my laptop to the Sun is the only true TCP connection (made easier with a ssh keypair, private key on laptop). Other things can be tunnelled through the ssh connection. > If the Sun is compromised, then your ssh connection and password is > compromised. If your ssh password is compromised then an attacker can > use it to connect to the laptop and get your secret key. If the secret > key has no passphrase, then the attacker has all they need. If the Sun machine is compromised, then I can see nothing compromised. They can, of course, send mail as "me", but it won't be GPG signed. The GPG private key is stored on my laptop (with a passphrase), but is never sent over the ssh tunnel. When signing needs to occur the MUA will call out to some program or script that will use the ssh tunnel to send the email message back to the laptop, on which it will be displayed for checking. GPG signing takes place on the laptop, and the signed message is returned to the MUA. Dave. -- David Symonds USyd::SITRG::PhD Student http://www.it.usyd.edu.au/~dasymond/ From stuart.yeates at computing-services.oxford.ac.uk Tue Apr 20 14:28:35 2004 From: stuart.yeates at computing-services.oxford.ac.uk (Stuart A Yeates) Date: Tue Apr 20 14:27:34 2004 Subject: Remote signing In-Reply-To: <20040420121005.GB10970@it.usyd.edu.au> References: <20040419234757.GA23117@it.usyd.edu.au> <4084D75E.8070503@computing-services.oxford.ac.uk> <20040420084111.GA28365@Update.UU.SE> <4084F3B4.2030507@computing-services.oxford.ac.uk> <20040420121005.GB10970@it.usyd.edu.au> Message-ID: <40851773.20005@computing-services.oxford.ac.uk> Dave Symonds wrote: > On Tue, Apr 20, 2004 at 10:56:04AM +0100, Stuart A Yeates wrote: > >>Let me see if I've understood you correctly: > > > I think you have it slightly backwards... > > >>You have a Sun which does all your you normal mail handling, and at >>whose console you do do your computing. You have another laptop >>connected to the Sun via a public network which has your secret key. >>When you wish to use your public keys you use ssh to start a bash shell >>(or similar) on the laptop, transfer your data, perform your operation >>and transfer your data back. Possibly you have a script to automate some >>of these connection/transfer/opration/transfer steps. > > > The Sun does all the normal mail handling, but there is no console that I > use to access it directly (only via ssh). The ssh connection from my laptop > to the Sun is the only true TCP connection (made easier with a ssh keypair, > private key on laptop). Other things can be tunnelled through the ssh > connection. It's still not clear where are the (ssh, gpg) keys stored, where are the gpg cryptographic operations are performed and where the passwords and passphrases are input and transfered. > The GPG private > key is stored on my laptop (with a passphrase), but is never sent over the ssh > tunnel. When signing needs to occur the MUA will call out to some program or > script that will use the ssh tunnel to send the email message back to the > laptop, on which it will be displayed for checking. GPG signing takes place > on the laptop, and the signed message is returned to the MUA. Where does the MUA (or the program it calls out to) get the ssh password and the gpg passphrase from? If the ssh connection is establishable automatically, then an attacker who has compromised the Sun can estalish a connection to the laptop. If the connection is to a general purpose account, all sorts of badness may result. stuart -- Stuart Yeates stuart.yeates@computing-services.oxford.ac.uk OSS Watch http://www.oss-watch.ac.uk/ Oxford Text Archive http://ota.ahds.ac.uk/ Humbul Humanities Hub http://www.humbul.ac.uk/ From dasymond at it.usyd.edu.au Tue Apr 20 14:39:51 2004 From: dasymond at it.usyd.edu.au (Dave Symonds) Date: Tue Apr 20 14:37:01 2004 Subject: Remote signing In-Reply-To: <40851773.20005@computing-services.oxford.ac.uk> References: <20040419234757.GA23117@it.usyd.edu.au> <4084D75E.8070503@computing-services.oxford.ac.uk> <20040420084111.GA28365@Update.UU.SE> <4084F3B4.2030507@computing-services.oxford.ac.uk> <20040420121005.GB10970@it.usyd.edu.au> <40851773.20005@computing-services.oxford.ac.uk> Message-ID: <20040420123950.GA4404@it.usyd.edu.au> On Tue, Apr 20, 2004 at 01:28:35PM +0100, Stuart A Yeates wrote: > >The Sun does all the normal mail handling, but there is no console that I > >use to access it directly (only via ssh). The ssh connection from my > >laptop > >to the Sun is the only true TCP connection (made easier with a ssh keypair, > >private key on laptop). Other things can be tunnelled through the ssh > >connection. > > It's still not clear where are the (ssh, gpg) keys stored, where are the > gpg cryptographic operations are performed and where the passwords and > passphrases are input and transfered. All private keys are kept on the laptop, as is the GPG operations. The passphrases are entered directly onto the laptop. Emails are entered into the Sun via the ssh tunnel, but the actual privacy is not really an issue, only the integrity. > If the ssh connection is establishable automatically, then an attacker > who has compromised the Sun can estalish a connection to the laptop. If > the connection is to a general purpose account, all sorts of badness may > result. The ssh connection that is opened from the laptop to the Sun is opened on the laptop, and the "returning" connection for the signing is tunneled through this existing connection. Dave. -- David Symonds USyd::SITRG::PhD Student http://www.it.usyd.edu.au/~dasymond/ From pt at radvis.nu Tue Apr 20 15:02:45 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 20 14:59:53 2004 Subject: twofish keysize In-Reply-To: <87fzb0h1ii.fsf@vigenere.g10code.de> References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> Message-ID: <6.0.3.0.2.20040420144800.027dd950@localhost> At 09:26 2004-04-19, Werner Koch wrote: >On Mon, 19 Apr 2004 00:19:58 +0200, Malte Gell said: > >> Maybe he expects a performance advantage when using 128 bit keys instead >> of 256 bit. By the way, are there estimations how much slower a 256 bit >> Twofish key is? I guess one will only notice it whith huge amount of > >here are encrypt/decrypt result from libgcrypt. > > ECB CBC CFB CTR > --------------- --------------- --------------- ---------------- >3DES 140ms 150ms 170ms 190ms 150ms 160ms 180ms 180ms >CAST5 50ms 50ms 70ms 80ms 60ms 60ms 90ms 80ms >BLOWFISH 90ms 100ms 110ms 120ms 110ms 110ms 130ms 130ms >AES 40ms 40ms 50ms 60ms 40ms 50ms 80ms 80ms >AES192 40ms 40ms 60ms 60ms 50ms 50ms 80ms 80ms >AES256 40ms 50ms 60ms 70ms 50ms 60ms 90ms 80ms >TWOFISH 50ms 50ms 60ms 60ms 60ms 60ms 100ms 100ms >TWOFISH128 50ms 50ms 60ms 60ms 60ms 60ms 100ms 100ms > >As you can see, there is no difference between twofish (256 bit) and >twofish 128. > >The disadvantage of a 256 bit key is that is requires double as much >entropy to create a session key. > How large files where used in this performance test? I recently read a NIST evaluation: For 256-bit keys TWOFISH was slightly faster than AES on (very) large files. BTW I've been told it isn't wise to encrypt files larger than a few MB using a block size of 64 bits. What's the limit for the block size 128 bits, used in the new AES and TWOFISH ciphers? And what's the problem? Per Tunedal From Jon.Morisey at serono.com Tue Apr 20 15:44:28 2004 From: Jon.Morisey at serono.com (Jon.Morisey@serono.com) Date: Tue Apr 20 15:48:00 2004 Subject: How To: Create a batch mode, signed, encryted file on AIX? Message-ID: Skipped content of type multipart/alternative-------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/jpeg Size: 7628 bytes Desc: not available Url : /pipermail/attachments/20040420/9b1e32cb/attachment-0001.jpeg From toor at izardsnest.org Tue Apr 20 16:18:25 2004 From: toor at izardsnest.org (Alexander Komarov) Date: Tue Apr 20 16:15:49 2004 Subject: How To: Create a batch mode, signed, encryted file on AIX? Message-ID: <1082470705.40853131daa14@mail.izardsnest.org> I am using this: echo password | gpg .... --passphrase-fd 0 Could anybody suggest if it is secure? Quoting Jon.Morisey@serono.com: > Hi all, > > I am new to GNU and have a question regarding their encryption tools? I > have an options file in place. How can I encrypt and sign a file in batch > mode in the most secure way? > > I have considered: > GPG=`/usr/local/bin/gpg --encrypt ${FILE_TO_ENCRYPT}< ${GNUGPPASS} > EOF` > that doesn't work > > I also checked: > /usr/local/bin/gpg --encrypt --passphrase-fd ${GNUPASSFILE} > ${FILE_TO_ENCRYPT} > this one is always waiting for input, I have not been able to get it > to read from the file containing the passphrase. Does anyone know how to > make it accept the passphrase from the file descriptor parameter? > > I know that is not so secure anyway so I also found: > http://www.gnupg.org/(en)/documentation/faqs.html > this one fails for me on step 3. The command errors out. Does anyone > know how to make this one work? > > > > 4.14) How can I use GnuPG in an automated environment? > You should use the option --batch and don't use passphrases as there is > usually no way to store it more securely than on the secret keyring > itself. The suggested way to create keys for an automated environment is: > On a secure machine: > 1. If you want to do automatic signing, create a signing subkey for your > key > (use the interactive key editing menu by issueing the command 'gpg > --edit-key keyID', enter "addkey" and select the DSA key type). > 2. Make sure that you use a passphrase (needed by the current > implementation). > 3. gpg --export-secret-subkeys --no-comment foo >secring.auto > 4. Copy secring.auto and the public keyring to a test directory. > 5. Change to this directory. > 6. gpg --homedir . --edit foo and use "passwd" to remove the passphrase > from > the subkeys. You may also want to remove all unused subkeys. > 7. Copy secring.auto to a floppy and carry it to the target box. > On the target machine: > 1. Install secring.auto as the secret keyring. > 2. Now you can start your new service. It's also a good idea to install > an > intrusion detection system so that you hopefully get a notice of an > successful intrusion, so that you in turn can revoke all the subkeys > installed on that machine and install new subkeys. > > > Regards, > > > ********************************************************************************** > > * * > * Jon Morisey, OCP E-mail: > jon.morisey@serono.com * > * Serono, Project Manager Office: 781-681-2336 > * > * One Technology Place Cellular: 781-308-9295 > * > * Rockland, MA 02370 Fax: 781-681-2922 > * > * * > > ********************************************************************************** > > > > > > > ******************************************************************************************** > S - This message contains confidential information and is intended only for > the individual > named. If you are not the named addressee, you should not disseminate, > distribute or copy > this e-mail. Please notify the sender immediately by e-mail if you have > received this > e-mail by mistake and delete this e-mail from your system. > e-mail transmission cannot be guaranteed to be secure or error-free as > information could be > intercepted, corrupted, lost, destroyed, arrive late or incomplete, or > contain malware. The > presence of this disclaimer is not a proof that it was originated at Serono > International S.A. > or one of its affiliates. Serono International S.A and its affiliates > therefore do not accept > liability for any errors or omissions in the content of this message, which > arise as a result > of e-mail transmission. If verification is required, please request a > hard-copy version. > Serono International SA, 15bis Chemin Des Mines, Geneva, Switzerland, > www.serono.com. > ********************************************************************************************* From cedar at 3web.net Tue Apr 20 16:26:03 2004 From: cedar at 3web.net (C. D. Rok) Date: Tue Apr 20 16:23:27 2004 Subject: twofish keysize In-Reply-To: <6.0.3.0.2.20040420144800.027dd950@localhost> References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> <6.0.3.0.2.20040420144800.027dd950@localhost> Message-ID: <408532FB.6070809@3web.net> Per Tunedal Casual wrote: > BTW I've been told it isn't wise to encrypt files larger than a few MB > using a block size of 64 bits. I wasn't aware of this; is it cipher dependent od does it apply to all block ciphers? Any pointers to additinal info? TIA, C. Rok From kenta at mit.edu Tue Apr 20 17:47:54 2004 From: kenta at mit.edu (Ken T Takusagawa) Date: Tue Apr 20 17:45:06 2004 Subject: encrypting files in a script In-Reply-To: References: <4DCE15B9C4E66F4CA967EBF64C53D64D1902C6@bstn-exch1.forumsys.com> Message-ID: On Fri, 16 Apr 2004, Atom 'Smasher' wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Fri, 16 Apr 2004, Hasnain Mujtaba wrote: > > I am encrypting some files using a shell script. GPG asks the question: > > > > "It is NOT certain that the key belongs to the person named > > in the user ID. If you *really* know what you are doing, > > you may answer the next question with yes" > > > > How can I supress this question so the script can run through the end? > =============================== > > this will do the trick: > --trust-model always > > alternatively, you could mark the key as trusted. How do I mark the key as trusted? I still get "the question" even after I do gpg --edit-key foo trust 4 ( = "I trust fully") quit --ken From pt at radvis.nu Tue Apr 20 18:45:43 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Tue Apr 20 18:42:38 2004 Subject: Blocksize versus file size was: Re: twofish keysize In-Reply-To: <408532FB.6070809@3web.net> References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> <6.0.3.0.2.20040420144800.027dd950@localhost> <408532FB.6070809@3web.net> Message-ID: <6.0.3.0.2.20040420184215.027e44d0@localhost> At 16:26 2004-04-20, C. D. Rok wrote: >Per Tunedal Casual wrote: > >> BTW I've been told it isn't wise to encrypt files larger than a few MB >> using a block size of 64 bits. > >I wasn't aware of this; is it cipher dependent od does it apply to >all block ciphers? Any pointers to additinal info? > >TIA, >C. Rok > It appears to apply to all block ciphers, but it should be "a few HUNDRED megabytes", according to what I found with Google: "Blowfish has a 64 bit block size and a key length of anywhere from 32 bits to 448 bits. ... It should be noted that, like other block cyphers with a 64 bit block, it is risky to encrypt extremely large texts (more than a few hundred megabytes) with a single key. If more data is to be encrypted a block cipher with a larger block size should be used." (http://www.djhd.com/onlinehelp/ac/encryption_overview.htm) Per Tunedal From hmujtaba at forumsys.com Tue Apr 20 19:45:59 2004 From: hmujtaba at forumsys.com (Hasnain Mujtaba) Date: Tue Apr 20 19:43:42 2004 Subject: AES cipher support 1.0.6 and 1.0.7 Message-ID: <4DCE15B9C4E66F4CA967EBF64C53D64D67B326@bstn-exch1.forumsys.com> Hi, I encrypted a 2GB file using a DH/DSS 2048/1024 (AES-128) keypair using a third party tool. I gave the encrypted file to GPG 1.0.6 but it failed to decrypt the message with the following reason: gpg: protection algorithm 254 is not supported gpg: encrypted with 2048-bit ELG-E key, ID 3F29C018, created 2004-04-16 "GPG-DH-2048 " gpg: public key decryption failed: unknown cipher algorithm gpg: decryption failed: secret key not available Then I gave the encrypted file to GPG 1.0.7 and the decryption succeeded. I'm wondering if this is due to AES support, or lack of, in the two versions. If anyone knows, please let me know. Thanks Hasnain. ---- The information contained in this electronic mail and any attached document is the confidential and proprietary business information of Forum Systems, Inc. It is intended solely for the addressed recipient listed above. It may not be distributed in any manner without the express written consent of Forum Systems, Inc. From gnupg at ml0402.albert.uni.cc Tue Apr 20 19:49:58 2004 From: gnupg at ml0402.albert.uni.cc (Albert) Date: Tue Apr 20 19:47:53 2004 Subject: encrypting files in a script In-Reply-To: References: <4DCE15B9C4E66F4CA967EBF64C53D64D1902C6@bstn-exch1.forumsys.com> Message-ID: <200404201949.58773.gnupg@ml0402.albert.uni.cc> Am Dienstag, 20. April 2004 17:47 schrieb Ken T Takusagawa: > How do I mark the key as trusted I didn't read the whole thread, but maybe this helps: I use the following: echo "Password" | gpg --quiet --no-tty --pgp8 --batch --passphrase-fd 0 --charset iso-8859-15 --force-mdc --no-secmem-warning --symmetric --output "createdfile" "originalfile" | gpg --local-user $SENDER --no-secmem-warning --quiet --textmode --clearsign --trusted-key "$LONGKEYID" \ ... Albert From ian_bruce at fastmail.fm Wed Apr 21 01:21:45 2004 From: ian_bruce at fastmail.fm (Ian Bruce) Date: Wed Apr 21 01:16:53 2004 Subject: Remote signing In-Reply-To: <20040420123950.GA4404@it.usyd.edu.au> References: <20040419234757.GA23117@it.usyd.edu.au> <4084D75E.8070503@computing-services.oxford.ac.uk> <20040420084111.GA28365@Update.UU.SE> <4084F3B4.2030507@computing-services.oxford.ac.uk> <20040420121005.GB10970@it.usyd.edu.au> <40851773.20005@computing-services.oxford.ac.uk> <20040420123950.GA4404@it.usyd.edu.au> Message-ID: <20040420162145.560f6feb.ian_bruce@fastmail.fm> On Tue, 20 Apr 2004 22:39:51 +1000 Dave Symonds wrote: > On Tue, Apr 20, 2004 at 01:28:35PM +0100, Stuart A Yeates wrote: > > >The Sun does all the normal mail handling, but there is no console > > >that I use to access it directly (only via ssh). The ssh > > >connection from my laptop > > >to the Sun is the only true TCP connection (made easier with a ssh > > >keypair, private key on laptop). Other things can be tunnelled > > >through the ssh connection. > > > > It's still not clear where are the (ssh, gpg) keys stored, where are > > the gpg cryptographic operations are performed and where the > > passwords and passphrases are input and transfered. > > All private keys are kept on the laptop, as is the GPG operations. The > passphrases are entered directly onto the laptop. Emails are entered > into the Sun via the ssh tunnel, but the actual privacy is not really > an issue, only the integrity. It appears to me that there is some mutual misunderstanding involved in this conversation. As far as I understand, the situation is as follows: The human user is physically seated in front of his own laptop computer, on which is stored his GPG secret key. There is an SSH session connecting that laptop to a remote Sun computer. The mail client is running on the Sun machine, with its console I/O tunnelled over SSH to a terminal process running on the laptop. It is proposed that when the mail client sends a message, it will first provide the message text to a script running on the laptop, using the same SSH session. That script will then prompt the human user for the GPG passphrase, sign the message text, and return it to the mail client on the Sun. The message will then be sent using the normal SMTP protocols. As far as security is concerned, the important point is that neither the GPG key nor passphrase will ever leave the laptop. They are entirely separate from the SSH key. If the Sun machine or the SSH session are compromised, the worst that can happen is that the human user will be tricked into signing a bogus message. Is that about right? Assuming that it is, I have to ask why you wouldn't just run the mail client on the laptop and avoid all the extra complication. You can then either tunnel the IMAP and SMTP connections over SSH or just use IMAP/SSL and SMTP/SSL directly. Ian Bruce From DHill at StudentLoan.org Tue Apr 20 19:15:58 2004 From: DHill at StudentLoan.org (David Hill) Date: Wed Apr 21 08:25:10 2004 Subject: RSA Encryption / exchange with PGP Message-ID: We are trying to exchange files with a company using PGP Command Line - Freeware version 6.5 with RSA keys generated with version 2.6.2, file exchange remains compatible with the old 2.6.2 version. We cannot decrypt their files and they cannot decrypt ours. We could easily generate a new key, but our trading partner is very established and will not want to regenerate. We are using GnuPG 1.2.4 to generate keys with the following options Type: DSA & ElGamal (the only option that allowed encryption, RSA was only for signatures) key length: 1024 bit key does not expire Real Name: Iowa Student Loan Liquidity Corp. Email: dhill@studentloan.org Comment: iLink Dev I am sending the trading partner a new text file encrypted with the following options: gpg --rfc1991 --force-v3-sigs --compress-algo 1 --recipient "THEM" --recipient "US" --output "readme.gpg" -es "readme.txt" In previous attempts I didn't use the first three commands, I am hoping these will help our trading partner decrypt, but I don't have high hopes of success. Even if by some miracle that worked, we still can't decrypt their files. Do we need to use different keys or protocols in order to interoperate? This is a sample decryption session: C:\gnupg>gpg --verbose --decrypt "install.log.pgp" gpg: public key is F58C3B48 gpg: using secondary key F58C3B48 instead of primary key 4D1D68AC You need a passphrase to unlock the secret key for user: "Iowa Student Loan Liquidity Corp. (iLink Dev) " gpg: using secondary key F58C3B48 instead of primary key 4D1D68AC 1024-bit ELG-E key, ID F58C3B48, created 2004-04-19 (main key ID 4D1D68AC) gpg: encrypted with 1024-bit ELG-E key, ID F58C3B48, created 2004-04-19 "Iowa Student Loan Liquidity Corp. (iLink Dev) " gpg: CAST5 encrypted data gpg: block_filter 024BCB68: read error (size=7302,a->size=536874118) gpg: WARNING: message was not integrity protected gpg: block_filter: pending bytes! C:\gnupg> From atom at suspicious.org Wed Apr 21 01:12:56 2004 From: atom at suspicious.org (Atom 'Smasher') Date: Wed Apr 21 08:25:15 2004 Subject: encrypting files in a script In-Reply-To: References: <4DCE15B9C4E66F4CA967EBF64C53D64D1902C6@bstn-exch1.forumsys.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tue, 20 Apr 2004, Ken T Takusagawa wrote: > On Fri, 16 Apr 2004, Atom 'Smasher' wrote: > > On Fri, 16 Apr 2004, Hasnain Mujtaba wrote: > > > I am encrypting some files using a shell script. GPG asks the question: > > > > > > "It is NOT certain that the key belongs to the person named > > > in the user ID. If you *really* know what you are doing, > > > you may answer the next question with yes" > > > > > > How can I supress this question so the script can run through the end? > > =============================== > > > > this will do the trick: > > --trust-model always > > > > alternatively, you could mark the key as trusted. > > How do I mark the key as trusted? I still get "the > question" even after I do > > gpg --edit-key foo > trust > 4 ( = "I trust fully") > quit ================================= "full" trust isn't enough... you'll have to select: 5 = I trust ultimately or use: --trust-model always ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Those who make peaceful revolution impossible will make violent revolution inevitable." -- John F. Kennedy -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCFrn0ACgkQnCgLvz19QePaCQCfSm00mGnjwB/RwBnvDvZakPZK 5yUAnRhp7UNoiwaOvOkw88elLgYM2QRD =sYo9 -----END PGP SIGNATURE----- From wk at gnupg.org Wed Apr 21 11:05:15 2004 From: wk at gnupg.org (Werner Koch) Date: Wed Apr 21 10:51:19 2004 Subject: twofish keysize In-Reply-To: <6.0.3.0.2.20040420144800.027dd950@localhost> (Per Tunedal Casual's message of "Tue, 20 Apr 2004 15:02:45 +0200") References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> <6.0.3.0.2.20040420144800.027dd950@localhost> Message-ID: <87zn9566r8.fsf@vigenere.g10code.de> On Tue, 20 Apr 2004 15:02:45 +0200, Per Tunedal Casual said: > How large files where used in this performance test? I recently read a > NIST evaluation: For 256-bit keys TWOFISH was slightly faster than AES That are not files but benchmarks of the actual encryption function. IIRC, a million bytes for each test are used. > on (very) large files. That depends on the key setup which is only done once per encryption. For most applications this is irrelevant. Furthermore OpenPGP does use CFB mode and thus the more expensive AES key setup for _decrytion_ is not required. > BTW I've been told it isn't wise to encrypt files larger than a few MB > using a block size of 64 bits. What's the limit for the block size > 128 Not a few MB but several GB: Due to the birthday paradoxon you will notice on average identical blocks after 2^32 blocks (32 GB). This yields patterns which help in cryptanalysis. It is also the reason why ssh re-negotiates a new key after 1 gig. For a 128 bit block cipher (AES or Twofish) this limit is a pretty reasonable value (2^64 blocks). Shalom-Salam, Werner From boldyrev+nospam at cgitftp.uiggm.nsc.ru Wed Apr 21 08:57:40 2004 From: boldyrev+nospam at cgitftp.uiggm.nsc.ru (Ivan Boldyrev) Date: Wed Apr 21 11:36:54 2004 Subject: Why GPG doesn't use filename inside encrypted file? References: <20040420043055.GC5727@psilocybe.teonanacatl.org> Message-ID: <6kihl1x16v.ln2@ibhome.cgitftp.uiggm.nsc.ru> On 8720 day of my life Freedom_Lover@pobox.com wrote: > Ivan Boldyrev wrote: >> How can I decrypt file so that original name of file is restored >> automatically? > > Does --use-embedded-filename do what you want? Thanks!! And manpage explains why it is not default :) -- Ivan Boldyrev XML -- new language of ML family. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 480 bytes Desc: not available Url : /pipermail/attachments/20040421/8e0b3bec/attachment.bin From kenta at MIT.EDU Wed Apr 21 16:28:34 2004 From: kenta at MIT.EDU (Ken T Takusagawa) Date: Wed Apr 21 16:25:51 2004 Subject: How To: Create a batch mode, signed, encryted file on AIX? In-Reply-To: <1082470705.40853131daa14@mail.izardsnest.org> References: <1082470705.40853131daa14@mail.izardsnest.org> Message-ID: Someone running the "ps" command at just the right time will see the command "echo password" in the list of processes, thereby learning your password. Maybe if you are the only user of your machine, and you aren't too paranoid, then it's an OK thing to do. The point about --passphrase-fd is that it expects a number, not a file name. --ken On Tue, 20 Apr 2004, Alexander Komarov wrote: > I am using this: > echo password | gpg .... --passphrase-fd 0 > Could anybody suggest if it is secure? > > Quoting Jon.Morisey@serono.com: > > Hi all, > > > > I am new to GNU and have a question regarding their encryption tools? I > > have an options file in place. How can I encrypt and sign a file in batch > > mode in the most secure way? > > > > I have considered: > > GPG=`/usr/local/bin/gpg --encrypt ${FILE_TO_ENCRYPT}< > ${GNUGPPASS} > > EOF` > > that doesn't work > > > > I also checked: > > /usr/local/bin/gpg --encrypt --passphrase-fd ${GNUPASSFILE} > > ${FILE_TO_ENCRYPT} > > this one is always waiting for input, I have not been able to get it > > to read from the file containing the passphrase. Does anyone know how to > > make it accept the passphrase from the file descriptor parameter? > > > > I know that is not so secure anyway so I also found: > > http://www.gnupg.org/(en)/documentation/faqs.html > > this one fails for me on step 3. The command errors out. Does anyone > > know how to make this one work? > > > > > > > > 4.14) How can I use GnuPG in an automated environment? > > You should use the option --batch and don't use passphrases as there is > > usually no way to store it more securely than on the secret keyring > > itself. The suggested way to create keys for an automated environment is: > > On a secure machine: > > 1. If you want to do automatic signing, create a signing subkey for your > > key > > (use the interactive key editing menu by issueing the command 'gpg > > --edit-key keyID', enter "addkey" and select the DSA key type). > > 2. Make sure that you use a passphrase (needed by the current > > implementation). > > 3. gpg --export-secret-subkeys --no-comment foo >secring.auto > > 4. Copy secring.auto and the public keyring to a test directory. > > 5. Change to this directory. > > 6. gpg --homedir . --edit foo and use "passwd" to remove the passphrase > > from > > the subkeys. You may also want to remove all unused subkeys. > > 7. Copy secring.auto to a floppy and carry it to the target box. > > On the target machine: > > 1. Install secring.auto as the secret keyring. > > 2. Now you can start your new service. It's also a good idea to install > > an > > intrusion detection system so that you hopefully get a notice of an > > successful intrusion, so that you in turn can revoke all the subkeys > > installed on that machine and install new subkeys. > > > > > > Regards, > > > > > > ********************************************************************************** > > > > * * > > * Jon Morisey, OCP E-mail: > > jon.morisey@serono.com * > > * Serono, Project Manager Office: 781-681-2336 > > * > > * One Technology Place Cellular: 781-308-9295 > > * > > * Rockland, MA 02370 Fax: 781-681-2922 > > * > > * * > > > > ********************************************************************************** > > > > > > > > > > > > > > > ******************************************************************************************** > > S - This message contains confidential information and is intended only for > > the individual > > named. If you are not the named addressee, you should not disseminate, > > distribute or copy > > this e-mail. Please notify the sender immediately by e-mail if you have > > received this > > e-mail by mistake and delete this e-mail from your system. > > e-mail transmission cannot be guaranteed to be secure or error-free as > > information could be > > intercepted, corrupted, lost, destroyed, arrive late or incomplete, or > > contain malware. The > > presence of this disclaimer is not a proof that it was originated at Serono > > International S.A. > > or one of its affiliates. Serono International S.A and its affiliates > > therefore do not accept > > liability for any errors or omissions in the content of this message, which > > arise as a result > > of e-mail transmission. If verification is required, please request a > > hard-copy version. > > Serono International SA, 15bis Chemin Des Mines, Geneva, Switzerland, > > www.serono.com. > > > ********************************************************************************************* > > > > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users > From Jon.Morisey at serono.com Wed Apr 21 19:15:46 2004 From: Jon.Morisey at serono.com (Jon.Morisey@serono.com) Date: Wed Apr 21 19:16:38 2004 Subject: How To: Create a batch mode, signed, encryted file on AIX? Message-ID: Skipped content of type multipart/alternative-------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/jpeg Size: 7628 bytes Desc: not available Url : /pipermail/attachments/20040421/2a6e7c59/attachment-0001.jpeg From kenta at MIT.EDU Wed Apr 21 20:19:34 2004 From: kenta at MIT.EDU (Ken T Takusagawa) Date: Wed Apr 21 20:16:50 2004 Subject: How To: Create a batch mode, signed, encryted file on AIX? In-Reply-To: References: Message-ID: Unfortunately this is beyond my knowledge. I know 0=stdin, 1=stdout, 2=stderr (so 0 is the only one immediately useful for passphrase). There may be higher numbers, but I don't know how to use them. Have you tried: gpg -passphrase-fd 0 etc. << EOF password EOF --ken On Wed, 21 Apr 2004 Jon.Morisey@serono.com wrote: > Dear Ken, > > What does the number represent? I know you pass "0" which means stdin, > but are there other #'s that your can pass to make it point to a file at a > specified location? Is there a way to compile the passphrase in some code > that makes the correct call the gpg? > > Regards, > > > ********************************************************************************** > * * > * Jon Morisey, OCP E-mail: > jon.morisey@serono.com * > * Serono, Project Manager Office: 781-681-2336 > * > * One Technology Place Cellular: 781-308-9295 > * > * Rockland, MA 02370 Fax: 781-681-2922 > * > * * > > ********************************************************************************** > > > > > > > > > Ken T Takusagawa > 04/21/2004 10:28 AM > > > To: Alexander Komarov > cc: gnupg-users@gnupg.org, Jon.Morisey@serono.com > Subject: Re: How To: Create a batch mode, signed, encryted file on AIX? > > > Someone running the "ps" command at just the right time will > see the command "echo password" in the list of processes, > thereby learning your password. > > Maybe if you are the only user of your machine, and you > aren't too paranoid, then it's an OK thing to do. > > The point about --passphrase-fd is that it expects a number, > not a file name. > > --ken > > On Tue, 20 Apr 2004, Alexander Komarov wrote: > > > I am using this: > > echo password | gpg .... --passphrase-fd 0 > > Could anybody suggest if it is secure? > > > > Quoting Jon.Morisey@serono.com: > > > Hi all, > > > > > > I am new to GNU and have a question regarding their encryption tools? > I > > > have an options file in place. How can I encrypt and sign a file in > batch > > > mode in the most secure way? > > > > > > I have considered: > > > GPG=`/usr/local/bin/gpg --encrypt ${FILE_TO_ENCRYPT}< > > ${GNUGPPASS} > > > EOF` > > > that doesn't work > > > > > > I also checked: > > > /usr/local/bin/gpg --encrypt --passphrase-fd ${GNUPASSFILE} > > > ${FILE_TO_ENCRYPT} > > > this one is always waiting for input, I have not been able to get > it > > > to read from the file containing the passphrase. Does anyone know how > to > > > make it accept the passphrase from the file descriptor parameter? > > > > > > I know that is not so secure anyway so I also found: > > > http://www.gnupg.org/(en)/documentation/faqs.html > > > this one fails for me on step 3. The command errors out. Does > anyone > > > know how to make this one work? > > > > > > > > > > > > 4.14) How can I use GnuPG in an automated environment? > > > You should use the option --batch and don't use passphrases as there > is > > > usually no way to store it more securely than on the secret keyring > > > itself. The suggested way to create keys for an automated environment > is: > > > On a secure machine: > > > 1. If you want to do automatic signing, create a signing subkey > for your > > > key > > > (use the interactive key editing menu by issueing the command 'gpg > > > --edit-key keyID', enter "addkey" and select the DSA key type). > > > 2. Make sure that you use a passphrase (needed by the current > > > implementation). > > > 3. gpg --export-secret-subkeys --no-comment foo >secring.auto > > > 4. Copy secring.auto and the public keyring to a test directory. > > > 5. Change to this directory. > > > 6. gpg --homedir . --edit foo and use "passwd" to remove the > passphrase > > > from > > > the subkeys. You may also want to remove all unused subkeys. > > > 7. Copy secring.auto to a floppy and carry it to the target box. > > > On the target machine: > > > 1. Install secring.auto as the secret keyring. > > > 2. Now you can start your new service. It's also a good idea to > install > > > an > > > intrusion detection system so that you hopefully get a notice of an > > > successful intrusion, so that you in turn can revoke all the subkeys > > > installed on that machine and install new subkeys. > > > > > > > > > Regards, > > > > > > > > > > ********************************************************************************** > > > > > > * * > > > * Jon Morisey, OCP E-mail: > > > jon.morisey@serono.com * > > > * Serono, Project Manager Office: 781-681-2336 > > > * > > > * One Technology Place Cellular: 781-308-9295 > > > * > > > * Rockland, MA 02370 Fax: 781-681-2922 > > > * > > > * * > > > > > > > ********************************************************************************** > > > > > > > > > > > > > > > > > > > > > > > > ******************************************************************************************** > > > S - This message contains confidential information and is intended > only for > > > the individual > > > named. If you are not the named addressee, you should not disseminate, > > > distribute or copy > > > this e-mail. Please notify the sender immediately by e-mail if you > have > > > received this > > > e-mail by mistake and delete this e-mail from your system. > > > e-mail transmission cannot be guaranteed to be secure or error-free as > > > information could be > > > intercepted, corrupted, lost, destroyed, arrive late or incomplete, or > > > contain malware. The > > > presence of this disclaimer is not a proof that it was originated at > Serono > > > International S.A. > > > or one of its affiliates. Serono International S.A and its affiliates > > > therefore do not accept > > > liability for any errors or omissions in the content of this message, > which > > > arise as a result > > > of e-mail transmission. If verification is required, please request a > > > hard-copy version. > > > Serono International SA, 15bis Chemin Des Mines, Geneva, Switzerland, > > > www.serono.com. > > > > > > ********************************************************************************************* > > > > > > > > _______________________________________________ > > Gnupg-users mailing list > > Gnupg-users@gnupg.org > > http://lists.gnupg.org/mailman/listinfo/gnupg-users > > > > > > > ******************************************************************************************** > S - This message contains confidential information and is intended only for the individual > named. If you are not the named addressee, you should not disseminate, distribute or copy > this e-mail. Please notify the sender immediately by e-mail if you have received this > e-mail by mistake and delete this e-mail from your system. > e-mail transmission cannot be guaranteed to be secure or error-free as information could be > intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain malware. The > presence of this disclaimer is not a proof that it was originated at Serono International S.A. > or one of its affiliates. Serono International S.A and its affiliates therefore do not accept > liability for any errors or omissions in the content of this message, which arise as a result > of e-mail transmission. If verification is required, please request a hard-copy version. > Serono International SA, 15bis Chemin Des Mines, Geneva, Switzerland, www.serono.com. > ********************************************************************************************* > > From malte.gell at gmx.de Thu Apr 22 00:11:40 2004 From: malte.gell at gmx.de (Malte Gell) Date: Thu Apr 22 00:10:09 2004 Subject: twofish keysize In-Reply-To: <87zn9566r8.fsf@vigenere.g10code.de> References: <200404162004.13385.jluehr@gmx.net> <6.0.3.0.2.20040420144800.027dd950@localhost> <87zn9566r8.fsf@vigenere.g10code.de> Message-ID: <200404220012.22771.malte.gell@gmx.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am Mittwoch, 21. April 2004 11:05 schrieb Werner Koch: > On Tue, 20 Apr 2004 15:02:45 +0200, Per Tunedal Casual said: > > BTW I've been told it isn't wise to encrypt files larger than a few > > MB using a block size of 64 bits. What's the limit for the block > > size 128 > Not a few MB but several GB: Due to the birthday paradoxon you will > notice on average identical blocks after 2^32 blocks (32 GB). This > yields patterns which help in cryptanalysis. It is also the reason > why ssh re-negotiates a new key after 1 gig. Isn't this issue one of the reasons why OpenPGP compresses data before encrypting it? If a large cleartext chunk contains much redundant data, compression may reduce it and helps to reduce such patterns. So encrypting with "only" a 64 bit block size shouldn't be a problem. Is this correct? Malte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAhvGmGzg12gD8wBYRAlnKAJ9R5QNbzy8TuU6sakvSSaroiL74wwCfctkm bPxxHKXXInpLTECvEMzAk0g= =YYnI -----END PGP SIGNATURE----- From malte.gell at gmx.de Thu Apr 22 00:28:47 2004 From: malte.gell at gmx.de (Malte Gell) Date: Thu Apr 22 00:28:09 2004 Subject: twofish keysize In-Reply-To: <87fzb0h1ii.fsf@vigenere.g10code.de> References: <200404162004.13385.jluehr@gmx.net> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> Message-ID: <200404220029.41619.malte.gell@gmx.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am Montag, 19. April 2004 09:26 schrieb Werner Koch: > On Mon, 19 Apr 2004 00:19:58 +0200, Malte Gell said: > > Maybe he expects a performance advantage when using 128 bit keys > > instead of 256 bit. By the way, are there estimations how much > > slower a 256 bit Twofish key is? I guess one will only notice it > > whith huge amount of >[...] > As you can see, there is no difference between twofish (256 bit) and > twofish 128. > The disadvantage of a 256 bit key is that is requires double as much > entropy to create a session key. Would it be much hassle to introduce 2Fish-128,196,256, just according to AES-128,196,256 to adress this? Since many encryption keys are 1024/2048 bit a 128 bit session key should be sufficient and wouldn't "waste" entropy? On the one hand asym. keys > 2048 are said to be useful only in certain situation, on the other hand PGP/GnuPG amply uses system's entropy for AES-256/Twofish(256)...(sorry for key length discussion now on the symm. side,I'll immediately duck and cover;-) Malte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAhvWgGzg12gD8wBYRApo6AJ0TOvg39qZINLLJrfat8Qm9wWZ5wgCfaMGJ X6muWGSiEhOnSnr1Pv1C28w= =zmnM -----END PGP SIGNATURE----- From linux at codehelp.co.uk Thu Apr 22 00:34:19 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Thu Apr 22 00:30:40 2004 Subject: trust management script Message-ID: <200404212334.24999.linux@codehelp.co.uk> Just in case this is useful to anyone else: (Replace 28BCB3E3 in line 11 with the keyid of your main signing key) #!/bin/bash echo "Printing details of fully trusted keys that are set to FULL trust" echo "but which do NOT carry your signature" KEYS=`gpg --list-keys --with-colons | grep "pub:f:" | grep "::f:" | cut \ -d: -f5` for check in $KEYS do SIGS='' if [ "$check" ]; then OUTPUT=`gpg --list-key $check | grep "^pub"` SIGS="`gpg --list-sigs $check | grep 28BCB3E3`" if [ -z "$SIGS" ]; then echo "$OUTPUT" fi fi done I wrote it to check that I hadn't inadvertently set a key to fully trusted for verifying other keys when I hadn't actually signed and therefore verified the key itself. I worked on the idea that if I hadn't verified the key, I probably had never met the person behind the key and therefore I was unlikely to be able to make much of a decision on how carefully that person would verify someone else's key. (Not having had the chance to see them verify my own key.) It could be used alongside --update-trustdb which catches those keys with no user trust setting but a full GnuPG trust value. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040421/2bd3df24/attachment.bin From pt at radvis.nu Thu Apr 22 09:01:38 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Thu Apr 22 08:58:52 2004 Subject: twofish keysize In-Reply-To: <87zn9566r8.fsf@vigenere.g10code.de> References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> <6.0.3.0.2.20040420144800.027dd950@localhost> <87zn9566r8.fsf@vigenere.g10code.de> Message-ID: <6.0.3.0.2.20040422084123.027e0720@localhost> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 11:05 2004-04-21, Werner Koch wrote: >On Tue, 20 Apr 2004 15:02:45 +0200, Per Tunedal Casual said: > >> How large files where used in this performance test? I recently >> read a >> NIST evaluation: For 256-bit keys TWOFISH was slightly faster than >> AES > >That are not files but benchmarks of the actual encryption >function. IIRC, a million bytes for each test are used. > >> on (very) large files. > >That depends on the key setup which is only done once per encryption. >For most applications this is irrelevant. Furthermore OpenPGP does >use CFB mode and thus the more expensive AES key setup for >_decrytion_ >is not required. I was more concerned about encryption. What about the performance AES-256 compared to TWOFISH for very large files with GPG? AES makes more rounds for larger keys, TWOFISH does the same number of rounds for all key lengths. That might make TWOFISH attractive for large files. > >> BTW I've been told it isn't wise to encrypt files larger than a few >> MB >> using a block size of 64 bits. What's the limit for the block size >> 128 > >Not a few MB but several GB: Due to the birthday paradoxon you will >notice on average identical blocks after 2^32 blocks (32 GB). This >yields patterns which help in cryptanalysis. It is also the reason >why ssh re-negotiates a new key after 1 gig. > >For a 128 bit block cipher (AES or Twofish) this limit is a pretty >reasonable value (2^64 blocks). > A Swedish cryptographer explained the issue to me yesterday. Further he told me that the recommendation "not more than a few hundred Megabytes" for was to have marginal against "bad luck". On average 32 GB is the limit i.e. the probability is 50 % that an adversary finds two identical blocks. But you might have bad luck and he might find two identical blocks in somewhat smaller files. He didn't tell me how to calculate the risk: a) What is the risk (probability) if I encrypt a file of the size S with the blocksize B? and reversed: b) If I want to set the risk to P and use the block size S, how large files can I encrypt? Per Tunedal -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) iD8DBQFAh22PaDDfzFT+2PIRAj5CAJ0Vi9O5Us6QYczof1h97FsVN/qjlgCfRtEb HuqdUMXv8MUz/0HkbMLR6gs= =OGlE -----END PGP SIGNATURE----- From malte.gell at gmx.de Thu Apr 22 11:53:34 2004 From: malte.gell at gmx.de (Malte Gell) Date: Thu Apr 22 11:51:29 2004 Subject: AES256 vs Twofish performance (Was: twofish keysize) In-Reply-To: <6.0.3.0.2.20040422084123.027e0720@localhost> References: <200404162004.13385.jluehr@gmx.net> <87zn9566r8.fsf@vigenere.g10code.de> <6.0.3.0.2.20040422084123.027e0720@localhost> Message-ID: <200404221153.47930.malte.gell@gmx.de> Am Donnerstag, 22. April 2004 09:01 schrieb Per Tunedal Casual: > I was more concerned about encryption. What about the performance > AES-256 compared to TWOFISH for very large files with GPG? AES makes > more rounds for larger keys, TWOFISH does the same number of rounds > for all key lengths. That might make TWOFISH attractive for large > files. AES-256 has 14, Twofish(256) has 16 rounds. BTW, there's some nice Information about AES at Wikipedia http://en.wikipedia.org/wiki/Advanced_Encryption_Standard Just made a quick "real-worl-test", test.tar has about 55 MB and is a tar'ed /home directory on a Linux box with different types of data. Actually Twofish was really slightly faster, this is interesting since Rijndael's believed speed was one of NISTS's major reasons why it became AES and 2Fish-256 has more rounds. *AES-256* time echo test | gpg --passphrase-fd 0 --cipher-algo aes256 --symmetric test.tar real 0m28.687s user 0m26.280s sys 0m0.930s *Twofish* time echo test |gpg --passphrase-fd 0 --cipher-algo twofish --symmetric test.tar real 0m27.055s user 0m24.800s sys 0m1.000s It shows the difference is negligible and AES/Rijndael doesn't seem to be the top performer it is sometimes said to be, at least with large files. I think I may later make a test with a 1000 MB tar-file to see what happens with really large files. HTH Malte -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040422/4da58e30/attachment.bin From atom-gpg at suspicious.org Wed Apr 21 08:37:08 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Thu Apr 22 13:27:12 2004 Subject: ez-pine-gpg v0.4a Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 some messages weren't being displayed properly in v0.3g. v0.4a makes some cosmetic changes to ez-pine-gpg-incoming and changes the way that signed messages are displayed. read the README for more info and a slight change in pine's "display-filters" configuration. http://business-php.com/opensource/ez-pine-gpg/ ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Beware, a record of the books you borrow may end up in the hands of the FBI. And if the FBI requests your records, librarians are prohibited by law from telling you about it. Questions about this policy should be directed to Attorney General John Ashcroft, Department of Justice, Washington, D.C. 20530." -- Sign greeting patrons entering all 10 of the county libraries in Santa Cruz, California -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCGFpkACgkQnCgLvz19QeN5owCgnzFNem7workyJQe8QXSlCcmz uE4AnArImykkBNe9E+m5lBvmfvXnkeYA =ErWl -----END PGP SIGNATURE----- From wk at gnupg.org Thu Apr 22 16:16:48 2004 From: wk at gnupg.org (Werner Koch) Date: Thu Apr 22 16:01:05 2004 Subject: AES256 vs Twofish performance In-Reply-To: <200404221153.47930.malte.gell@gmx.de> (Malte Gell's message of "Thu, 22 Apr 2004 11:53:34 +0200") References: <200404162004.13385.jluehr@gmx.net> <87zn9566r8.fsf@vigenere.g10code.de> <6.0.3.0.2.20040422084123.027e0720@localhost> <200404221153.47930.malte.gell@gmx.de> Message-ID: <871xmg2j3j.fsf@vigenere.g10code.de> On Thu, 22 Apr 2004 11:53:34 +0200, Malte Gell said: > tar'ed /home directory on a Linux box with different types of data. > Actually Twofish was really slightly faster, this is interesting since > Rijndael's believed speed was one of NISTS's major reasons why it Our Twofish implementation is highly optimized but the AES code is pretty much the reference code without any serious optimizations. Werner From wk at gnupg.org Thu Apr 22 16:21:48 2004 From: wk at gnupg.org (Werner Koch) Date: Thu Apr 22 16:06:03 2004 Subject: twofish keysize In-Reply-To: <200404220012.22771.malte.gell@gmx.de> (Malte Gell's message of "Thu, 22 Apr 2004 00:11:40 +0200") References: <200404162004.13385.jluehr@gmx.net> <6.0.3.0.2.20040420144800.027dd950@localhost> <87zn9566r8.fsf@vigenere.g10code.de> <200404220012.22771.malte.gell@gmx.de> Message-ID: <87wu4814ar.fsf@vigenere.g10code.de> On Thu, 22 Apr 2004 00:11:40 +0200, Malte Gell said: > Isn't this issue one of the reasons why OpenPGP compresses data before > encrypting it? If a large cleartext chunk contains much redundant data, > compression may reduce it and helps to reduce such patterns. So OTOH, compressed data has a well known structure and you know also that there is not much entropy in the text. This also gives hints for cryptanalysis. Both, AES and Twofish solve the problem with very large files in a simple and efficient way. You don't need a 256 bit key but well that is what the standard says for OpenPGP's Twofish. Actually it was more a marketing reason that NAI at the time used the 256 bit version. Shalom-Salam, Werner From jharris at widomaker.com Fri Apr 23 02:18:10 2004 From: jharris at widomaker.com (Jason Harris) Date: Fri Apr 23 02:15:46 2004 Subject: revised: new (2004-04-18) keyanalyze results (+sigcheck) Message-ID: <20040423001809.GA10980@pm1.ric-41.lft.widomaker.com> New keyanalyze results are available at: http://keyserver.kjsl.com/~jharris/ka/2004-04-18/ Signatures are now being checked using keyanalyze+sigcheck: http://dtype.org/~aaronl/ This report was revised to fix a signature expiration bug in keyanalyze+sigcheck-0.9. That report is now at: http://keyserver.kjsl.com/~jharris/ka/2004-04-18-bugged/ SHA-1 hashes and sizes for all the "permanent" files: a62faf053aa29ca1611eb47fda27f2e02cdfe4b0 9587286 preprocess.keys a1b5970422513eb9441f50107b004b83fe07bd33 6506410 othersets.txt 2b103c560b2e4b7b04252430584f24c77797228f 2450482 msd-sorted.txt b0f152cbac2bff77aeed70a933fec6d7ac3e7b71 1484 index.html e6b8bb2dcac213b742bca8ac6e2fb52adf823fef 2289 keyring_stats 81c3984231cab5c95854d3da82629d5660bf61d1 967599 msd-sorted.txt.bz2 29b32c8e893d7eb6471994ee34ca795425e780d7 26 other.txt e4e5fed331bca59e5953074776e042ab95e7250f 1387160 othersets.txt.bz2 e75a06a06e2418eb15a067af4b0985c50a3ca339 3915333 preprocess.keys.bz2 4f4cc8c361a79092c3674e6de1e73d511d5df17d 9397 status.txt e433c43d6fde5a1bf5adedd66961e34b2cac2d5f 212005 top1000table.html d3ee55bf0e723c1f480ee57cf283b3822041a9f4 30563 top1000table.html.gz 904a97b96a7dbfe43d4cd2ebe122c14125aa0e5d 11037 top50table.html 41c2692012632c94e011fc2ca77b859ce10cea42 2254 D3/D39DA0E3 -- Jason Harris | NIC: JH329, PGP: This _is_ PGP-signed, isn't it? jharris@widomaker.com _|_ web: http://keyserver.kjsl.com/~jharris/ Got photons? (TM), (C) 2004 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: not available Url : /pipermail/attachments/20040422/8eae8e4a/attachment.bin From wzab at ise.pw.edu.pl Fri Apr 23 15:54:33 2004 From: wzab at ise.pw.edu.pl (Wojciech Zabolotny) Date: Sat Apr 24 10:08:32 2004 Subject: Encrypting for multiple recipients eith GnuPG Message-ID: <40892019.1080201@ise.pw.edu.pl> Hi All, I'd like to encrypt a data file for multiple recipients, using the GPG. It should be possible, just encrypt the file with a symmetric key, and then attach multiple copies of that key (with some random data added), each copy encrypted with public key of the other recipient (the random data should vary to make attack more difficult). However are there any options in GnuPG to automate that process? Meybe such a functionality should be added to the GnuPG? -- TIA & regards, Wojtek Zabolotny From atom-gpg at suspicious.org Sun Apr 25 10:24:17 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Sun Apr 25 10:21:42 2004 Subject: verifying signature after decryption Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 calling on our good friends alice and bob.... alice sends me (and only me) a message that's signed+encrypted. i need to show bob that this message is signed by alice. i can think of two ways to do this: 1) i give bob a copy of the encrypted message, my secret key and my password, so he can decrypt the message and see that it's signed by alice. of course, this would be dumb. 2) i can give bob a copy of the encrypted message, and the session key. i can instruct bob how to use the "--override-session-key" option. this requires that bob can understand and follow instructions. question: is there a way to extract the signed message, including the signature, from an encrypted message? in other words, can i take alice's signed+encrypted message, and pass it to bob either in plaintext or encrypted to bob's key, while still maintaining alice's signature over her message? of course, it must be ~possible~ to do this, but is there any ~practical~ way to do this? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "Vietnam was the first war ever fought without any censorship. Without censorship, things can get terribly confused in the public mind." -- General William Westmoreland -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCLdbYACgkQnCgLvz19QeP5DACgmwxijFAjJ8aqCKaXRwnrPgrp ryEAoJY6nLPBF8VW4m2zGRkMwngDX5Sp =rcfI -----END PGP SIGNATURE----- From malte.gell at gmx.de Sun Apr 25 15:49:13 2004 From: malte.gell at gmx.de (Malte Gell) Date: Sun Apr 25 15:48:28 2004 Subject: MUA option "encrypt to self" weakness in certain situations? Message-ID: <200404251549.31856.malte.gell@gmx.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Many MUA's or plugins allow to encrypt a message not only to the recipient's key, but to your own key to keep the sent message secret on your own machine. But, can this strategy be a possible weakness? When using such an option the message is now encrypted to 2 keys with the same session key: an attacker can now chose which key to attack to restore the message. Example: I have a 2048 bit ELG-E subkey and send a message to a person whose encryption key has 4096 bit, I have enabled the MUA's "encrypt to self" option. So the encrypted message contains the session key twice, encrypted to 2 keys. The overall security is now limited to the shorter key. The recipient is confident his long key protects the message to him, but actually the "encrypt to self" option limits this protection to the shorter key. This could mean that if someone (=recipient) uses a long key it may be rendered "useless" without intention if the sender has a short(er) key and uses such an "encrypt to self" option in his MUA. Is this thought correct? Malte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAi8HnGzg12gD8wBYRAlcvAJ9/7HAvEWl4lLj8njy4CE2gMi2vCQCgnjCL lDMZSLaKikKqPgh9fyr99sk= =z9xd -----END PGP SIGNATURE----- From sbutler at fchn.com Mon Apr 26 16:37:47 2004 From: sbutler at fchn.com (Steve Butler) Date: Mon Apr 26 16:35:58 2004 Subject: Encrypting for multiple recipients eith GnuPG Message-ID: <9A86613AB85FF346BB1321840DB42B4B046D43E2@jupiter.fchn.com> Why go to all that work? Just give gpg the list of recipients (multiple -r entries on the command line) before asking it to encrypt the file. gpg -r x -r y -r z -o output.gpg -e input.txt -----Original Message----- From: Wojciech Zabolotny [mailto:wzab@ise.pw.edu.pl] Sent: Friday, April 23, 2004 6:55 AM To: gnupg-users@gnupg.org Subject: Encrypting for multiple recipients eith GnuPG Hi All, I'd like to encrypt a data file for multiple recipients, using the GPG. It should be possible, just encrypt the file with a symmetric key, and then attach multiple copies of that key (with some random data added), each copy encrypted with public key of the other recipient (the random data should vary to make attack more difficult). However are there any options in GnuPG to automate that process? Meybe such a functionality should be added to the GnuPG? -- TIA & regards, Wojtek Zabolotny _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. From rasmith1959 at charter.net Mon Apr 26 17:39:56 2004 From: rasmith1959 at charter.net (Roy Smith) Date: Mon Apr 26 17:37:55 2004 Subject: Introduction Message-ID: <408D2D4C.4050406@charter.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi everyone! Just wanted to introduce myself and ask a few questions. First, is there a FAQ for this mailing list? Second: Is signing messages allowed, or even encrypting them using a group key? - -- Roy Smith I use the Mozilla Suite of Internet tools because I want to use software that complies with Web standards. See . -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) Comment: GnuPG 1.2.4 - GPGkeys 3.10 - Enigmail - Mozilla 1.6 iD8DBQFAjSyzIRbmvL066eYRApMkAJ4yJ7QDeu/H3N1pfpOKO1uqm2FNBQCgsMHQ P3mUalBpECJv4I9H+psBXRc= =Ba34 -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Mon Apr 26 17:35:12 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Mon Apr 26 17:38:05 2004 Subject: AES cipher support 1.0.6 and 1.0.7 In-Reply-To: <4DCE15B9C4E66F4CA967EBF64C53D64D67B326@bstn-exch1.forumsys.com> References: <4DCE15B9C4E66F4CA967EBF64C53D64D67B326@bstn-exch1.forumsys.com> Message-ID: <20040426153512.GC5711@jabberwocky.com> On Tue, Apr 20, 2004 at 01:45:59PM -0400, Hasnain Mujtaba wrote: > Hi, > > I encrypted a 2GB file using a DH/DSS 2048/1024 (AES-128) keypair using > a third party tool. > > I gave the encrypted file to GPG 1.0.6 but it failed to decrypt the > message with the following reason: > > gpg: protection algorithm 254 is not supported > gpg: encrypted with 2048-bit ELG-E key, ID 3F29C018, created 2004-04-16 > "GPG-DH-2048 " > gpg: public key decryption failed: unknown cipher algorithm > gpg: decryption failed: secret key not available > > Then I gave the encrypted file to GPG 1.0.7 and the decryption > succeeded. > > I'm wondering if this is due to AES support, or lack of, in the two > versions. No. It's due to the new secret key format in 1.0.7 and later. David From dshaw at jabberwocky.com Mon Apr 26 17:34:05 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Mon Apr 26 17:39:04 2004 Subject: verifying signature after decryption In-Reply-To: References: Message-ID: <20040426153405.GB5711@jabberwocky.com> On Sun, Apr 25, 2004 at 04:24:17AM -0400, Atom 'Smasher' wrote: > calling on our good friends alice and bob.... > > alice sends me (and only me) a message that's signed+encrypted. i need to > show bob that this message is signed by alice. > > i can think of two ways to do this: > 1) i give bob a copy of the encrypted message, my secret key and my > password, so he can decrypt the message and see that it's signed > by alice. of course, this would be dumb. > 2) i can give bob a copy of the encrypted message, and the session key. i > can instruct bob how to use the "--override-session-key" option. > this requires that bob can understand and follow instructions. > > question: is there a way to extract the signed message, including the > signature, from an encrypted message? > > in other words, can i take alice's signed+encrypted message, and pass it > to bob either in plaintext or encrypted to bob's key, while still > maintaining alice's signature over her message? > > of course, it must be ~possible~ to do this, but is there any ~practical~ > way to do this? There is nothing in the OpenPGP protocol that prevents this. In fact, it's quite easy to do. However, the code in GnuPG doesn't currently allow it (it's not a generally useful feature). David From dshaw at jabberwocky.com Mon Apr 26 17:28:51 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Mon Apr 26 17:40:02 2004 Subject: MUA option "encrypt to self" weakness in certain situations? In-Reply-To: <200404251549.31856.malte.gell@gmx.de> References: <200404251549.31856.malte.gell@gmx.de> Message-ID: <20040426152851.GA5711@jabberwocky.com> On Sun, Apr 25, 2004 at 03:49:13PM +0200, Malte Gell wrote: > Many MUA's or plugins allow to encrypt a message not only to the > recipient's key, but to your own key to keep the sent message secret on > your own machine. > > But, can this strategy be a possible weakness? When using such an option > the message is now encrypted to 2 keys with the same session key: an > attacker can now chose which key to attack to restore the message. > > Example: I have a 2048 bit ELG-E subkey and send a message to a person > whose encryption key has 4096 bit, I have enabled the MUA's "encrypt to > self" option. So the encrypted message contains the session key twice, > encrypted to 2 keys. The overall security is now limited to the shorter > key. The recipient is confident his long key protects the message to > him, but actually the "encrypt to self" option limits this protection > to the shorter key. > > This could mean that if someone (=recipient) uses a long key it may be > rendered "useless" without intention if the sender has a short(er) key > and uses such an "encrypt to self" option in his MUA. Is this thought > correct? It's sort of correct, but not really a problem in the real world. It is true that the message is only as "safe" as the smallest key it is encrypted to, but even the smaller key in your example is vastly stronger than most attackers. To put it another way, if your attacker can't climb more than 1000 feet, adding alligators and a moat to your 1001 foot wall doesn't matter much. It all depends on the attacker. David From atom-gpg at suspicious.org Mon Apr 26 17:49:25 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Mon Apr 26 17:56:55 2004 Subject: MUA option "encrypt to self" weakness in certain situations? In-Reply-To: <200404251549.31856.malte.gell@gmx.de> References: <200404251549.31856.malte.gell@gmx.de> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Sun, 25 Apr 2004, Malte Gell wrote: > This could mean that if someone (=recipient) uses a long key it may be > rendered "useless" without intention if the sender has a short(er) key > and uses such an "encrypt to self" option in his MUA. Is this thought > correct? ===================================== correct. this is the case *anytime* multiple recipients are specified, not only the case of encrypt-to-self. if an attacker wants to know the session key, they can attack the "weakest" public that a message is encrypted to. as i understand it, the public key implementations does not help an attacker compute another private key, used in the same message. part of the security of ANY encryption scheme depends not just on how secure/paranoid ~you~ are, but also how secure/paranoid is the person you're communicating with. a mistake on *their* end can cause *your* secrets to not be secret. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "When one tries to rise above Nature one is liable to fall below it." -- Sherlock Holmes (Arthur Conan Doyle) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCNL4wACgkQnCgLvz19QeNacgCeKEJp86xOl9o9RiV2/Zh1G7US mW8An2XUqNRe5PjyyWq57z/qUgJmQqLg =TeOg -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Mon Apr 26 18:54:34 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Mon Apr 26 18:51:47 2004 Subject: block-cipher weakness? - was: Re: twofish keysize In-Reply-To: <6.0.3.0.2.20040422084123.027e0720@localhost> References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> <6.0.3.0.2.20040420144800.027dd950@localhost> <87zn9566r8.fsf@vigenere.g10code.de> <6.0.3.0.2.20040422084123.027e0720@localhost> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thu, 22 Apr 2004, Per Tunedal Casual wrote: > On average 32 GB is the limit i.e. the probability is 50 % that an > adversary finds two identical blocks. But you might have bad luck and > he might find two identical blocks in somewhat smaller files. =============================== can someone explain to me why/how it's a vulnerability if an attacker finds more than one block of cipher text that are the same? would there be a benefit of using a stream cipher for large files? is there a reason why OpenPGP doesn't specify any stream ciphers? thanks... ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- _ _(_)_ wWWWw _ @@@@ (_)@(_) vVVVv _ @@@@ (___) _(_)_ @@()@@ wWWWw (_)\ (___) _(_)_ @@()@@ Y (_)@(_) @@@@ (___) `|/ Y (_)@(_) @@@@ \|/ (_)\ / Y \| \|/ /(_) \| |/ | \ | \ |/ | / \ | / \|/ |/ \| \|/ \\|// \\|/// \\\|//\\\|/// \|/// \\\|// \\|// \\\|// ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCNPs8ACgkQnCgLvz19QeNY7gCdExdCYs5LFApqwkVEvOA89Tv4 JjYAnjMB7gTyMSAQFrkEeQM7JrdAtj6Q =FngB -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Mon Apr 26 18:57:22 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Mon Apr 26 18:54:33 2004 Subject: verifying signature after decryption In-Reply-To: <20040426153405.GB5711@jabberwocky.com> References: <20040426153405.GB5711@jabberwocky.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 26 Apr 2004, David Shaw wrote: > > question: is there a way to extract the signed message, including the > > signature, from an encrypted message? > > There is nothing in the OpenPGP protocol that prevents this. In fact, > it's quite easy to do. However, the code in GnuPG doesn't currently > allow it (it's not a generally useful feature). ==================================== is there any way to do this without writing my own code? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "The unleashed power of the atom has changed everything save our modes of thinking and we thus drift toward unparalleled catastrophe." -- Albert Einstein -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCNP3YACgkQnCgLvz19QeP1OACfYaykatP8eGjq9BBhj4zAJfE+ 2bYAnjrMSgYFXBZrPvL8IeP1AztCkZjk =bauV -----END PGP SIGNATURE----- From dshaw at jabberwocky.com Mon Apr 26 19:29:25 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Mon Apr 26 19:26:39 2004 Subject: verifying signature after decryption In-Reply-To: References: <20040426153405.GB5711@jabberwocky.com> Message-ID: <20040426172925.GE5711@jabberwocky.com> On Mon, Apr 26, 2004 at 12:57:22PM -0400, Atom 'Smasher' wrote: > On Mon, 26 Apr 2004, David Shaw wrote: > > > question: is there a way to extract the signed message, including the > > > signature, from an encrypted message? > > > > There is nothing in the OpenPGP protocol that prevents this. In fact, > > it's quite easy to do. However, the code in GnuPG doesn't currently > > allow it (it's not a generally useful feature). > ==================================== > > is there any way to do this without writing my own code? Nothing I can think of offhand, sorry. David From jerry.windrel at verizon.net Mon Apr 26 19:31:06 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Mon Apr 26 19:26:51 2004 Subject: verifying signature after decryption References: <20040426153405.GB5711@jabberwocky.com> Message-ID: <017401c42bb4$41edfd00$6401a8c0@Windows> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Someone mentioned on this list a while back that GPG supports extracting the symmetric session key used for a message. If you provide this session key to someone, they can use GPG to decrypt that particular message only, including the signature. I'm not sure what the exact commands are. - ----- Original Message ----- From: "Atom 'Smasher'" To: Sent: Monday, April 26, 2004 12:57 PM Subject: Re: verifying signature after decryption > On Mon, 26 Apr 2004, David Shaw wrote: > > > question: is there a way to extract the signed message, > > > including the signature, from an encrypted message? > > > > There is nothing in the OpenPGP protocol that prevents this. In > > fact, it's quite easy to do. However, the code in GnuPG doesn't > > currently allow it (it's not a generally useful feature). > ==================================== > > is there any way to do this without writing my own code? -----BEGIN PGP SIGNATURE----- Version: PGP 8.0.3 - not licensed for commercial use: www.pgp.com iQA/AwUBQI1HRYlVuABWWiqVEQI5tQCgmQlksMq0KfVA9vxbzFhH1XJLbTQAnRc1 a/0i//VEGX1fwjqHmPyV635T =Vpkg -----END PGP SIGNATURE----- From atom-gpg at suspicious.org Mon Apr 26 19:38:13 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Mon Apr 26 19:35:31 2004 Subject: verifying signature after decryption In-Reply-To: <017401c42bb4$41edfd00$6401a8c0@Windows> References: <20040426153405.GB5711@jabberwocky.com> <017401c42bb4$41edfd00$6401a8c0@Windows> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Someone mentioned on this list a while back that GPG supports > extracting the symmetric session key used for a message. If you > provide this session key to someone, they can use GPG to decrypt that > particular message only, including the signature. I'm not sure what > the exact commands are. ==================================== the session key is extracted with --show-session-key and used with - --override-session-key. this requires some extra work on the recipients end. ideally, i'd like to make it idiot-resistant enough that it would automagically work, like in someones MUA. some people are deathly afraid of the command line. some people can be taught how to use pgp in their MUA, but might not trust instructions that a geek tells them to run on the command line. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "We do many things at the federal level that would be considered dishonest and illegal if done in the private sector." -- Ronald Reagan -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCNSQoACgkQnCgLvz19QeNOxACePmL2zW6bjhp41NxxcV3Qm9tt 1pcAnR7/FnFI0h6fvbPEPYfDPlu9r9PR =3p7R -----END PGP SIGNATURE----- From linux at codehelp.co.uk Mon Apr 26 19:43:12 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Mon Apr 26 19:39:30 2004 Subject: Introduction In-Reply-To: <408D2D4C.4050406@charter.net> References: <408D2D4C.4050406@charter.net> Message-ID: <200404261843.18776.linux@codehelp.co.uk> On Monday 26 April 2004 4:39, Roy Smith wrote: > Hi everyone! > > Just wanted to introduce myself and ask a few questions. First, is > there a FAQ for this mailing list? No, but there's an archive and plenty of documentation on the website: http://www.gnupg.org/(en)/documentation/index.html http://www.gnupg.org/(en)/documentation/faqs.html http://lists.gnupg.org/pipermail/gnupg-users/ > Second: Is signing messages allowed, Yes. > or even encrypting them using a > group key? There is no single group key for all users, perhaps because new users are using new keys which, by nature, are untrusted. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040426/c343b310/attachment.bin From vedaal at hush.com Mon Apr 26 19:48:41 2004 From: vedaal at hush.com (vedaal@hush.com) Date: Mon Apr 26 19:45:50 2004 Subject: verifying signature after decryption Message-ID: <200404261748.i3QHmfhq042513@mailserver2.hushmail.com> >Message: 5 >Date: Sun, 25 Apr 2004 04:24:17 -0400 (EDT) >From: Atom 'Smasher' >Subject: verifying signature after decryption >alice sends me (and only me) a message that's signed+encrypted. >i need to >show bob that this message is signed by alice. [...] >question: is there a way to extract the signed message, including >the >signature, from an encrypted message? > >in other words, can i take alice's signed+encrypted message, and >pass it >to bob either in plaintext or encrypted to bob's key, while still >maintaining alice's signature over her message? > >of course, it must be ~possible~ to do this, but is there any ~practical~ >way to do this? yes, but not (yet) with gnupg, but very easily, for 'any' key type/signature hash/ or symmetric algorithm, by using a pgp command-line program called 'filecrypt' {the free demo download is all that is needed} detailed example description is here: http://www.angelfire.com/pr/pgpf/fcs.html 'filecrypt' is here: http://www.veridis.com/openpgp/en/index.asp#.html hth, vedaal Concerned about your privacy? Follow this link to get FREE encrypted email: https://www.hushmail.com/?l=2 Free, ultra-private instant messaging with Hush Messenger https://www.hushmail.com/services.php?subloc=messenger&l=434 Promote security and make money with the Hushmail Affiliate Program: https://www.hushmail.com/about.php?subloc=affiliate&l=427 From jerry.windrel at verizon.net Mon Apr 26 20:04:59 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Mon Apr 26 20:00:42 2004 Subject: group key? References: <408D2D4C.4050406@charter.net> <200404261843.18776.linux@codehelp.co.uk> Message-ID: <01ae01c42bb8$fd7003d0$6401a8c0@Windows> Can you please expand on the notion of what you mean by a "group key"? Is such a thing commonly used and in what situations? Just to give you a reference point for possible correction... what I'm imagining you mean is that a key pair is generated and assosicated, not with a person as is customery, but rather with a group. The private key (and any passphrase) would then be distributed to all members of the group (which would be quite an unorthodox practice, in terms of PGP). Then, all messages to the group would be encrypted to that "group key". The ramifications of this would be: the group private key would be only as secure as the weakest practices of any of the group members; you would be able to add but not remove members (unless you generated a new group key). Please clarify and correct as necessary ------------------------------------------------------------------------ There is no single group key for all users, perhaps because new users are using new keys which, by nature, are untrusted. -- Neil Williams From atom-gpg at suspicious.org Mon Apr 26 20:08:59 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Mon Apr 26 20:06:15 2004 Subject: group key? In-Reply-To: <01ae01c42bb8$fd7003d0$6401a8c0@Windows> References: <408D2D4C.4050406@charter.net> <200404261843.18776.linux@codehelp.co.uk> <01ae01c42bb8$fd7003d0$6401a8c0@Windows> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 if you want to find some people to practice sending/receiving encrypted messages with, check out - http://groups.yahoo.com/group/PGP-Basics/ ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "The whole aim of practical politics is to keep the populace alarmed (and hence clamorous to be led to safety) by menacing it with an endless series of hobgoblins, all of them imaginary." -- H.L. Mencken -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCNUEAACgkQnCgLvz19QeO/TQCZASkG2Ou+4GKZEN3FzkzYYgTR 0QIAoJzkqCFgbDnUq3rs8T1Q/JTtY5Vz =6e6H -----END PGP SIGNATURE----- From linux at codehelp.co.uk Mon Apr 26 22:32:00 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Mon Apr 26 22:28:18 2004 Subject: group key? In-Reply-To: <01ae01c42bb8$fd7003d0$6401a8c0@Windows> References: <408D2D4C.4050406@charter.net> <200404261843.18776.linux@codehelp.co.uk> <01ae01c42bb8$fd7003d0$6401a8c0@Windows> Message-ID: <200404262132.04808.linux@codehelp.co.uk> On Monday 26 April 2004 7:04, Jerry Windrel wrote: To clarify: Personally, I don't see the point of a group key for a mailing list - it's generally unworkable, more than a little pointless and unnecessarily exclusive. Lists are meant to be open and openly archived. However, the question of a group key is wider than just a list. There have been discussions about corporate keys on this list before, a group key is a similar idea. > imagining you mean is that a key pair is generated and assosicated, not > with a person as is customery, but rather with a group. The private key > (and any passphrase) would then be distributed to all members of the group > (which would be quite an unorthodox practice, in terms of PGP). Then, all > messages to the group would be encrypted to that "group key". Or, more simply, a single database would keep a tally of current keyids and encrypt all messages to a vast list of keys using the same session key. Not surprising that this doesn't sound good. At least it removes the need for a distributed private key AND it allows for user additions and removals. You would be able to exclude old keys from decrypting old messages and allow new keys to decrypt old messages, however, re-encrypting archived messages to the amended group would be too burdensome, (being recursive and all). This would, in effect, simply be a group of keys (or key group), not a group key. GnuPG can already do this and has group functionality in place. I don't think the group was ever considered to be of sufficient size for an entire list/company! > The ramifications of this would be: the group private key would be only as > secure as the weakest practices of any of the group members; you would be As has been discussed elsewhere on this list today, encrypting to more than one person/key always renders the entire chain vulnerable to the weakest link - it has always been this way, no matter what tool is used to create any chain. > able to add but not remove members (unless you generated a new group key). > > Please clarify and correct as necessary The problems and limitations of a group key are not technological but practical and social. It echoes other comments about using a key on a server to sign or decrypt messages automatically - the passphrase is either entirely absent or stored in a file on the server. Neither are particularly appealing because you substitute the security of GnuPG for the security of the server alone instead of adding. (If anyone can crack the server, they can obtain the private key + passphrase, if any.) More feasible were ideas around corporate keys that had a single main key that had a private key only accessible to a named high-up bod. That key would sign subordinate keys and revoke signatures if the subordinate left the company. At no time would more than one individual ever have access to more than one private key in the corporate keyring and no one private key would ever available to more than one person. This is workable but is it worthwhile? The problem remains of controlling the signatures and what happens when the underlings get together and sign each other's keys thereby becoming trusted without the boss' signature? That's a procedural / management / policy issue. It also didn't address the issue of encrypting to everyone, just trusting the signatures / keys. Encryption would still use some form of key group - which leaves it right back at square 1. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040426/38d6dd09/attachment.bin From dshaw at jabberwocky.com Tue Apr 27 01:57:26 2004 From: dshaw at jabberwocky.com (David Shaw) Date: Tue Apr 27 01:54:44 2004 Subject: RSA Encryption / exchange with PGP In-Reply-To: References: Message-ID: <20040426235725.GB10272@jabberwocky.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tue, Apr 20, 2004 at 12:15:58PM -0500, David Hill wrote: > I am sending the trading partner a new text file encrypted with the following options: > gpg --rfc1991 > --force-v3-sigs > --compress-algo 1 > --recipient "THEM" > --recipient "US" > --output "readme.gpg" > -es "readme.txt" > > In previous attempts I didn't use the first three commands, I am > hoping these will help our trading partner decrypt, but I don't have > high hopes of success. Unfortunately, PGP 6 does not follow the PGP specification very closely. GnuPG has code to work around this. Try setting the "--pgp6" option: gpg -r THEM -r US --pgp6 --output readme.gpg -es readme.txt > Even if by some miracle that worked, we still can't decrypt their > files. Do we need to use different keys or protocols in order to > interoperate? This is a sample decryption session: > > C:\gnupg>gpg --verbose --decrypt "install.log.pgp" > gpg: public key is F58C3B48 > gpg: using secondary key F58C3B48 instead of primary key 4D1D68AC > > You need a passphrase to unlock the secret key for > user: "Iowa Student Loan Liquidity Corp. (iLink Dev) " > gpg: using secondary key F58C3B48 instead of primary key 4D1D68AC > 1024-bit ELG-E key, ID F58C3B48, created 2004-04-19 (main key ID 4D1D68AC) > > gpg: encrypted with 1024-bit ELG-E key, ID F58C3B48, created 2004-04-19 > "Iowa Student Loan Liquidity Corp. (iLink Dev) " > gpg: CAST5 encrypted data > gpg: block_filter 024BCB68: read error (size=7302,a->size=536874118) > gpg: WARNING: message was not integrity protected > gpg: block_filter: pending bytes! How did you get the encrypted file (install.log.pgp)? More than anything else this looks like the encrypted file was transferred to you via text mode instead of binary FTP and so the file was corrupted before you got it. David -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.3.6-cvs (GNU/Linux) Comment: Key available at http://www.jabberwocky.com/david/keys.asc iHEEARECADEFAkCNoeUqGGh0dHA6Ly93d3cuamFiYmVyd29ja3kuY29tL2Rhdmlk L2tleXMuYXNjAAoJEOJmXIdJ4cvJOakAnRPS+0SSC0GhVslMCg9w+WlLarZ9AJ9D gfcdosRoumYm87KEsu/OlibU0w== =iARg -----END PGP SIGNATURE----- From dasymond at it.usyd.edu.au Tue Apr 27 02:34:09 2004 From: dasymond at it.usyd.edu.au (Dave Symonds) Date: Tue Apr 27 02:31:20 2004 Subject: Remote signing In-Reply-To: <20040420162145.560f6feb.ian_bruce@fastmail.fm> References: <20040419234757.GA23117@it.usyd.edu.au> <4084D75E.8070503@computing-services.oxford.ac.uk> <20040420084111.GA28365@Update.UU.SE> <4084F3B4.2030507@computing-services.oxford.ac.uk> <20040420121005.GB10970@it.usyd.edu.au> <40851773.20005@computing-services.oxford.ac.uk> <20040420123950.GA4404@it.usyd.edu.au> <20040420162145.560f6feb.ian_bruce@fastmail.fm> Message-ID: <20040427003409.GB19564@it.usyd.edu.au> On Tue, Apr 20, 2004 at 04:21:45PM -0700, Ian Bruce wrote: > It appears to me that there is some mutual misunderstanding involved in > this conversation. As far as I understand, the situation is as follows: [snip] > Is that about right? Yep, that all looks correct. > Assuming that it is, I have to ask why you wouldn't just run the mail > client on the laptop and avoid all the extra complication. You can then > either tunnel the IMAP and SMTP connections over SSH or just use > IMAP/SSL and SMTP/SSL directly. That is a good point. I thought about that, but the three factors that stop me doing that are: (1) The laptop is a (primarily) WinXP machine, which is a PITA to run mutt on to my satisfaction. (2) I'm not always using the laptop to read my mail, and I want to avoid the trouble of keeping mail settings (of which I have quite a lot) consistent. (3) IMAP/SMTP access to the mail server is not possible (firewalled) directly, so it would have to be tunnelled. Combined with (2), it seems a *lot* easier (except for the crypto stuff) to just ssh to the Sun and do all the mail stuff there. I'm going to look into (1), but surely this kind of setup has been faced by people before. Anyone seen something like this done? Dave. -- David Symonds USyd::SITRG::PhD Student http://www.it.usyd.edu.au/~dasymond/ From greg at turnstep.com Tue Apr 27 04:27:28 2004 From: greg at turnstep.com (Greg Sabino Mullane) Date: Tue Apr 27 04:25:05 2004 Subject: Remote signing In-Reply-To: <20040419234757.GA23117@it.usyd.edu.au> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > However, I don't want to entrust my GPG private key(s) to that system, and > would prefer to keep them on my laptop (or a USB key). > ... > This seems reasonably secure to me, and shouldn't be too difficult for > me to code up. Does this sound like a reasonable approach? Has it been > done before? The only real weak point is that fact that your key is on a networked machine. Your best bet is to keep your signing box off of a network, and use some other medium to transfer the plaintext to this box. You then sign the message, and use the medium to return the signed message. Good candidates for transferral are USB keys, infrared, and the good old floppy disk. It can be a real pain to go through all that for each message, but a rule of thumb is that the amount of automation is inversely proportional to the amount of security. Since encryption is not important in your case, another option is to do make sure that only plain text can transfer between the two machines, and no logging in is possible. This could be done with a simple cgi script on the signing box, and keeping only port 80 open. It's sniffable, but you don't care because all you are revealing is the plaintext and the signed plaintext. - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200404262222 -----BEGIN PGP SIGNATURE----- iD8DBQFAjcVlvJuQZxSWSsgRAqg6AJ42boxG+yX+yezVMc89jGJlCXn6bQCdEC2R 4TkwctwE9k1RjZJviGE5Up0= =RMte -----END PGP SIGNATURE----- From Holger.Sesterhenn at smgwtest.aachen.utimaco.de Tue Apr 27 08:41:22 2004 From: Holger.Sesterhenn at smgwtest.aachen.utimaco.de (Holger Sesterhenn) Date: Tue Apr 27 08:39:01 2004 Subject: verifying signature after decryption In-Reply-To: <20040426172925.GE5711@jabberwocky.com> References: <20040426153405.GB5711@jabberwocky.com> <20040426172925.GE5711@jabberwocky.com> Message-ID: <408E0092.8020308@smgwtest.aachen.utimaco.de> Hi, >>>>question: is there a way to extract the signed message, including the >>>>signature, from an encrypted message? [...] >>is there any way to do this without writing my own code? > > > Nothing I can think of offhand, sorry. If I remember correctly there was a feature request on Gnupg-devel at the end of february: --unwrap This would decrypt the message but leave the signed data with the signature packet untouched. Of couse you need again an OpenPGP client to verify the signature and get the data. -- Best Regards, Holger Sesterhenn --- Internet http://www.utimaco.com From avbidder at fortytwo.ch Tue Apr 27 09:02:04 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Tue Apr 27 08:59:20 2004 Subject: trust management script In-Reply-To: <200404212334.24999.linux@codehelp.co.uk> References: <200404212334.24999.linux@codehelp.co.uk> Message-ID: <200404270902.10391@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday 22 April 2004 00.34, Neil Williams wrote: Hi, Thanks for that script, this is certainly useful. > I worked on the idea that if I > hadn't verified the key, I probably had never met the person behind > the key and therefore I was unlikely to be able to make much of a > decision on how carefully that person would verify someone else's > key. (Not having had the chance to see them verify my own key.) I think this assumption does not always hold. I have many keys set to full trust in my keyring - these are mostly keys from prominent Debian developers. I trust them because I feel I can judge from seeing their email on the mailing list how competent they are - certainly better than I can judge from seeing them once at a keysigning party. Trust is always a personal thing, and everybody needs to find their own working compromise between usability and paranoia. In my case, the fact that I rarely need to transmit really secret data helps - encrypting ordinary email by default is different from encrypting account passwords etc.; in the latter case I take an extra look at the trust path. greetings - -- vbi - -- The content of this message may or may not reflect the opinion of me, my employer, my girlfriend, my cat or anybody else, regardless of the fact whether such an employer, girlfriend, cat, or anybody else exists. I (or my employer, girlfriend, cat or whoever) disclaim any legal obligations resulting from the above message. You, as the reader of this message, may or may not have the permission to redistribute this message as a whole or in parts, verbatim or in modified form, or to distribute any message at all. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkCOBXFgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6imoAn0wQoGAOwbQ57cQtN7hqrd8t U4feAJ0U9QusPDlKXRyxRw9Tl/BknoNH5g== =w1Xt -----END PGP SIGNATURE----- From wk at gnupg.org Tue Apr 27 11:30:32 2004 From: wk at gnupg.org (Werner Koch) Date: Tue Apr 27 11:15:50 2004 Subject: twofish keysize In-Reply-To: <200404220029.41619.malte.gell@gmx.de> (Malte Gell's message of "Thu, 22 Apr 2004 00:28:47 +0200") References: <200404162004.13385.jluehr@gmx.net> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> <200404220029.41619.malte.gell@gmx.de> Message-ID: <878yghwyx3.fsf@vigenere.g10code.de> On Thu, 22 Apr 2004 00:28:47 +0200, Malte Gell said: > Would it be much hassle to introduce 2Fish-128,196,256, just according > to AES-128,196,256 to adress this? Since many encryption keys are It is not possible because OpenPGP does only specify Twofish 256. > situation, on the other hand PGP/GnuPG amply uses system's entropy for > AES-256/Twofish(256)...(sorry for key length discussion now on the > symm. side,I'll immediately duck and cover;-) Ask the folks over at NAI (most likely the marketing dept.) why they used the 256 bit version of Twofish. Salam-Shalom, Werner From wk at gnupg.org Tue Apr 27 11:34:07 2004 From: wk at gnupg.org (Werner Koch) Date: Tue Apr 27 11:16:02 2004 Subject: verifying signature after decryption In-Reply-To: (atom-gpg@suspicious.org's message of "Mon, 26 Apr 2004 12:57:22 -0400 (EDT)") References: <20040426153405.GB5711@jabberwocky.com> Message-ID: <874qr5wyr4.fsf@vigenere.g10code.de> On Mon, 26 Apr 2004 12:57:22 -0400 (EDT), Atom 'Smasher' said: > is there any way to do this without writing my own code? Yes: Use PGP/MIME. The standard way there is to sign and encrypt using separate MIME containers. This makes it very easy to strip of the encryption layer. Shalom-Salam, Werner From geert.hendrickx at ua.ac.be Tue Apr 27 10:22:02 2004 From: geert.hendrickx at ua.ac.be (Geert Hendrickx) Date: Tue Apr 27 13:05:46 2004 Subject: can you deny you sent a signed e-mail? Message-ID: <20040427082201.GB27223@lori.mine.nu> Hi, GPG gives us a method for signing our e-mails so the addressee can be sure it was sent by yourself. But what about the inverse question: can you deny that you sent an e-mail which has been signed using you secret key? Not that I need this :-) it's just out of curiousity. GH From thomas at northernsecurity.net Tue Apr 27 13:41:59 2004 From: thomas at northernsecurity.net (Thomas =?iso-8859-1?Q?Sj=F6gren?=) Date: Tue Apr 27 13:39:40 2004 Subject: can you deny you sent a signed e-mail? In-Reply-To: <20040427082201.GB27223@lori.mine.nu> References: <20040427082201.GB27223@lori.mine.nu> Message-ID: <20040427114159.GA10394@northernsecurity.net> On Tue, Apr 27, 2004 at 10:22:02AM +0200, Geert Hendrickx wrote: > But what about the inverse question: can you deny that you sent an > e-mail which has been signed using you secret key? Sure you can, but who would believe you? If you send a signed mail that you dont want to be blamed for later (sending company info to a competitor for example), revoke you key and tell everyone that someone broke into your machine. /Thomas -- == thomas@northernsecurity.net | thomas@se.linux.org == Encrypted e-mails preferred | GPG KeyID: 114AA85C -- -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 481 bytes Desc: Digital signature Url : /pipermail/attachments/20040427/076fda78/attachment.bin From fw at deneb.enyo.de Tue Apr 27 14:20:41 2004 From: fw at deneb.enyo.de (Florian Weimer) Date: Tue Apr 27 14:17:52 2004 Subject: can you deny you sent a signed e-mail? In-Reply-To: <20040427082201.GB27223@lori.mine.nu> (Geert Hendrickx's message of "Tue, 27 Apr 2004 10:22:02 +0200") References: <20040427082201.GB27223@lori.mine.nu> Message-ID: <874qr5lihy.fsf@deneb.enyo.de> Geert Hendrickx writes: > But what about the inverse question: can you deny that you sent an > e-mail which has been signed using you secret key? Yes, you leak your secret key and claim that someone else has used it. -- Current mail filters: many dial-up/DSL/cable modem hosts, and the following domains: atlas.cz, bigpond.com, di-ve.com, hotmail.com, netscape.net, postino.it, tiscali.co.uk, tiscali.cz, tiscali.it, voila.fr. From atom-gpg at suspicious.org Tue Apr 27 15:14:36 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Tue Apr 27 15:11:54 2004 Subject: verifying signature after decryption In-Reply-To: <408E0092.8020308@smgwtest.aachen.utimaco.de> References: <20040426153405.GB5711@jabberwocky.com> <20040426172925.GE5711@jabberwocky.com> <408E0092.8020308@smgwtest.aachen.utimaco.de> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tue, 27 Apr 2004, Holger Sesterhenn wrote: > >>is there any way to do this without writing my own code? > > If I remember correctly there was a feature request on Gnupg-devel at > the end of february: --unwrap > > This would decrypt the message but leave the signed data with the > signature packet untouched. Of couse you need again an OpenPGP client to > verify the signature and get the data. ========================= does anyone know if that will make it into gpg 1.4.x? ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "The hottest places in hell are reserved for those who in times of great moral crises maintain their neutrality." -- Dante Aleghieri (1265-1321) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCOXMIACgkQnCgLvz19QeMU1ACeJcwF2UrNgg1FrXqMp5DZYJHN K74An2ogSDrsKuubfeJSRXts2uFJ307g =w4V6 -----END PGP SIGNATURE----- From DHill at StudentLoan.org Tue Apr 27 16:35:09 2004 From: DHill at StudentLoan.org (David Hill) Date: Tue Apr 27 16:32:47 2004 Subject: RSA Encryption / exchange with PGP Message-ID: Actually we got the two to interoperate nicely, they are using PGP version 6.5 freeware, RSA legacy keys, and encrypting with IDEA. We used the PGP product to generate the newer RSA 2048 keys and imported those keys into GnuPG. That worked a lot better for allowing them to decrypt our traffic, the DSA/ElGamel gave them trouble decrypting our files. The real problem was that they were unwilling to change their preferred encryption method from IDEA to TripleDES, so we downloaded and tested the IDEA dll from ftp://ftp.gnupg.dk/pub/contrib-dk/ and that worked perfectly. Then I went to the www.mediacrypt.com site and found I could license the IDEA algorithm for 15 euros per seat, and presto, the rouge IDEA dll was legal. -----Original Message----- From: David Shaw [mailto:dshaw@jabberwocky.com] Sent: Monday, April 26, 2004 6:57 PM To: David Hill Cc: gnupg-users@gnupg.org Subject: Re: RSA Encryption / exchange with PGP -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tue, Apr 20, 2004 at 12:15:58PM -0500, David Hill wrote: > I am sending the trading partner a new text file encrypted with the following options: > gpg --rfc1991 > --force-v3-sigs > --compress-algo 1 > --recipient "THEM" > --recipient "US" > --output "readme.gpg" > -es "readme.txt" > > In previous attempts I didn't use the first three commands, I am > hoping these will help our trading partner decrypt, but I don't have > high hopes of success. Unfortunately, PGP 6 does not follow the PGP specification very closely. GnuPG has code to work around this. Try setting the "--pgp6" option: gpg -r THEM -r US --pgp6 --output readme.gpg -es readme.txt > Even if by some miracle that worked, we still can't decrypt their > files. Do we need to use different keys or protocols in order to > interoperate? This is a sample decryption session: > > C:\gnupg>gpg --verbose --decrypt "install.log.pgp" > gpg: public key is F58C3B48 > gpg: using secondary key F58C3B48 instead of primary key 4D1D68AC > > You need a passphrase to unlock the secret key for > user: "Iowa Student Loan Liquidity Corp. (iLink Dev) " > gpg: using secondary key F58C3B48 instead of primary key 4D1D68AC > 1024-bit ELG-E key, ID F58C3B48, created 2004-04-19 (main key ID 4D1D68AC) > > gpg: encrypted with 1024-bit ELG-E key, ID F58C3B48, created 2004-04-19 > "Iowa Student Loan Liquidity Corp. (iLink Dev) " > gpg: CAST5 encrypted data > gpg: block_filter 024BCB68: read error (size=7302,a->size=536874118) > gpg: WARNING: message was not integrity protected > gpg: block_filter: pending bytes! How did you get the encrypted file (install.log.pgp)? More than anything else this looks like the encrypted file was transferred to you via text mode instead of binary FTP and so the file was corrupted before you got it. David -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.3.6-cvs (GNU/Linux) Comment: Key available at http://www.jabberwocky.com/david/keys.asc iHEEARECADEFAkCNoeUqGGh0dHA6Ly93d3cuamFiYmVyd29ja3kuY29tL2Rhdmlk L2tleXMuYXNjAAoJEOJmXIdJ4cvJOakAnRPS+0SSC0GhVslMCg9w+WlLarZ9AJ9D gfcdosRoumYm87KEsu/OlibU0w== =iARg -----END PGP SIGNATURE----- From rainman at hod.aarg.net Tue Apr 27 18:43:23 2004 From: rainman at hod.aarg.net (Denis Green) Date: Tue Apr 27 18:40:36 2004 Subject: Moving from PGP to GnuPG and other questions ... Message-ID: <200404271643.i3RGhNBp027790@www.aarg.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello Group I'm a new user here : 1/ How do I choose "No mail", web only option - for this list ? 2/ What are the most recommended Windows front ends for GnuPG (Windows XP user) 3/ Is there a way to use the IDEA algorithm with GnuPG - if so how ? [wish to be fully compatible with PGP users ] 4/ I presently use Pegasus for Windows as my e mail client. Are there better free e mail clients - that integrate with GPG especially ? 5/ I'm presently forced to use PGP mainly for Nym and Re-mailer messages. Presently I'm using Jack B Nymble as my mail client for Nym related mail. JBN seems to integrate ONLY with PGP. Are there equivalent mail clients (that can generate re-mailer and nym messages) which interface with GnuPG 5.1/ If anyone is using Jack B Nymble with GnuPg please let me know - any tips / pointers are most welcome 6/ any GnuPG equivalent for the PGPDisk ? un related / general - --------------------------- 7/ Any free English language Dictionaries for English meanings / spelling / pronunciation etc. ? (not spell checker - I have Aspell already) 8/ Free stuff preferred. If not available please recommend paid versions as well Sorry if some of these questions are repeated May thanks in advance DG send a blank mail to rainmankey@eml.cc for my public key -----BEGIN PGP SIGNATURE----- Version: 6.5.8ckt http://www.ipgpp.com/ iQA/AwUBQI49Uh1/6GqH2TSrEQJ9OwCg+TnIPNeO9FrJ6Bdon09qNVuR2isAn1Zj QQE+Htpp3VWZVbF8n3A7YOyh =o9gP -----END PGP SIGNATURE----- From SGates at olbh.com Tue Apr 27 16:43:09 2004 From: SGates at olbh.com (Gates, Scott) Date: Tue Apr 27 18:48:24 2004 Subject: can you deny you sent a signed e-mail? Message-ID: <4E5A2C0A6997D811B58800A0C9C551501A30FF@ashland01msx> Unless you gave out the passphrase that protects your secret key--which is almost equal to giving away ones ID, house keys, car keys, and bank account numbers, then griping you got 'robbed'. . . . I'm not sure how one COULD deny. -----Original Message----- From: Geert Hendrickx [mailto:geert.hendrickx@ua.ac.be] Sent: Tuesday, April 27, 2004 4:22 AM To: gnupg-users@gnupg.org Subject: can you deny you sent a signed e-mail? Hi, GPG gives us a method for signing our e-mails so the addressee can be sure it was sent by yourself. But what about the inverse question: can you deny that you sent an e-mail which has been signed using you secret key? Not that I need this :-) it's just out of curiousity. GH _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users From wk at gnupg.org Tue Apr 27 18:21:21 2004 From: wk at gnupg.org (Werner Koch) Date: Tue Apr 27 18:55:23 2004 Subject: RSA Encryption / exchange with PGP In-Reply-To: (David Hill's message of "Tue, 27 Apr 2004 09:35:09 -0500") References: Message-ID: <878yghqtmm.fsf@vigenere.g10code.de> On Tue, 27 Apr 2004 09:35:09 -0500, David Hill said: > ftp://ftp.gnupg.dk/pub/contrib-dk/ and that worked perfectly. Then I > went to the www.mediacrypt.com site and found I could license the > IDEA algorithm for 15 euros per seat, and presto, the rouge IDEA dll Just for the record: The primary reason GnuPG was written is to avoid exactly that patent claim. I am obviously not very lucky to hear that money is put into the chests of the patent holder by means of software written to avoid that. There is no technical reason for IDEA and actually it is a weaker cipher than the modern ones (i.e. AES or Twofish) at least in terms of block size. http://swpat.ffii.org, http://noepatents.org Werner From jerry.windrel at verizon.net Tue Apr 27 19:16:25 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Tue Apr 27 19:12:21 2004 Subject: verifying signature after decryption References: <20040426153405.GB5711@jabberwocky.com><20040426172925.GE5711@jabberwocky.com><408E0092.8020308@smgwtest.aachen.utimaco.de> Message-ID: <008d01c42c7b$5f1d1ec0$6401a8c0@Windows> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 - ----- Original Message ----- From: "Atom 'Smasher'" To: Sent: Tuesday, April 27, 2004 9:14 AM Subject: Re: verifying signature after decryption > On Tue, 27 Apr 2004, Holger Sesterhenn wrote: > > >>is there any way to do this without writing my own code? > > > > If I remember correctly there was a feature request on > > Gnupg-devel at the end of february: --unwrap > > > > This would decrypt the message but leave the signed data with the > > signature packet untouched. Of couse you need again an OpenPGP > > client to verify the signature and get the data. Could there be a feature that would be similar to the "unwrap" feature, but that would give you just a standard signature, without it being in a packet with the data? Then you could decrypt the data yourself and send its cleartext, along with the signature, to others for verification. They could verify the signature as they would for any clear-signed message, as if it had never been enrypted. -----BEGIN PGP SIGNATURE----- Version: PGP 8.0.3 - not licensed for commercial use: www.pgp.com iQA/AwUBQI6VVIlVuABWWiqVEQKQTgCeKwfDZwx8KZbxDdaIG3uWpjhfpFgAoOrj 7YnlNr3hxCnHpUYv22+U7Qnm =9Umf -----END PGP SIGNATURE----- From jerry.windrel at verizon.net Tue Apr 27 19:37:03 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Tue Apr 27 19:32:58 2004 Subject: can you deny you sent a signed e-mail? References: <20040427082201.GB27223@lori.mine.nu> <20040427114159.GA10394@northernsecurity.net> Message-ID: <00c101c42c7e$413f1090$6401a8c0@Windows> That's actually a crucial question that needs to be resolved in order for digital signatures to be more widely adopted. If anyone can just revoke their key and claim a hacker break-in, then what good does it do to generate signatures to begin with? From svwright+lists at amtp.liv.ac.uk Tue Apr 27 19:54:49 2004 From: svwright+lists at amtp.liv.ac.uk (Stewart V. Wright) Date: Tue Apr 27 20:53:11 2004 Subject: can you deny you sent a signed e-mail? In-Reply-To: <00c101c42c7e$413f1090$6401a8c0@Windows> References: <20040427082201.GB27223@lori.mine.nu> <20040427114159.GA10394@northernsecurity.net> <00c101c42c7e$413f1090$6401a8c0@Windows> Message-ID: <20040427175449.GA3303@amtp.liv.ac.uk> G'day Jerry, * Jerry Windrel [040427 18:44]: > That's actually a crucial question that needs to be resolved in order for > digital signatures to be more widely adopted. If anyone can just revoke > their key and claim a hacker break-in, then what good does it do to generate > signatures to begin with? Really? How does this differ from a regular (handwritten) signature? Handwriting experts can be fooled, so even in the situation where you get an "expert" opinion (i.e. court) a signature is only as trustworthy as the participants define it. At least with a digital signature you can revoke it and (providing your correspondents update your key regularly) the potential for fraudulently signatures is time limited. I'd love to be corrected by someone who knows more about this than me though... Cheers, S. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 274 bytes Desc: Digital signature Url : /pipermail/attachments/20040427/005ca805/attachment.bin From jrhendri at maine.rr.com Tue Apr 27 20:59:04 2004 From: jrhendri at maine.rr.com (Jim Hendrick) Date: Tue Apr 27 20:56:49 2004 Subject: can you deny you sent a signed e-mail? In-Reply-To: <00c101c42c7e$413f1090$6401a8c0@Windows> Message-ID: <003301c42c89$b68d45f0$05211e0a@keane7926> It is usually more useful to be able to indicate you *did* send an email, sign a document, etc. To be useful for non-repudiation, you need to have a policy that states you are responsible for all actions using your account/key/whatever and you are expected to use proper care to protect that account/key/whatever. Similar to if you are issued a physical door key and claim "it wasn't me, somebody stole my keys". You may introduce reasonable doubt, but you also reduce your own credibility (in that event and in the future). Jim > -----Original Message----- > From: gnupg-users-bounces@gnupg.org > [mailto:gnupg-users-bounces@gnupg.org]On Behalf Of Jerry Windrel > Sent: Tuesday, April 27, 2004 1:37 PM > To: gnupg-users@gnupg.org > Subject: Re: can you deny you sent a signed e-mail? > > > That's actually a crucial question that needs to be resolved > in order for > digital signatures to be more widely adopted. If anyone can > just revoke > their key and claim a hacker break-in, then what good does it > do to generate > signatures to begin with? > > > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users > From iam-est-hora-surgere at despammed.com Tue Apr 27 21:04:38 2004 From: iam-est-hora-surgere at despammed.com (Marcus Frings) Date: Tue Apr 27 21:01:51 2004 Subject: Moving from PGP to GnuPG and other questions ... References: <200404271643.i3RGhNBp027790@www.aarg.net> Message-ID: * Denis Green wrote: > 1/ > How do I choose "No mail", web only option - for this list ? Forget your old behaviour, read this list via gmane.org as newsgroup. > 2/ > What are the most recommended Windows front ends for GnuPG > (Windows XP user) On Windows I prefer GPGShell which offers all the features I want: > 3/ > Is there a way to use the IDEA algorithm with GnuPG - if > so how ? [wish to be fully compatible with PGP users ] Yes, have a look at the FAQ: > 4/ > I presently use Pegasus for Windows as my e mail client. > Are there better free e mail clients - that integrate with > GPG especially ? Gnus is not the question, Gnus is the answer. > 6/ > any GnuPG equivalent for the PGPDisk ? AFAIK no. Regards, Marcus -- "Alice pressed against the wall So she can see the door In case the laughing strangers crawl and Crush the petals on the floor" From linux at codehelp.co.uk Tue Apr 27 21:21:22 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Tue Apr 27 21:17:41 2004 Subject: can you deny you sent a signed e-mail? In-Reply-To: <00c101c42c7e$413f1090$6401a8c0@Windows> References: <20040427082201.GB27223@lori.mine.nu> <20040427114159.GA10394@northernsecurity.net> <00c101c42c7e$413f1090$6401a8c0@Windows> Message-ID: <200404272021.26077.linux@codehelp.co.uk> On Tuesday 27 April 2004 6:37, Jerry Windrel wrote: > That's actually a crucial question that needs to be resolved in order for > digital signatures to be more widely adopted. If anyone can just revoke Adoption will only move as fast as the trust. Signing with an untrusted key declares your willingness to use GnuPG/PGP but the crucial element here is about keysigning and trust. > their key and claim a hacker break-in, then what good does it do to > generate signatures to begin with? Claiming a compromise when none occurred is deceitful, the costs of such lies will impact on the keyowner but how much depends on the web of trust. If the key is part of the strong set, i.e. if it's signed by other strong keys rather than a long list of nonsense keys, then there is a lot to lose by revoking the key. Denying a digital signature on such an email (like this one) is going to be costly in terms of the trust held in that key and it's owner. Someone who habitually revokes keys is going to find it hard to get his key repeatedly signed by other strong keys, so far less people will end up trusting the new key. A signature is much more than just anti-tamper - a valid signature from an untrusted key is not as useful as a valid signature from a trusted key. Anyone can create a new key that carries the same name and email address as an untrusted key. Until the key (and therefore the keyid) can be trusted, you can't know if this is a new key for an existing untrusted user or a new key from an unknown user. Keyservers can also provide clues here, signatures can only be verified if the public key is public so each revoked key will show up on the keyserver. As these don't get deleted, a search for the user might shed some light if anyone has cause to doubt the sincerity of the claim of a compromise. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040427/8df9b29b/attachment.bin From linux at codehelp.co.uk Tue Apr 27 21:28:42 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Tue Apr 27 21:24:56 2004 Subject: trust management script In-Reply-To: <200404270902.10391@fortytwo.ch> References: <200404212334.24999.linux@codehelp.co.uk> <200404270902.10391@fortytwo.ch> Message-ID: <200404272028.42651.linux@codehelp.co.uk> On Tuesday 27 April 2004 8:02, Adrian 'Dagurashibanipal' von Bidder wrote: > On Thursday 22 April 2004 00.34, Neil Williams wrote: > > Thanks for that script, this is certainly useful. Excellent. > > I worked on the idea that if I > > hadn't verified the key, I probably had never met the person behind > > the key and therefore I was unlikely to be able to make much of a > > decision on how carefully that person would verify someone else's > > key. (Not having had the chance to see them verify my own key.) > > I think this assumption does not always hold. I have many keys set to True. I didn't say this in the first email, but I have also left some keys at full trust even though I haven't met the person - people like Werner Koch - for the same reason as you. > full trust in my keyring - these are mostly keys from prominent Debian > developers. I trust them because I feel I can judge from seeing their > email on the mailing list how competent they are - certainly better > than I can judge from seeing them once at a keysigning party. Very good point. Although the list doesn't help verify the key itself, it does help in getting to know the person and their attitudes towards their key(s). In that sense, it does allow for a more considered judgement than a single brief meeting. However, a lot of the keys I have signed are people from my local LUG and I meet and correspond with them on a regular basis. Even those I've met at Expo are mostly still in occasional contact. > Trust is always a personal thing, and everybody needs to find their own > working compromise between usability and paranoia. In my case, the > fact that I rarely need to transmit really secret data helps - I use encryption mostly for local files. > encrypting ordinary email by default is different from encrypting > account passwords etc.; in the latter case I take an extra look at the > trust path. I agree. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040427/3b765974/attachment.bin From linux at codehelp.co.uk Tue Apr 27 22:06:15 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Tue Apr 27 22:02:30 2004 Subject: Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404271643.i3RGhNBp027790@www.aarg.net> References: <200404271643.i3RGhNBp027790@www.aarg.net> Message-ID: <200404272106.18365.linux@codehelp.co.uk> On Tuesday 27 April 2004 5:43, Denis Green wrote: > Hello Group > 3/ > Is there a way to use the IDEA algorithm with GnuPG - if Why compromise GnuPG with the patented IDEA? If you want to use free software, don't fund those whose patents would prevent the use of free software. Free = free speech, not just free beer. You want free? Then help GnuPG STAY free. http://www.dclug.org.uk/linux_doc/patent.html http://lpf.ai.mit.edu/Patents/rms-pto.html > so how ? [wish to be fully compatible with PGP users ] If you want to use the patent you'll have to pay the piper. The best option is to not use IDEA and educate the PGP users. > 4/ > I presently use Pegasus for Windows as my e mail client. > Are there better free e mail clients - that integrate with > GPG especially ? Heard good things about Turnpike and Eudora but I'm GNU/Linux only. > 7/ > Any free English language Dictionaries for English > meanings / spelling / pronunciation etc. ? (not spell > checker - I have Aspell already) Google? > 8/ > Free stuff preferred. If not available please recommend > paid versions as well Free does not mean not-paid! Free means liberty. http://www.codehelp.co.uk/html/about.html#principles Free software can still require payment but never reduces your freedom. Proprietary or patented software usually costs a fee and ALWAYS reduces your freedom. Please don't use 'paid versions' - what you mean is proprietary (or in the case of IDEA, patented). Free software (like Linux) can still be sold with a price. http://www.fsf.org/philosophy/free-sw.html -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040427/f6a7f2e6/attachment-0001.bin From eulenburg at gmx.de Tue Apr 27 22:28:17 2004 From: eulenburg at gmx.de (Mortimer Graf zu Eulenburg) Date: Tue Apr 27 22:27:38 2004 Subject: can you deny you sent a signed e-mail? In-Reply-To: <200404272021.26077.linux@codehelp.co.uk> References: <20040427082201.GB27223@lori.mine.nu> <00c101c42c7e$413f1090$6401a8c0@Windows> <200404272021.26077.linux@codehelp.co.uk> Message-ID: <200404272228.33244.eulenburg@gmx.de> Am Dienstag, 27. April 2004 21:21 schrieb Neil Williams: >On Tuesday 27 April 2004 6:37, Jerry Windrel wrote: >> That's actually a crucial question that needs to be resolved in order for >> digital signatures to be more widely adopted. If anyone can just revoke >> their key and claim a hacker break-in, then what good does it do to >> generate signatures to begin with? >Keyservers can also provide clues here, signatures can only be verified if > the public key is public so each revoked key will show up on the keyserver. > As these don't get deleted, a search for the user might shed some light if > anyone has cause to doubt the sincerity of the claim of a compromise. Hi Neal, Yes, but if people go in and fake complete signing networks that sign and revoke signs to make the keys appear sinister then it will be hard to argue around that your key underwent some sort of attack. It would be also very hard to defend such attacks, at last everybody is free to sign whomever key he wants to and revoke it with "key compromised" or such reason.. Greetz from Berlin, Mortimer -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040427/1a329e87/attachment.bin From linux at codehelp.co.uk Tue Apr 27 23:02:40 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Tue Apr 27 22:58:56 2004 Subject: can you deny you sent a signed e-mail? In-Reply-To: <200404272228.33244.eulenburg@gmx.de> References: <20040427082201.GB27223@lori.mine.nu> <200404272021.26077.linux@codehelp.co.uk> <200404272228.33244.eulenburg@gmx.de> Message-ID: <200404272202.43999.linux@codehelp.co.uk> On Tuesday 27 April 2004 9:28, Mortimer Graf zu Eulenburg wrote: > Yes, but if people go in and fake complete signing networks that sign and That is a seriously difficult task - a completely separate signing network is easy - getting people in the strong set to trust it requires a lot of security-aware people to be duped. > revoke signs to make the keys appear sinister then it will be hard to argue You cannot make someone else's strong key appear sinister by adding and revoking signatures, you just make their key very cluttered. It's revoked KEYS that are important in denying a signed email. If you revoke your signature on my key, that does not affect the validity of my digital signature and it may even have no particular effect on the level of trust in my key because there are lots of other signatures on my key. > around that your key underwent some sort of attack. It would be also very > hard to defend such attacks, at last everybody is free to sign whomever key Have you misunderstood signing, perhaps? If you sign my key, you can only revoke your signature on my key. That doesn't affect the validity of my key as atested by other signatures. So if you 'attack' my key, all that happens is that I get another nonsense signature added to the key. It doesn't invalidate the signatures made by others and in particular is does NOT make your key trusted - I have to sign YOUR key to make it trusted. That requires YOU to prove to me that you are the physical person declared in the key. I will not sign your key without full verification, so your key never becomes trusted. A long collection of revoked signatures on a key is NOT the same as a long list of revoked keys by a single user. It again comes back to trust - how trustworthy are the signatures that remain? GnuPG disregards any revoked signatures as well as signatures by untrusted keys, when calculating trust. > he wants to and revoke it with "key compromised" or such reason.. You are free to revoke your key, but that alone cannot affect my key. That's the strength of the web of trust, it is a web, not a chain. There is more than one path from my key to X key in the strong set, say Werner's. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040427/61c628e4/attachment.bin From vedaal at hush.com Tue Apr 27 23:10:37 2004 From: vedaal at hush.com (vedaal@hush.com) Date: Tue Apr 27 23:07:48 2004 Subject: can you deny you sent a signed e-mail? Message-ID: <200404272110.i3RLAbgZ003482@mailserver2.hushmail.com> >Message: 2 >Date: Tue, 27 Apr 2004 13:37:03 -0400 >From: "Jerry Windrel" >Subject: Re: can you deny you sent a signed e-mail? >To: >Message-ID: <00c101c42c7e$413f1090$6401a8c0@Windows> >Content-Type: text/plain; charset="iso-8859-1" >If anyone can just >revoke >their key and claim a hacker break-in, then what good does it do >to generate >signatures to begin with? it is more important for 'trust' than for 'legality' if someone trusts you, and your ability to sign and encrypt, and safely guard your key, then he/she can can be reasonably certain that a message that verifies with your signature, came from 'you', and not an mailicious impersonator if someone 'doesn't' trust you, then there is a 'notary' signature type that is being worked on, that can be requested, and can date and record a signature as being done at a certain time, which would be hard to claim later that it was stolen/forged if the signer admitted to using the key any time after the notarized signature, before revoking it/reporting it stolen (but who says that the 'notary' is trustworthy ... ?) my grandmother used to say, (back in the olden days when people were not in the habit of going to lawyers): "if you really know and trust someone, you don't need the 'papers' if you don't trust them, the 'papers' will not help you ..." vedaal Concerned about your privacy? Follow this link to get FREE encrypted email: https://www.hushmail.com/?l=2 Free, ultra-private instant messaging with Hush Messenger https://www.hushmail.com/services.php?subloc=messenger&l=434 Promote security and make money with the Hushmail Affiliate Program: https://www.hushmail.com/about.php?subloc=affiliate&l=427 From atom-gpg at suspicious.org Wed Apr 28 03:04:43 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 28 03:02:05 2004 Subject: Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404271643.i3RGhNBp027790@www.aarg.net> References: <200404271643.i3RGhNBp027790@www.aarg.net> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > 3/ > Is there a way to use the IDEA algorithm with GnuPG - if > so how ? [wish to be fully compatible with PGP users ] ====================== unless it's an ancient version of PGP(tm) it should auto-magically handle other algorithms. from the OpenPGP spec: Implementations MUST implement Triple-DES. Implementations SHOULD implement IDEA and CAST5. Implementations MAY implement any other algorithm. IDEA is the *only* symmetric algorithm mentioned in the spec that gpg doesn't handle "out of the box." you can use: 3DES (168), CAST5 (128), blowfish (128), AES-128, AES-192, AES-256, twofish (256). > 7/ > Any free English language Dictionaries for English > meanings / spelling / pronunciation etc. ? (not spell > checker - I have Aspell already) ====================== http://dictionary.com/ ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "They tell us that we live in a great free republic; that our institutions are democratic; that we are a free and self-governing people. That is too much, even for a joke. Wars throughout history have been waged for conquest and plunder. And that is war in a nutshell. The master class has always declared the wars; the subject class has always fought the battles." -- Eugene V. Debs, 1918 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCPAzAACgkQnCgLvz19QePzQACbBPwHNOyjR1g3yiiOL2XpC5++ VNQAoJOWA/mqqihGIjflgXTJnCJrUSU1 =EiSs -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Wed Apr 28 09:12:28 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 28 09:09:44 2004 Subject: trust management script In-Reply-To: <200404212334.24999.linux@codehelp.co.uk> References: <200404212334.24999.linux@codehelp.co.uk> Message-ID: <200404280912.28460@fortytwo.ch> Skipped content of type multipart/mixed-------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040428/fc52d9f8/attachment.bin From avbidder at fortytwo.ch Wed Apr 28 09:14:15 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 28 09:11:27 2004 Subject: can you deny you sent a signed e-mail? In-Reply-To: <20040427082201.GB27223@lori.mine.nu> References: <20040427082201.GB27223@lori.mine.nu> Message-ID: <200404280914.19208@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday 27 April 2004 10.22, Geert Hendrickx wrote: > Hi, > > GPG gives us a method for signing our e-mails so the addressee can be > sure it was sent by yourself. > > But what about the inverse question: can you deny that you sent an > e-mail which has been signed using you secret key? You can always publish the secret key (but of course, explicitly publishing your secret key will make anybody think that something is fishy.) Other than that, plead a break-in and hope somebody believes you. cheers - -- vbi - -- The content of this message may or may not reflect the opinion of me, my employer, my girlfriend, my cat or anybody else, regardless of the fact whether such an employer, girlfriend, cat, or anybody else exists. I (or my employer, girlfriend, cat or whoever) disclaim any legal obligations resulting from the above message. You, as the reader of this message, may or may not have the permission to redistribute this message as a whole or in parts, verbatim or in modified form, or to distribute any message at all. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkCPWctgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6KRsAn083AC2uQN23kD1HEwco4Ebs 1xLoAKCfPe9/iMxhbrUnTn5X4YzH3Roqeg== =GaQ1 -----END PGP SIGNATURE----- From avbidder at fortytwo.ch Wed Apr 28 09:26:13 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 28 09:23:20 2004 Subject: Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404271643.i3RGhNBp027790@www.aarg.net> References: <200404271643.i3RGhNBp027790@www.aarg.net> Message-ID: <200404280926.13541@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday 27 April 2004 18.43, Denis Green wrote: > 3/ > Is there a way to use the IDEA algorithm with GnuPG - if > so how ? [wish to be fully compatible with PGP users ] If your main goal is full compatibility with PGP users, then you don't need the IDEA algorithm, afaict. Just be sure that your key preferences don't specify IDEA as a cipher you support (if so, change the preferences, which will create a new selfsignature on the key, and then make sure everybody you communicate with has the new version of your key.) (Can some PGP-expert confirm that PGP does handle multiple selfsigs with differing preferences in a sane way?) The result will be that nobody sends you IDEA-encrypted email. Of course, if you have old data files with IDEA encryption, you'll need to either compile a GnuPG with IDEA, or keep PGP around until these are all converted. greetings - -- vbi - -- Die Sch?nheit des Alltags entdecken, ist eine Kunst des Lebens. -- A. Bartsch -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkCPXJVgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l61QkAoJn6jCpY7uXN9anO4Mx6GggY ONi4AJ9szu7sQ4OmjX7ZNfWp2BBf+y0oXw== =ITH2 -----END PGP SIGNATURE----- From brian at mahern.net Tue Apr 27 15:53:20 2004 From: brian at mahern.net (Brian Mahern) Date: Wed Apr 28 09:23:40 2004 Subject: new install - old version number Message-ID: <200404271353.i3RDrKHK025122@mmm2623.sbc-webhosting.com> I downloaded, compiled and installed the latest version of GnuPG (1.2.4). When I run it from the command line with the --version command it still shows the older version. (1.0.6), but when it used in through Enigmail the signed message indicates the the newer version. What gives? Which is it? Thanks. From jose.betances at rbccm.com Tue Apr 27 21:52:51 2004 From: jose.betances at rbccm.com (Betances, Jose) Date: Wed Apr 28 09:23:54 2004 Subject: Corrupted Files -- Need Help Message-ID: <3B979D87D713194585AF9BA0CE2052A70B2B0AB3@sew39031.oak.fg.rbc.com> I was wondering if you had been able to resolve the issue you had decrypting a file using gpg. If so, can you tell me how you resolved it; I am having the same problem. Thanks

------------------------------------------------------------------------------

This E-Mail (including any attachments) may contain privileged or confidential information. It is intended only for the addressee(s) indicated above. The sender does not waive any of its rights, privileges or other protections respecting this information. Any distribution, copying or other use of this E-Mail or the information it contains, by other than an intended recipient, is not sanctioned and is prohibited. If you received this E-Mail in error, please delete it and advise the sender (by return E-Mail or otherwise) immediately.

This E-Mail (including any attachments) has been scanned for viruses. It is believed to be free of any virus or other defect that might affect any computer system into which it is received and opened. However, it is the responsibility of the recipient to ensure that it is virus free. The sender accepts no responsibility for any loss or damage arising in any way from its use.

E-Mail received by or sent from RBC Capital Markets is subject to review by Supervisory personnel. Such communications are retained and may be produced to regulatory authorities or others with legal rights to the information.

====================================================

-------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20040427/03a48fb2/attachment.html From malsyned at uofr.net Wed Apr 28 04:29:01 2004 From: malsyned at uofr.net (Dennis Lambe Jr.) Date: Wed Apr 28 09:24:05 2004 Subject: encrypting files in a script In-Reply-To: References: <4DCE15B9C4E66F4CA967EBF64C53D64D1902C6@bstn-exch1.forumsys.com> Message-ID: <1083119341.27533.121.camel@localhost> On Tue, 2004-04-20 at 19:12, Atom 'Smasher' wrote: > > How do I mark the key as trusted? I still get "the > > question" even after I do > > > > gpg --edit-key foo > > trust > > 4 ( = "I trust fully") > > quit > ================================= That's because you haven't signed the key. GPG's "trust" is actually a measure of how much you trust that key to sign other people's keys responsibly. What you're trying to do is assure GPG that the key really does belong to the person it says it does, which is accomplished through signing it (or trusting someone who signed it, thanks to the Web of Trust). > "full" trust isn't enough... you'll have to select: > 5 = I trust ultimately This is not a good solution in this case. I don't even think it will solve the problem, and it's giving the key's owner a lot of credit that you may not have good reason to give him or her. > or use: > --trust-model always If your script is supposed to handle arbitrary keys gracefully, this is probably the best way. It ignores all issues of whether the key belongs to the person it claims to, which is a security hazard in many circumstances. Can you give a little more information about what this script does? --D -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 279 bytes Desc: This is a digitally signed message part Url : /pipermail/attachments/20040427/c01f5461/attachment.bin From malsyned at uofr.net Wed Apr 28 04:30:52 2004 From: malsyned at uofr.net (Dennis Lambe Jr.) Date: Wed Apr 28 09:24:11 2004 Subject: Remote signing In-Reply-To: <20040420162145.560f6feb.ian_bruce@fastmail.fm> References: <20040419234757.GA23117@it.usyd.edu.au> <4084D75E.8070503@computing-services.oxford.ac.uk> <20040420084111.GA28365@Update.UU.SE> <4084F3B4.2030507@computing-services.oxford.ac.uk> <20040420121005.GB10970@it.usyd.edu.au> <40851773.20005@computing-services.oxford.ac.uk> <20040420123950.GA4404@it.usyd.edu.au> <20040420162145.560f6feb.ian_bruce@fastmail.fm> Message-ID: <1083119451.27536.126.camel@localhost> On Tue, 2004-04-20 at 19:21, Ian Bruce wrote: > If the Sun machine or the SSH session are > compromised, the worst that can happen is that the human user will be > tricked into signing a bogus message. This risk can be mitigated by having the client-side script print out the to-be-signed text for verification. --D -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 279 bytes Desc: This is a digitally signed message part Url : /pipermail/attachments/20040427/a852f31f/attachment.bin From atom-gpg at suspicious.org Wed Apr 28 09:37:56 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Wed Apr 28 09:35:09 2004 Subject: new install - old version number In-Reply-To: <200404271353.i3RDrKHK025122@mmm2623.sbc-webhosting.com> References: <200404271353.i3RDrKHK025122@mmm2623.sbc-webhosting.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > I downloaded, compiled and installed the latest version of GnuPG > (1.2.4). When I run it from the command line with the --version command > it still shows the older version. (1.0.6), but when it used in through > Enigmail the signed message indicates the the newer version. What > gives? Which is it? ============================= you might have them both installed, in different places. rebuild the "locate" database *, and run: $ locate gpg | egrep 'bin' i suspect you'll find two copies. run them each like this: $ /full/path/to/gpg --version * you can also `find` gpg by doing this: $ find / -type f -name gpg 2> /dev/null | egrep 'bin' ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "The universe is not hostile, nor yet is it friendly. It is simply indifferent." -- John Hughes Holmes -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) Comment: What is this gibberish? - http://atom.smasher.org/links/#digital_signatures iEYEARECAAYFAkCPX1kACgkQnCgLvz19QePTSgCeICf5Bl/ab9J7Y8Wy5hV4xNON OC0An3Y1n665zTtMbxPf9d+kkWMqen2Z =b/pn -----END PGP SIGNATURE----- From samuel at Update.UU.SE Wed Apr 28 10:14:39 2004 From: samuel at Update.UU.SE (Samuel ]slund) Date: Wed Apr 28 10:12:08 2004 Subject: new install - old version number In-Reply-To: <200404271353.i3RDrKHK025122@mmm2623.sbc-webhosting.com> References: <200404271353.i3RDrKHK025122@mmm2623.sbc-webhosting.com> Message-ID: <20040428081438.GM27722@Update.UU.SE> On Tue, Apr 27, 2004 at 09:53:20AM -0400, Brian Mahern wrote: > I downloaded, compiled and installed the latest version of GnuPG (1.2.4). When I run it from the command line with the --version command it still shows the older version. (1.0.6), but when it used in through Enigmail the signed message indicates the the newer version. What gives? Which is it? You probably have the older version in your path. Try using the full path to the gpg executable you compiled, that should give you the correct version. If you are using an Unix-like system try "which gpg" to see what executable you are using from the command line. On windows I would try to search for gpg.exe to see if any unexpected or old turns up. HTH //Samuel From stuart.yeates at computing-services.oxford.ac.uk Wed Apr 28 10:15:17 2004 From: stuart.yeates at computing-services.oxford.ac.uk (Stuart A Yeates) Date: Wed Apr 28 10:14:37 2004 Subject: trust management script In-Reply-To: <200404280912.28460@fortytwo.ch> References: <200404212334.24999.linux@codehelp.co.uk> <200404280912.28460@fortytwo.ch> Message-ID: <408F6815.8020002@computing-services.oxford.ac.uk> Adrian 'Dagurashibanipal' von Bidder wrote: > Incidentally: is gpg's database handling good enough, or will I get a > speedup by starting again from an empty keyring? You will get a speed-up. Be aware that keys with multiple signatures and uids consume far more space that keys with fewer signatures and uids. The safe way to start with an empty keyring is: 0) send any new signatures to the keyservers (gpg --send) 1) rename pubring.gpg 2) rename trustdb.gpg 3) import your public key(s) from a keyserver 4) edit your key(s) to make them trusted cheers stuart -- Stuart Yeates stuart.yeates@computing-services.oxford.ac.uk OSS Watch http://www.oss-watch.ac.uk/ Oxford Text Archive http://ota.ahds.ac.uk/ Humbul Humanities Hub http://www.humbul.ac.uk/ From linux at codehelp.co.uk Wed Apr 28 10:55:15 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Wed Apr 28 10:51:26 2004 Subject: trust management script In-Reply-To: <408F6815.8020002@computing-services.oxford.ac.uk> References: <200404212334.24999.linux@codehelp.co.uk> <200404280912.28460@fortytwo.ch> <408F6815.8020002@computing-services.oxford.ac.uk> Message-ID: <200404280955.16327.linux@codehelp.co.uk> On Wednesday 28 April 2004 9:15, Stuart A Yeates wrote: > Adrian 'Dagurashibanipal' von Bidder wrote: > > Incidentally: is gpg's database handling good enough, or will I get a > > speedup by starting again from an empty keyring? > > You will get a speed-up. Be aware that keys with multiple signatures and > uids consume far more space that keys with fewer signatures and uids. > The safe way to start with an empty keyring is: > > 0) send any new signatures to the keyservers (gpg --send) > 1) rename pubring.gpg > 2) rename trustdb.gpg > 3) import your public key(s) from a keyserver > 4) edit your key(s) to make them trusted That will take a while! In a keyring of 3,000 keys where 2,000 have multiple signatures, just how much speed is going to be gained over simply deleting the 1,000 keys? Each time --update-trustdb gets new trust information, the entire trust database needs to be checked so the closer you get to that 2,000th multiple sig key, the longer the trust check takes. Instead of a single trust update after deleting 1,000 keys with a script, you are using a recursively longer trust check 2,000 times. Me thinks that would take longer than a simple find, identify, delete script. :-) You would have to have a very considerable speed improvement to justify the effort! -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040428/42fe5f03/attachment.bin From Holger.Sesterhenn at smgwtest.aachen.utimaco.de Wed Apr 28 12:16:40 2004 From: Holger.Sesterhenn at smgwtest.aachen.utimaco.de (Holger Sesterhenn) Date: Wed Apr 28 12:14:30 2004 Subject: Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404280926.13541@fortytwo.ch> References: <200404271643.i3RGhNBp027790@www.aarg.net> <200404280926.13541@fortytwo.ch> Message-ID: <408F8488.9070707@smgwtest.aachen.utimaco.de> Hi, Adrian 'Dagurashibanipal' von Bidder wrote: > (Can some PGP-expert confirm that PGP does handle multiple selfsigs with > differing preferences in a sane way?) I have tried to remove IDEA from older V4 keys (RSA or DSA) using GnuPGs --edit command and failed using PGP 8.03. Of course there are no preferences on V3 keys. I have deleted the old key from PGP, exported the public key with new preferences from GnuPG an then imported again this key to PGP 8.03. PGP still showed IDEA as prefered algorithm (just set 3DES, CAST5 and AES with GnuPG). Don't know whether there is a special feature to rebuild the keyring with PGP. I have in mind that there is somewhere a switch to force PGP to use a certain algorithm as default. Some time ago I have created 3DES encrypted data with PGP 8.0x. -- Best Regards, Holger Sesterhenn --- Internet http://www.utimaco.com From g-r-v at ukr.net Wed Apr 28 12:29:15 2004 From: g-r-v at ukr.net (Robert Golovniov) Date: Wed Apr 28 12:25:19 2004 Subject: pksd error Message-ID: <200404281036.i3SAaWbS014982@gw.core> Hello, After installing PKS server and trying to add a key from a remote machine, I get this error message: kd_open: Error opening num_keydb (errno = 2) Could anybody give me a hint on how to solve it? -- -=Robert & Beata Golovniov | Lviv, Ukraine=- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ mailto:golovniov@interia.pl?subject=PGP%20Key&Body=Embedded%20key ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From anonymous at remailer.metacolo.com Wed Apr 28 13:24:13 2004 From: anonymous at remailer.metacolo.com (Anonymous Sender) Date: Wed Apr 28 13:21:28 2004 Subject: Moving from PGP to GnuPG and other questions ... Message-ID: Neil Williams: > On Tuesday 27 April 2004 5:43, Denis Green wrote: > > 4/ > > I presently use Pegasus for Windows as my e mail client. > > Are there better free e mail clients - that integrate with > > GPG especially ? Dennis, give Becky! 2 (with the GnuPG plugin) a try. From avbidder at fortytwo.ch Wed Apr 28 13:30:33 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Wed Apr 28 13:27:45 2004 Subject: trust management script In-Reply-To: <200404280955.16327.linux@codehelp.co.uk> References: <200404212334.24999.linux@codehelp.co.uk> <408F6815.8020002@computing-services.oxford.ac.uk> <200404280955.16327.linux@codehelp.co.uk> Message-ID: <200404281330.38355@fortytwo.ch> On Wednesday 28 April 2004 10.55, Neil Williams wrote: > On Wednesday 28 April 2004 9:15, Stuart A Yeates wrote: > > Adrian 'Dagurashibanipal' von Bidder wrote: > > > Incidentally: is gpg's database handling good enough, or will I > > > get a speedup by starting again from an empty keyring? > > > > You will get a speed-up. Be aware that keys with multiple > > signatures and uids consume far more space that keys with fewer > > signatures and uids. The safe way to start with an empty keyring > > is: > > > > 0) send any new signatures to the keyservers (gpg --send) > > 1) rename pubring.gpg > > 2) rename trustdb.gpg > > 3) import your public key(s) from a keyserver > > 4) edit your key(s) to make them trusted > > That will take a while! > > In a keyring of 3,000 keys where 2,000 have multiple signatures, just > how much speed is going to be gained over simply deleting the 1,000 > keys? Well, it doesn't take that long, and in my case gpg speed is really quite bad (sometimes it takes 1 second or so for a simple gpg --list-key , and a --check-trustdb takes ages (90s or so.) So I guess I'll try it - the process is simple enough $ gpg --export > keys.gpg $ gpg --export-ownertrust > trust $ mv trustdb.gpg trustdb.gpg.old $ mv pubring.gpg pubring.gpg.old $ gpg --import < keys.gpg $ gpg --import-ownertrust < trust $ gpg --check-trustdb If key lookup is starting to be the slow part of gpg operations, and not the crypto, I find this a bit annoying. cheers -- vbi -- My uncle was the town drunk -- and we lived in Chicago. -- George Gobel -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 331 bytes Desc: signature Url : /pipermail/attachments/20040428/3e00be0e/attachment.bin From linux at codehelp.co.uk Wed Apr 28 14:10:55 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Wed Apr 28 14:07:45 2004 Subject: Moving from PGP to GnuPG and other questions ... In-Reply-To: <408F8488.9070707@smgwtest.aachen.utimaco.de> References: <200404271643.i3RGhNBp027790@www.aarg.net> <200404280926.13541@fortytwo.ch> <408F8488.9070707@smgwtest.aachen.utimaco.de> Message-ID: <200404281310.56460.linux@codehelp.co.uk> On Wednesday 28 April 2004 11:16, Holger Sesterhenn wrote: > PGP still showed IDEA as prefered algorithm (just set 3DES, CAST5 and > AES with GnuPG). PGP might want to prefer IDEA but what happens when you use PGP to try and encrypt to the GnuPG key? Preferred might mean preferred as a general rule rather than as a specific preference for this key. > I have in mind that there is somewhere a switch to force PGP to use a > certain algorithm as default. Some time ago I have created 3DES > encrypted data with PGP 8.0x. Doesn't it depend on what the recipient key can use? PGP must be smart enough not to encrypt to a key using an algorithm that it knows the key does not handle. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040428/12e1fb5c/attachment.bin From Holger.Sesterhenn at smgwtest.aachen.utimaco.de Wed Apr 28 15:26:08 2004 From: Holger.Sesterhenn at smgwtest.aachen.utimaco.de (Holger Sesterhenn) Date: Wed Apr 28 15:23:35 2004 Subject: Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404281310.56460.linux@codehelp.co.uk> References: <200404271643.i3RGhNBp027790@www.aarg.net> <200404280926.13541@fortytwo.ch> <408F8488.9070707@smgwtest.aachen.utimaco.de> <200404281310.56460.linux@codehelp.co.uk> Message-ID: <408FB0F0.1020907@smgwtest.aachen.utimaco.de> Hi, Neil Williams wrote: > PGP might want to prefer IDEA but what happens when you use PGP to try and > encrypt to the GnuPG key? Preferred might mean preferred as a general rule > rather than as a specific preference for this key. If I import a new key to PGP which has been generated by GnuPG with its default preferences ( Cipher: AES256, AES192, AES, CAST5, 3DES Digest: SHA1, RIPEMD160 Compression: ZLIB, ZIP, Uncompressed Features: MDC ,done with GnuPG 1.2.3) PGP will encrypt with CAST5 or 3DES (maybe AES). --------------------- BUT I have changed the prefs of a V4 key which has been generated with PGP (with IDEA as default!), exported to GnuPG, *deleted* entirely from PGP, exported from GnuPG with new prefs set and again imported from PGP. Will say: there was a time where PGP did not know about the existence of this key. But somewhere in PGP's key database information can be found that this key has had IDEA as the default algo. > Doesn't it depend on what the recipient key can use? PGP must be smart enough > not to encrypt to a key using an algorithm that it knows the key does not > handle. That exactly is the question! -- Best Regards, Holger Sesterhenn --- Internet http://www.utimaco.com From rainman at hod.aarg.net Wed Apr 28 16:14:45 2004 From: rainman at hod.aarg.net (Denis Green) Date: Wed Apr 28 16:11:53 2004 Subject: re. Moving from PGP to GnuPG and other questions ... Message-ID: <200404281414.i3SEEjWo021233@www.aarg.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello group 1/ Thanks for the responses. I've downloaded GPG and installed the same. - - Since I don't have an earlier installation of *trusted GPG* and - - couldn't succeed with the MD5sum command, I haven't verified the signatures. So If someone can help me with the verification of signatures, I'd be grateful 2/ Though I've been using PGP for a short while, I've never checked detached signatures. 3/ I've installed GPG and GPG shell in a temporary directory (to test out till I verify signatures). I hope I can un- install and re install GPG / shell, if need be (if the signature check end up bad) I've imported public / private keys from PGP using GPGKeys I can see the keys when I start GPGKeys (though I see each key repeated in two lines) 4/ When I try to encrypt files using GPGtools, I get to select the file(to be encrypted), but when the " GPG Tools - Encrypt " window opens up, I don't see any public keys in the dialog box 5/ What I've checked so far - ------------------------------ - - GPG.exe is in the path - - GnuPG-Homedir has the config files http://www.gnupg.org/(en)/documentation/faqs.html#q5.8 - - the correct directory (having config and keyrings ) is specified as the - GnuPG-Homedir in the GPgconfig window 6/ I've visited http://www.gnupg.org/(en)/documentation/faqs.html but can't find an answer to my problem 7/ any clues on where I've gone wrong ? any help / directions are appreciated 8/ sorry about the "free " software stuff in my last post 9/ How does one denote "free as in free beer" (i.e.) Software not requiring payment, while writing to groups like this ? [and how to distinguish **that** from Free as in Freedom of Speech ? -say a version of Linux that needs payment ] thanks DG send a blank mail to rainman_(1)_eml_(2)_cc to get my public key. replace _(1)_ with @ and _(2)_ with . -----BEGIN PGP SIGNATURE----- Version: 6.5.8ckt http://www.ipgpp.com/ iQA/AwUBQI9tSh1/6GqH2TSrEQIq6ACg2yvyhoMvjTs2IGwRpc7LtKd3ZtEAoPE6 0MSMqpR5FDnsj3g/ixq5q2zX =O/VP -----END PGP SIGNATURE----- From sbutler at fchn.com Wed Apr 28 17:00:27 2004 From: sbutler at fchn.com (Steve Butler) Date: Wed Apr 28 16:58:24 2004 Subject: Moving from PGP to GnuPG and other questions ... Message-ID: <9A86613AB85FF346BB1321840DB42B4B046D4404@jupiter.fchn.com> At one time a client was sending us files encrypted with IDEA (they were a PGP shop). Found that there were no prefs on the key that had been generated by GnuPG. So, I edited the key, set the preferences and had the client update their keyring. They no longer send us files encrypted using IDEA. So, yes, a PGP client will honor the preferences set by GnuPG. --Steve Butler Oracle Administrator First Choice Health Network Seattle, WA USA sbutler@fchn.com 206-268-2309 -----Original Message----- From: Holger Sesterhenn [mailto:Holger.Sesterhenn@smgwtest.aachen.utimaco.de] Sent: Wednesday, April 28, 2004 6:26 AM To: Gnupg-Users Subject: Re: Moving from PGP to GnuPG and other questions ... Hi, Neil Williams wrote: > PGP might want to prefer IDEA but what happens when you use PGP to try and > encrypt to the GnuPG key? Preferred might mean preferred as a general rule > rather than as a specific preference for this key. If I import a new key to PGP which has been generated by GnuPG with its default preferences ( Cipher: AES256, AES192, AES, CAST5, 3DES Digest: SHA1, RIPEMD160 Compression: ZLIB, ZIP, Uncompressed Features: MDC ,done with GnuPG 1.2.3) PGP will encrypt with CAST5 or 3DES (maybe AES). --------------------- BUT I have changed the prefs of a V4 key which has been generated with PGP (with IDEA as default!), exported to GnuPG, *deleted* entirely from PGP, exported from GnuPG with new prefs set and again imported from PGP. Will say: there was a time where PGP did not know about the existence of this key. But somewhere in PGP's key database information can be found that this key has had IDEA as the default algo. > Doesn't it depend on what the recipient key can use? PGP must be smart enough > not to encrypt to a key using an algorithm that it knows the key does not > handle. That exactly is the question! -- Best Regards, Holger Sesterhenn --- Internet http://www.utimaco.com _______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. From linux at codehelp.co.uk Wed Apr 28 18:07:44 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Wed Apr 28 18:04:06 2004 Subject: re. Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404281414.i3SEEjWo021233@www.aarg.net> References: <200404281414.i3SEEjWo021233@www.aarg.net> Message-ID: <200404281707.51219.linux@codehelp.co.uk> On Wednesday 28 April 2004 3:14, Denis Green wrote: > 4/ > When I try to encrypt files using GPGtools, I get to > select the file(to be encrypted), but when the > " GPG Tools - Encrypt " window opens up, I don't see > any public keys in the dialog box That's because no public keys are set as trusted. If you have your own secret key, import both the secret and public key and then use --edit-key to set your own key to ultimate trust. GnuPG will check through the other keys in the keyring and will only let you encrypt to those that can be trusted (without using command-line options intended for secure environments). These will be keys that you have already signed or keys that are signed by people you have signed, etc. There's no sense allowing encryption of sensitive data to a key that cannot be trusted! If you override the GnuPG security using '--trust-model always' or similar, you must still verify the key in some other manner. Trust begins with the secret keys - those are presumably yours so if you have the passphrase, these should be set as ultimate trust. I tend to consider ultimate trust as only for keys with a usable secret key. All other keys then have their trust calculated as starting from your ultimate keys, fully trusted (allowing encryption), marginally trusted (needing an override but still not recommended) or trust unknown (don't encrypt to these unless you *really* know what you are doing). Other trust factors like revoked and expired are hopefully clear in their meaning. BTW. General question: will '--trust-model always' allow encryption to a revoked key? i.e. after you've imported/refreshed the key with it's revocation certificate. (I'd guess that it won't but haven't got any revoked keys to test against.) > 5/ > What I've checked so far If you can list keys then GPG is working. Issues about encryption are actually about trust. > 9/ > How does one denote "free as in free beer" (i.e.) Software > not requiring payment, while writing to groups like this ? shareware, proprietary, demo, trojan (!), virus, (!) . . . Even shareware is proprietary - you don't generally have access to the source code and you are not free to redistribute the program with or without modifications either with or without payments - there are restrictions. > [and how to distinguish **that** from Free as in Freedom of > Speech ? -say a version of Linux that needs payment ] A free software program that requires payment is still free software. Think of 'free software' as one term, not two words. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040428/9d383b45/attachment-0001.bin From Holger.Sesterhenn at smgwtest.aachen.utimaco.de Wed Apr 28 18:44:48 2004 From: Holger.Sesterhenn at smgwtest.aachen.utimaco.de (Holger Sesterhenn) Date: Wed Apr 28 18:42:29 2004 Subject: Moving from PGP to GnuPG and other questions ... In-Reply-To: <9A86613AB85FF346BB1321840DB42B4B046D4404@jupiter.fchn.com> References: <9A86613AB85FF346BB1321840DB42B4B046D4404@jupiter.fchn.com> Message-ID: <408FDF80.4080101@smgwtest.aachen.utimaco.de> Hi, > At one time a client was sending us files encrypted with IDEA (they were a > PGP shop). Found that there were no prefs on the key that had been > generated by GnuPG. So, I edited the key, set the preferences and had the > client update their keyring. They no longer send us files encrypted using > IDEA. > > So, yes, a PGP client will honor the preferences set by GnuPG. Well, I have to admit that I have to correct my previous mails. Because my test have been a while ago I have done it again. The latest PGP 8.x seem to handle the key preferrences correctly. I have created a testkey with GnuPG, setting CAST5 as preferred algo, imported it to PGP. PGP shows CAST5 as preferred algo. Then I have used 'setpref' with GnuPG and switched to 3DES as preferred algo (CAST5 still at second position). After importing this changed public key, there was a second selfsig and the key show 3DES as preferred algo with PGP8. Don't need to delete the key before. Everything is merged. Looks like the old versions (6.5, 7.x) did not merge correctly and that was the reason why I said PGP has some bugs. I have found the 'default algo' switch for PGP8 under 'Options/Adavanced' (well, I have the german version! Maybe the translation is not exact). You have to set the algo *before* you create a new key. It's possible to disable specific algos, too. So tell your recipients to switch off IDEA and use AES, CAST5 or 3DES. -- Best Regards, Holger Sesterhenn --- Internet http://www.utimaco.com From rlaager at wiktel.com Wed Apr 28 19:15:05 2004 From: rlaager at wiktel.com (Richard Laager) Date: Wed Apr 28 19:12:15 2004 Subject: pksd error In-Reply-To: <200404281036.i3SAaWbS014982@gw.core> Message-ID: <004401c42d44$5a1f6710$d2f2fea9@umcrookston.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > After installing PKS server and trying to add a key from a > remote > machine, I get this error message: > > kd_open: Error opening num_keydb (errno = 2) > > Could anybody give me a hint on how to solve it? For new installations, you really should be using SKS. It was written from the ground up to fix all of the fatal flaws in PKS. See this website: http://www.nongnu.org/sks/ Richard Laager -----BEGIN PGP SIGNATURE----- Version: PGP 8.0.2 Comment: If you don't know what this is, you can safely ignore it. iQA/AwUBQI/mmW31OrleHxvOEQL04gCguz1qVuKfPmSFJHPwIiewcmleIsQAn31T 1QYacJSmfkqr6RgGFTE8mbWF =t9gZ -----END PGP SIGNATURE----- From ewheaton at rapidstability.com Wed Apr 28 21:04:48 2004 From: ewheaton at rapidstability.com (Eli Wheaton) Date: Wed Apr 28 21:02:30 2004 Subject: command window popup Message-ID: Hi All, Does anyone know how to use gnupg from within an application without having the black Windows command window popup? THANKS IN ADVANCE From rainman at hod.aarg.net Tue Apr 27 18:15:18 2004 From: rainman at hod.aarg.net (Denis Green) Date: Wed Apr 28 21:44:40 2004 Subject: an we delete (our own) posts sent here earlier ? Message-ID: <200404271615.i3RGFICN025964@www.aarg.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi folks : One more question that I did not add to my earlier mail... 1/ How do I delete a post that I sent to this list (if need be ?) 2/ for e.g. I've mentioned the full e mail address for my key - - autoresponder in the last post. I'd like to modify it as usual.. talking away the @ and replacing it with the _at_ etc.. Can I do that ? If so how ? thanks in advance DG -----BEGIN PGP SIGNATURE----- Version: 6.5.8ckt http://www.ipgpp.com/ iQA/AwUBQI5Acx1/6GqH2TSrEQLrGwCfWEOQ271Y9f6XzkzOlrLzGCwJkV0An3Qi BB0sFFYPViGgTwJLmg9G+JM9 =9faC -----END PGP SIGNATURE----- From pt at radvis.nu Wed Apr 28 21:48:41 2004 From: pt at radvis.nu (Per Tunedal Casual) Date: Wed Apr 28 21:45:29 2004 Subject: block-cipher weakness? - was: Re: twofish keysize In-Reply-To: References: <200404162004.13385.jluehr@gmx.net> <200404162114.57964.jluehr@gmx.net> <20040417021929.GC30100@jabberwocky.com> <200404190020.15022.malte.gell@gmx.de> <87fzb0h1ii.fsf@vigenere.g10code.de> <6.0.3.0.2.20040420144800.027dd950@localhost> <87zn9566r8.fsf@vigenere.g10code.de> <6.0.3.0.2.20040422084123.027e0720@localhost> Message-ID: <6.0.3.0.2.20040428203918.027ccec0@localhost> At 18:54 2004-04-26, you wrote: >On Thu, 22 Apr 2004, Per Tunedal Casual wrote: >> On average 32 GB is the limit i.e. the probability is 50 % that an >> adversary finds two identical blocks. But you might have bad luck and >> he might find two identical blocks in somewhat smaller files. >=============================== > >can someone explain to me why/how it's a vulnerability if an attacker >finds more than one block of cipher text that are the same? I am not a cryptographer, but I suppose it gives a hint to find the key. That's why certain encryption modes like ECB (electronic code book encryption) should not be used. With ECB all similar blocks will get the same encryption. > >would there be a benefit of using a stream cipher for large files? I don't think it matters, se below. > >is there a reason why OpenPGP doesn't specify any stream ciphers? > OpenPGP uses the encryption mode CFB , Werner Koch wrote recently. CFB is an encryption mode that turns a block cipher into a stream cipher ... Per Tunedal From jerry.windrel at verizon.net Wed Apr 28 23:36:12 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Wed Apr 28 23:32:02 2004 Subject: command window popup References: Message-ID: <001901c42d68$d3eefc80$6401a8c0@Windows> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Basically you have to call the Windows API to create a process. Here's how you would do it in Perl... there are similar techniques in other languages. ==================================== # run the command 'gpg --recipient harry --encrypt foo.txt' without # a command prompt box use Win32; use Win32::Process; Win32::Process::Create($Win32::Process::Create::ProcessObj, 'c:\gnupg\gpg.exe', '--recipient', 'harry', '--encrypt', 'foo.txt', 0, DETACHED_PROCESS, '.') || die &print_error; sub print_error { print Win32::FormatMessage(Win32::GetLastError()); } - ----- Original Message ----- From: "Eli Wheaton" To: Sent: Wednesday, April 28, 2004 3:04 PM Subject: command window popup > Hi All, > > Does anyone know how to use gnupg from within an application > without having the black Windows command window popup? > > THANKS IN ADVANCE > > > > _______________________________________________ > Gnupg-users mailing list > Gnupg-users@gnupg.org > http://lists.gnupg.org/mailman/listinfo/gnupg-users -----BEGIN PGP SIGNATURE----- Version: PGP 8.0.3 - not licensed for commercial use: www.pgp.com iQA/AwUBQJAjoYlVuABWWiqVEQK6pACgiXX++1o8OPPeylixht91237LcPoAnRbK vVR5SBv2FdwOa23uwJjdbmd1 =M2R1 -----END PGP SIGNATURE----- From carrie at databeast.com Thu Apr 29 02:02:45 2004 From: carrie at databeast.com (carrie) Date: Thu Apr 29 01:02:44 2004 Subject: first time gpg automation with perl Message-ID: <200404290002.45460.carrie@databeast.com> Hi all, I am using shared hosting with an ISP. gpg 1.06 is installed. I have got a ssh login and have used gpg via that in my area of the server and it works exactly as expected. but I am prototyping the script to use in an environment where the only access I have to the command line is via backticks from perl. I have encountered some trouble though and which has eventually resulted in the following script which I wrote to see if it would help me resolve the problem #!/usr/bin/perl my $gpgargs = " --version"; my $sentstring = "gpg $gpgargs"; my $returnedstring = `gpg $gpgargs`; my $returnedstring2 = `echo "hello carrie"`; my $returnedstring3 = `pwd`; my $returnedstring4 = `whoami`; my $returnedstring5 = `ls -l`; my $returnedstring6 = `gpg --list-keys`; print "Content-type: text/html\n\n"; print <

Message:

  • xxxx $sentstring xxxx $returnedstring
    xxx echo "hello carrie" xxx $returnedstring2
    pwd xxx $returnedstring3
    whoami xxx $returnedstring4
    gpg --list-keys xxx $returnedstring6
    ls -l xxx $returnedstring5
HTML the output demonstrates my trouble Message: * xxxx gpg --version xxxx gpg (GnuPG) 1.0.6 Copyright (C) 2001 Free Software Foundation, Inc. This program comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it under certain conditions. See the file COPYING for details. Home: ~/.gnupg Supported algorithms: Cipher: 3DES, CAST5, BLOWFISH, RIJNDAEL, RIJNDAEL192, RIJNDAEL256, TWOFISH Pubkey: RSA, RSA-E, RSA-S, ELG-E, DSA, ELG Hash: MD5, SHA1, RIPEMD160 xxx echo "hello carrie" xxx hello carrie pwd xxx /home/carrie/cgi-bin whoami xxx carrie gpg --list-keys xxx ls -l xxx total 96 -rwxr-xr-x 1 carrie carrie 1384 Apr 28 18:41 databeast.key -rw-r--r-- 1 carrie carri everything works as you would expect except --list-keys and the other gpg action commands. But I know that I have the priviliges to run these commands because I can run them from my ssh connection. and the script reported that the directory and user (carrie) are the same as they would be from the ssh shell. So why willt it let me execute some commands in gpg but not others? is there a security setting ? maybe something that stops script access? any other ideas? TIA Carrie From rainman at hod.aarg.net Thu Apr 29 08:46:43 2004 From: rainman at hod.aarg.net (Denis Green) Date: Thu Apr 29 08:43:52 2004 Subject: re. Moving from PGP to GnuPG and other questions ... Message-ID: <200404290646.i3T6khJq032119@www.aarg.net> Dear Neil Thanks for this detailed response. >re. Moving from PGP to GnuPG and other questions ... >Neil Williams linux at codehelp.co.uk >Wed Apr 28 18:07:44 CEST 2004 > > >On Wednesday 28 April 2004 3:14, Denis Green wrote: >> 4/ >> When I try to encrypt files using GPGtools, I get to >> select the file(to be encrypted), but when the >> " GPG Tools - Encrypt " window opens up, I don't see >> any public keys in the dialog box > >That's because no public keys are set as trusted. If you >have your own secret key, import both the secret and public >key and then use --edit-key to set your own key to ultimate >trust. done >GnuPG will check through the other keys in the >keyring and will only let you encrypt to those that can be >trusted (without using command-line options intended for >secure environments). These will be keys that you have >already signed or keys that are signed by people you have >signed, etc. > A little confused here. Is sign = Trust ? or Because when I use the "edit" function from GPG Keys window, I get a help list, which has both sign and trust as separate commands. Keys that I sign seem to be the ones I am allowed to encrypt to ? (am I right there ?) >There's no sense allowing encryption of sensitive data to >a key that cannot be trusted! Not very sure of what you mean by trusting a key ? Is this the same as trusting a person behind that key ? Then Oh ! no, there a are set of keys that I'd like to encrypt to ...but I have no knowledge of the person bend the key, either to trust them or ... NOT to trust them For e.g. a re-mailer key ... I've explained my thoughts on siging, later in this post >If you override the GnuPG >security using '--trust-model always' or similar, you must >still verify the key in some other manner. > Lets say I've got some 500 Public keys on my PGP keyring Lets say I may end up using some 100 of these public keys pretty regularly - like Nym keys, re-mailer keys, list member keys, news reporter's keys etc. Theoretically, in quite a few some cases, I'm not even expected to know who these people are and so the question of trusting or not trusting **the person** doesn't arise .. and a practical question .. ----------------------------------- How do I get to use GPG to encrypt to such keys ? Is there a command to sign all 100 keys in 1 go Can I choose multiple keys selectively with this command ? >Trust begins with the secret keys - those are presumably >yours so if you have the pass-phrase, these should be set as >ultimate trust. I tend to consider ultimate trust as only >for keys with a usable secret key. Ok. Set ultimate trust to 1 of my secret keys , for starters >All other keys then have >their trust calculated as starting from your ultimate keys, >fully trusted (allowing encryption), marginally trusted >(needing an override but still not recommended) or trust >unknown (don't encrypt to these unless you *really* know >what you are doing). Other trust factors like revoked and >expired are hopefully clear in their meaning. > I took a sample key. Set it to fully trusted (just one level below ultimate trust). Still I had to sign this key before I could encrypt to this key ...some additional questions .. -------------------------------------- 1/ I had imported all the 500+ public keys and my private key*s* to GPG (from PGP). The process went thru a 500 Y/N sequence ... but was ultimately completed. Then while trying to edit a key, I inadvertently deleted it (I had signed this key earlier - to test encrypting to this key) Since I still had this public key on my PGP ring, I tried importing this key from PGP (copy from clipboard - asc export from PGP and import from GnupG keys etc. etc.] Here's what I get every time I try to import **just** this key -------------------------------------------------------------- pub 1024D/abcdefgh.. created: 2002-06-15 expires: never Key fingerprint = aaaa bbbb cccc dddd eeee ffff gggg hhhh iiii jjjj abcdesfgh ... Do you want to import this key? (y/N) y gpg: renaming `C:/gnudirectory\pubring.gpg' to `C:/gnudirectory\pubring.bak' failed: Permission denied [***comment added : I have full rights to this directory. Checked the same by creating, renaming, deleting files in the directory **] gpg: error writing keyring `C:/gnudirectory\pubring.gpg': file rename error gpg: key abcdefgh ..: public key "[User id not found]" imported gpg: error reading `c:\docume~1\adminname\locals~1\temp\~gpgtray.asc': file rename error gpg: import from `c:\docume~1\adminname\locals~1\temp\~gpgtray.asc' failed: file rename error [ comment added : Tried searching for gpgtray.asc ... can't find the file on the machine !] gpg: Total number processed: 0 gpg: imported: 1 [comment added : Though it says 1 imported, I can't see_this_key in GuPG keys] Press any key to continue . . . -------------------------------------------------------------- 2/ I have multiple private keys. Lets say I have key pars a - used in office (in the Full name and office e mail id) b - used for nym and re-mailers (with my nym address) c - used in personal communications (with my short name and personal address) 3/ How do I choose different private_ keys to sign different public keys ? For e.g. my colleague's key with _my_private_(a), a re-mailer ops key with _my_private_(b) and my brother's key with _my_private__(c) and so on ? 4/ Don't we import public keys - basically to encrypt to them (i.e.) with an intention to use them. How else would a public key get into my key ring ? (without me getting a public key block and me importing it by certain willful actions, without me searching the key servers and importing etc.. etc..) so why should I go through another process of signing a key ? (i.e.) forgetting the cryptography involved behind the process of signing, as a user, why should I do a second action (call it signing or stamping or whatever **after** willfully importing a public key ? Thanks DG From stuart.yeates at computing-services.oxford.ac.uk Thu Apr 29 09:25:26 2004 From: stuart.yeates at computing-services.oxford.ac.uk (Stuart A Yeates) Date: Thu Apr 29 09:24:47 2004 Subject: first time gpg automation with perl In-Reply-To: <200404290002.45460.carrie@databeast.com> References: <200404290002.45460.carrie@databeast.com> Message-ID: <4090ADE6.40508@computing-services.oxford.ac.uk> carrie wrote: > > everything works as you would expect except --list-keys and the other gpg > action commands. But I know that I have the priviliges to run these commands > because I can run them from my ssh connection. and the script reported that > the directory and user (carrie) are the same as they would be from the ssh > shell. So why willt it let me execute some commands in gpg but not others? > is there a security setting ? > maybe something that stops script access? > > any other ideas? I'm guessing that some of the environmental variables are different. The man page lists $HOME, $GNUPGHOME and $GPG_AGENT_INFO as variables the gpg uses. Maybe you could check that these are set as you expect? cheers stuart -- Stuart Yeates stuart.yeates@computing-services.oxford.ac.uk OSS Watch http://www.oss-watch.ac.uk/ Oxford Text Archive http://ota.ahds.ac.uk/ Humbul Humanities Hub http://www.humbul.ac.uk/ From samuel at Update.UU.SE Thu Apr 29 10:30:32 2004 From: samuel at Update.UU.SE (Samuel ]slund) Date: Thu Apr 29 10:28:22 2004 Subject: re. Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404290646.i3T6khJq032119@www.aarg.net> References: <200404290646.i3T6khJq032119@www.aarg.net> Message-ID: <20040429083032.GR27722@Update.UU.SE> On Wed, Apr 28, 2004 at 11:46:43PM -0700, Denis Green wrote: > << Loads of questions basicly about what is the difference between Signing and trusting a key >> I have not been active for a while but if I remember correctly you sign keys that you have personaly verified belongs to who ever it says it belongs to, you trust a key owner to verify keys in a good way when signing them. That means that any key you have signed is a key you can encrypt to. To be able to encrypt to a key you have not signed some other key that you have signed and who's owner you trust to verify keys correctly must have signed that "unknown" key. Chains of these kinds of verifications is what builds the Web of Trust. About verifying keys before using them, there are several reasons to import keys that you can not be shure has not been modified in transmission. One reason is to check signatures on a mailing list, who actuauly is writing is not as interesting as knowing that it is the same entity all the time. Another could be that you have no way to check it but think that some of your acquaintances has, then you can import it and let GnuPG check if that is true, see Web of Trust. About "Ultimate trust", I think that is GnuPG's way of knowing which keys you know well enough to start the chains of signing. Earlier versions did this by assuming that if the program had the secret key it was secure enough. Using the ultimate trust setting instead allows (among other things) a user to keep the secret key in a secure place and still use the public key to start a chain in the web of trust. HTH //Samuel From wk at gnupg.org Thu Apr 29 11:07:26 2004 From: wk at gnupg.org (Werner Koch) Date: Thu Apr 29 10:50:43 2004 Subject: re. Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404281707.51219.linux@codehelp.co.uk> (Neil Williams's message of "Wed, 28 Apr 2004 17:07:44 +0100") References: <200404281414.i3SEEjWo021233@www.aarg.net> <200404281707.51219.linux@codehelp.co.uk> Message-ID: <87wu3zm9td.fsf@vigenere.g10code.de> On Wed, 28 Apr 2004 17:07:44 +0100, Neil Williams said: > BTW. General question: will '--trust-model always' allow encryption to a > revoked key? i.e. after you've imported/refreshed the key with it's > revocation certificate. (I'd guess that it won't but haven't got any revoked > keys to test against.) No, it should not. > Think of 'free software' as one term, not two words. Than;s why I prefer to write is like a proper name: Free Software Salam-Shalom, Werner From malsyned at uofr.net Thu Apr 29 15:30:04 2004 From: malsyned at uofr.net (Dennis Lambe Jr.) Date: Thu Apr 29 15:27:18 2004 Subject: re. Moving from PGP to GnuPG and other questions ... In-Reply-To: <200404290646.i3T6khJq032119@www.aarg.net> References: <200404290646.i3T6khJq032119@www.aarg.net> Message-ID: <1083245404.6912.15.camel@localhost> On Thu, 2004-04-29 at 02:46, Denis Green wrote: > A little confused here. > > Is sign = Trust ? or no. > Because when I use the "edit" function from GPG Keys > window, I get a help list, which has both sign and trust as > separate commands. Keys that I sign seem to be the ones I > am allowed to encrypt to ? (am I right there ?) Signing a key means that you believe that the key belongs to the person it says it does. Trusting a key means you trust that person's competency in signing other keys. If you don't sign a key, you may still be able to encrypt to it if you trust someone who has signed that key. > 4/ > Don't we import public keys - basically to encrypt to them > (i.e.) with an intention to use them. How else would a > public key get into my key ring ? (without me getting a > public key block and me importing it by certain willful > actions, without me searching the key servers and importing > etc.. etc..) > > so why should I go through another process of signing a key ? > > (i.e.) forgetting the cryptography involved behind the > process of signing, as a user, why should I do a second > action (call it signing or stamping or whatever **after** > willfully importing a public key ? The mistake you're making is assuming that you always have to sign a key to encrypt to it. This is not the case. You have to /either/ sign the key /or/ trust one of its signers. Let's take a real-world example: You have downloaded Lewis Powell's key (72007281), and then signed it (after duly verifying that it actually belongs to him via, for example, meeting him at a key-signing party and personally checking his government-issued ID and his key fingerprint). You know Lewis, and you know that he will only sign a key if he has made sure that it belongs to the person it says it does, so you've assigned him full trust (4). You get an email from me, Dennis Lambe (F53BA904). You download my key from a keyserver in order to verify the signature. Since you trust Lewis, and Lewis has signed my key, you don't need to sign my key in order to be confident that I am who I say I am. GPG will not complain that it can't be sure the key really belongs to me, since it knows Lewis is sure, and it knows that you trust Lewis's diligence at signing keys. This is all necessary because without either checking that the person who owns the key matches the name on the key, or trusting someone who has, there's really no telling who uploaded the key to the key server. I could create a key right now with a UID of "Dennis Green " and post it to a keyserver. If GPG let anyone encrypt to that key without a further verification step, then I could read all your email (and you couldn't). I hope that clears things up. --D -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 279 bytes Desc: This is a digitally signed message part Url : /pipermail/attachments/20040429/c3cbbcb8/attachment.bin From ewheaton at rapidstability.com Thu Apr 29 16:15:08 2004 From: ewheaton at rapidstability.com (Eli Wheaton) Date: Thu Apr 29 16:12:47 2004 Subject: --batch example Message-ID: Hi All, Could someone give me an example on how the --batch option works. Thanks in advance -eli From morse at sinauer.com Thu Apr 29 17:38:27 2004 From: morse at sinauer.com (Parker Morse) Date: Thu Apr 29 17:35:44 2004 Subject: Encrypting from web form to file: permissions error Message-ID: <4243DC51-99F3-11D8-A6C9-0003938174B4@sinauer.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'm working on encrypting the output of a form on our website to a file, with the idea of then downloading the files from the web server and decrypting locally. I'm using PHP, like so (edited and line-wrapped for email): shell_exec("echo $for_encryption | /usr/bin/gpg -e --no-secmem-warning - -ao \ /path/to/file.asc -r \"Keyname\""); I have successfully taken form output and encrypted it to an email message with a different form of this command. However, the line above now generates this error in the server error log: gpg: fatal: /root/.gnupg: can't create directory: Permission denied secmem usage: 0/0 bytes in 0/0 blocks of pool 0/16384 ...and the output of the shell_exec() above is nothing. Now, PHP is running as "nobody", so I'm not surprised that it can't create a directory in /root/. But why does it feel that it needs to? Isn't ~/.gnupg enough? How can I fix this? If this is a PHP problem, not a GPG problem, please let me know. Thanks in advance, pjm -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (Darwin) iD8DBQFAkSF4nRVGoRROKxIRAkvrAJ4xt4lyqz+1aLV/NaTLyWSbUifNnQCfVD2Y Wpm7QaFA8Jv/Ywc3+x6RKQs= =zOT+ -----END PGP SIGNATURE----- From hmujtaba at forumsys.com Thu Apr 29 17:54:38 2004 From: hmujtaba at forumsys.com (Hasnain Mujtaba) Date: Thu Apr 29 17:52:17 2004 Subject: interoperabilit test files Message-ID: <4DCE15B9C4E66F4CA967EBF64C53D64D1902ED@bstn-exch1.forumsys.com> Hi, I was doing some compatibility testing between GPG and OpenPGP tool that I am working with. Are there any test files that the GPG group uses to test interoperability that you can share with me? Appreciate it Hasnain. ---- The information contained in this electronic mail and any attached document is the confidential and proprietary business information of Forum Systems, Inc. It is intended solely for the addressed recipient listed above. It may not be distributed in any manner without the express written consent of Forum Systems, Inc. From jeff+gnupg at jeffenstein.org Thu Apr 29 19:11:41 2004 From: jeff+gnupg at jeffenstein.org (Jeff Fisher) Date: Thu Apr 29 19:09:14 2004 Subject: re. Moving from PGP to GnuPG and other questions ... In-Reply-To: <1083245404.6912.15.camel@localhost> References: <200404290646.i3T6khJq032119@www.aarg.net> <1083245404.6912.15.camel@localhost> Message-ID: <20040429171141.GC28312@frogger.jeffnet> On Thu, Apr 29, 2004 at 09:30:04AM -0400, Dennis Lambe Jr. wrote: > > You have downloaded Lewis Powell's key (72007281), and then signed it > (after duly verifying that it actually belongs to him via, for example, > meeting him at a key-signing party and personally checking his > government-issued ID and his key fingerprint). You know Lewis, and you > know that he will only sign a key if he has made sure that it belongs to > the person it says it does, so you've assigned him full trust (4). This is where my confusion comes from with signing keys... Yes, I can sign the key for Lewis Powell, after verifying that it is indeed a Lewis Powell that owns the key. However, a google search for "Lewis Powell" returns about 8,000 hits, including a Supreme Court justice, a criminal on death row, and a "Lewis Powell" award (and the few more I'll add by putting his name in this e-mail). I can safely assume he's not the one in jail, but he could be one of hundreds. Which "Jeff Fisher" am I? The one running for Congress in Florida, the american football coach, or yet another one? The reality is that his name, which is the identifying information everybody stresses, is far from unique. Yes, there may be only a handful of these "Lewis Powell"s that use pgp, but that is relying on chance more than anything. There is also the e-mail address in the user id field, which should be unique and relatively constant, but little or none of the information on signing keys mentions verifying that the e-mail address is actually this person. Yes, it's possible that someone else will take that e-mail address, but if several e-mail addresses are listed on the key, you can be relatively sure that you can still reach the person. So, (finally), the question is, in practice what's the use of verifying only the name of the person before signing their key? Personally, I've only signed the keys of people I know personally. However, after a bit of thought, this gives me pause to sign anybody else's key with or without a passport, or reason to sign keys based on only e-mail correspondence. -- jeff@jeffenstein.org http://www.jeffenstein.org/ Reality is that which, when you stop believing in it, doesn't go away. -- Philip K. Dick -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 793 bytes Desc: not available Url : /pipermail/attachments/20040429/c42b3271/attachment.bin From atom-gpg at suspicious.org Thu Apr 29 20:15:55 2004 From: atom-gpg at suspicious.org (Atom 'Smasher') Date: Thu Apr 29 20:13:19 2004 Subject: Encrypting from web form to file: permissions error In-Reply-To: <4243DC51-99F3-11D8-A6C9-0003938174B4@sinauer.com> References: <4243DC51-99F3-11D8-A6C9-0003938174B4@sinauer.com> Message-ID: > I'm working on encrypting the output of a form on our website to a > file, with the idea of then downloading the files from the web server > and decrypting locally. I'm using PHP, like so (edited and line-wrapped > for email): > > shell_exec("echo $for_encryption | /usr/bin/gpg -e --no-secmem-warning > -ao \ > /path/to/file.asc -r \"Keyname\""); ========================================== check out this app, and read the how-to... http://Business-PHP.com/opensource/gpg_encrytpt/ it should do what you need. in (almost) any case, you'll need to set the gpg directory to be writable by the web browser. ...atom _________________________________________ PGP key - http://atom.smasher.org/pgp.txt 3EBE 2810 30AE 601D 54B2 4A90 9C28 0BBF 3D7D 41E3 ------------------------------------------------- "If Jesus had been killed 20 years ago, Catholic school children would be wearing little electric chairs around their necks instead of crosses" -- Lenny Bruce From jerry.windrel at verizon.net Thu Apr 29 20:51:56 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Thu Apr 29 20:47:44 2004 Subject: Which Lewis Powell References: <200404290646.i3T6khJq032119@www.aarg.net><1083245404.6912.15.camel@localhost> <20040429171141.GC28312@frogger.jeffnet> Message-ID: <00d301c42e1b$0c020470$6401a8c0@Windows> The problem of multiple people with the same name is addressed by the ability to add a photo to your key (and have the photo signed) as well as the ability to add userids that include email addresses (also signed). From linux at codehelp.co.uk Thu Apr 29 23:28:40 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Thu Apr 29 23:24:57 2004 Subject: re. Moving from PGP to GnuPG and other questions ... In-Reply-To: <20040429171141.GC28312@frogger.jeffnet> References: <200404290646.i3T6khJq032119@www.aarg.net> <1083245404.6912.15.camel@localhost> <20040429171141.GC28312@frogger.jeffnet> Message-ID: <200404292228.45701.linux@codehelp.co.uk> On Thursday 29 April 2004 6:11, Jeff Fisher wrote: > On Thu, Apr 29, 2004 at 09:30:04AM -0400, Dennis Lambe Jr. wrote: > However, a google search for "Lewis Powell" returns about 8,000 hits, Keysigning is never advisable with partial data. Signing a key is a public declaration - you are publicly stating that you have no reason to doubt that the physical person who owns the secret key is precisely the same person identified in the key data. "Key data" in this sense should be defined as GnuPG itself deals with key data: The name, comment and email address are lumped together as one User ID which can have secondary UID's added as aliases for that one person. In turn, I may want to rely on your verification of that person, in combination with other signatures, to be able to trust an otherwise unknown key. In practice, it's never good to trust a single link in the chain and I set most keys to marginal trust unless I know the person personally. As mentioned elsewhere, signing = proof of your verification of that person and that key. trust = your assessment of how carefully that person will verify other keys. You need to be sure of all three components of a UID, although comments might not seem important, they can be used to express important details for some users. I always check: 1. The fingerprint - this is the basis of key identification - it's how you tell two similar keys apart if they have the same names and even the same/similar emails. 2. Photo ID - reputable and recognised source, not necessarily official or government but not easily forgeable or temporary. Must include the same name details as the key. 3. email verification, preferably off-list and using encryption. 4. comments: if the user specifically mentions a project or locality, perhaps a distinguishing facet - this should be checked during the face-to-face meeting to verify photo ID and key fingerprint. > So, (finally), the question is, in practice what's the use of verifying > only the name of the person before signing their key? No use whatsoever. The keysigning must verify all parts of the UID. If you cannot verify any part of a UID, do NOT sign that part of that key - use the uid command (uid 1 or uid 2 etc.) to select those UID's that you have been able to verify. Take my key. If you are unable to verify me as webmaster of DCLUG for whatever reason, you should not sign that UID. > Personally, I've only signed the keys of people I know personally. However, > after a bit of thought, this gives me pause to sign anybody else's key with > or without a passport, or reason to sign keys based on only e-mail > correspondence. I don't sign keys without seeing photo ID, exchange of fingerprint and some verification of the email. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040429/5737aad2/attachment.bin From linux at codehelp.co.uk Thu Apr 29 23:32:25 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Thu Apr 29 23:28:32 2004 Subject: first time gpg automation with perl In-Reply-To: <200404290002.45460.carrie@databeast.com> References: <200404290002.45460.carrie@databeast.com> Message-ID: <200404292232.25920.linux@codehelp.co.uk> On Thursday 29 April 2004 1:02, carrie wrote: > Hi all, > I am using shared hosting with an ISP. gpg 1.06 is installed. > > I have got a ssh login and have used gpg via that in my area of the server > and it works exactly as expected. > > but I am prototyping the script to use in an environment where the only > access I have to the command line is via backticks from perl. Yuk. There are Perl modules available from CPAN that will interface with GnuPG on your behalf in a reliable and tested manner. I've used GnuPG::Interface and it allows complete gpg key management and use. (You'd be advised to think carefully about allowing secret keys on remote machines before considering signing / decrypting!) Take a look at CPAN, there are other modules out there and they can all be installed in 'user space' even if your webhost doesn't like to install them. That's unlikely because if they've taken the time to install gpg as a package, they may already have installed a Perl module to use gpg. Check your webhost config, look for a perldiver.cgi script and try to find out which modules are already installed. Full examples are on CPAN. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040429/5cb5007b/attachment.bin From linux at codehelp.co.uk Fri Apr 30 00:19:23 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Fri Apr 30 00:15:39 2004 Subject: trustdb calculations Message-ID: <200404292319.28815.linux@codehelp.co.uk> gpg: checking at depth 0 signed=23 ot(-/q/n/m/f/u)=0/0/0/0/0/2 I understand this line to mean gpg found 2 keys set to ultimate trust (I have two keys of my own with matching secret keys) that have, in turn signed 23 other keys. Is that right? gpg: checking at depth 1 signed=85 ot(-/q/n/m/f/u)=0/0/0/1/22/0 gpg: checking at depth 2 signed=224 ot(-/q/n/m/f/u)=0/0/0/60/16/0 That jump to 224 - is that expected? Have I perhaps been too generous 'as-a-rule'? I tend to set keys at marginal trust during a run of --update-trustdb - gpg only lists those that it can fully trust but which have no trust value set. Up until now, I've been happy to consider these as keys belonging to people who have a reputation for good key management (to collect the signatures from the strong set that allow my gpg installation to trust the key) and that it is reasonable to set these to marginal trust unless I have reason to set full trust or have reason not to trust. Could this have led to an artificial increase in keys further down the chain becoming trusted when perhaps there isn't a good reason? i.e. it would enable someone to become fully trusted when 3 marginal keys have signed it, even though each of those 3 is as much as 3 links down the chain from me. The problem is that it is very hard to visualise these chains and to tell from the update-trustdb output just how far away you are from the key being queried. I've got to the stage now that trustdb checks have to be run manually because my mail client was waiting many minutes for the results of importing a new key. Now I've set no-auto-check-trustdb in gpg.conf and I run update-trustdb once all mail has been previewed (and new keys imported). Have I over-complicated my trustdb with trusted keys too far down the chain? (possibly by over-zealous use of the KGpg function "import missing keys from keyserver" on a trusted Debian developer key? - it requests and imports all keys that have signed the selected key but which are not already in the keyring. Up until now, I've only done it with keys with 5-8 missing signatures. Last night I did it with a stronger key and imported a lot more keys, most of which also became trusted.) Latest run of --check-trustdb gives: neil@garfield:~$ date Thu Apr 29 23:04:29 BST 2004 neil@garfield:~$ gpg --check-trustdb gpg: checking at depth 0 signed=23 ot(-/q/n/m/f/u)=0/0/0/0/0/2 gpg: checking at depth 1 signed=85 ot(-/q/n/m/f/u)=0/0/0/1/22/0 gpg: checking at depth 2 signed=224 ot(-/q/n/m/f/u)=0/0/0/60/16/0 gpg: checking at depth 3 signed=224 ot(-/q/n/m/f/u)=0/1/0/113/11/0 gpg: checking at depth 4 signed=154 ot(-/q/n/m/f/u)=0/0/0/84/4/0 gpg: next trustdb check due at 2004-05-14 neil@garfield:~$ date Thu Apr 29 23:09:22 BST 2004 Almost exactly 5 minutes. (This after a previous run of --update-trustdb and with no further changes to the keyring.) neil@garfield:~$ gpg --list-key | grep -c pub 471 471 keys after running my usual 'untrusted/expired/revoked' clear-out script. I'm now considering adding 'marginal' to the clear-out script - when a key is deemed by gpg to be of marginal trust, my email client still classifies it as untrusted. Am I right in thinking that until a key becomes fully trusted by gpg, it has no bearing on the trustdb calculations? If I delete all keys that are m/[m|-|n|q] will some of my fully trusted keys suddenly become untrusted? If I delete these keys, will it improve the trustdb speed at all? I'd suspect that the speed is down to the sheer number of verifiable signatures on the trusted keys - is that correct? Some of these are currently only marginal trust (trust m/m). -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040429/15d322bf/attachment.bin From linux at codehelp.co.uk Fri Apr 30 00:31:46 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Fri Apr 30 00:28:00 2004 Subject: an we delete (our own) posts sent here earlier ? In-Reply-To: <200404271615.i3RGFICN025964@www.aarg.net> References: <200404271615.i3RGFICN025964@www.aarg.net> Message-ID: <200404292331.50215.linux@codehelp.co.uk> On Tuesday 27 April 2004 5:15, Denis Green wrote: > Hi folks : > One more question that I did not add to my earlier mail... > 1/ > How do I delete a post that I sent to this list (if need > be ?) You can't. By the time you receive it back it's been sent to everyone else and has already been archived with all the others. I'm certainly not going to allow you to delete email that has already arrived in my inbox! > 2/ > for e.g. I've mentioned the full e mail address for my key > - autoresponder in the last post. I'd like to modify it as > usual.. talking away the @ and replacing it with the _at_ > etc.. > > Can I do that ? If so how ? Too late. You might be able to sweet-talk to webmaster for some archives but techniques like that are not spammer-proof (which is presumably your concern) because newsgroup users have been doing this for some time and it's easy to write a script that will re-form the original email address. However, whether the gnupg archives are actually a target for spammers trawling for email addresses is open to doubt. Besides, the archive already declares your verified email address in this format: Neil Williams linux at codehelp.co.uk - whether you quote your email address in your message or not. http://lists.gnupg.org/pipermail/gnupg-users/2004-April/022432.html That message was only sent 12 minutes before this one and it's already in the public archive for the entire internet to read. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040429/197ee50a/attachment.bin From huber+gpg at alum.wpi.edu Thu Apr 29 22:49:54 2004 From: huber+gpg at alum.wpi.edu (Josh Huber) Date: Fri Apr 30 01:26:16 2004 Subject: re. Moving from PGP to GnuPG and other questions ... In-Reply-To: <20040429171141.GC28312@frogger.jeffnet> (Jeff Fisher's message of "Thu, 29 Apr 2004 19:11:41 +0200") References: <200404290646.i3T6khJq032119@www.aarg.net> <1083245404.6912.15.camel@localhost> <20040429171141.GC28312@frogger.jeffnet> Message-ID: <87k6zygy7w.fsf@amalthea.paradoxical.net> Jeff Fisher writes: > [...] > There is also the e-mail address in the user id field, which should > be unique and relatively constant, but little or none of the > information on signing keys mentions verifying that the e-mail > address is actually this person. Yes, it's possible that someone > else will take that e-mail address, but if several e-mail addresses > are listed on the key, you can be relatively sure that you can still > reach the person. Before I sign someone's key, I like to verify the email addresses associated with each UID. I do this with a small emacs "plugin" which uses Gnus to generate and send encrypted challenge messages to each uid, given a keyid. It's here: http://www.paradoxical.net/~huber/gpg-party.el For each uid, I generate a random challenge string. I'll only sign a uid if I get a matching challenge string back. Maybe this is overkill, but I like it! -- Josh Huber -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 188 bytes Desc: not available Url : /pipermail/attachments/20040429/867a5663/attachment.bin From DougB at dougbarton.net Fri Apr 30 02:02:23 2004 From: DougB at dougbarton.net (Doug Barton) Date: Fri Apr 30 02:00:03 2004 Subject: re. Moving from PGP to GnuPG and other questions ... In-Reply-To: <87k6zygy7w.fsf@amalthea.paradoxical.net> References: <200404290646.i3T6khJq032119@www.aarg.net> <1083245404.6912.15.camel@localhost> <20040429171141.GC28312@frogger.jeffnet> <87k6zygy7w.fsf@amalthea.paradoxical.net> Message-ID: <4091978F.5000902@dougbarton.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Another way to approach this is to sign the key, then encrypt it with itself and mail it to the address specified in the identity. This accomplishes verification that the e-mail address is valid, verification that the person at the other end has control of the secret key, and puts the responsibility of uploading the signed key on them. This method gets a little cumbersome if the person has multiple identities on the same key, but in that situation I generally just sign each identity individually and mail that version of the signed key to that e-mail address. HTH, Doug - -- If you're never wrong, you're not trying hard enough -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) iD4DBQFAkZePyIakK9Wy8PsRAv6iAJjCND0LG6/xBcazQDF9HpyyttBiAKDyYKjg J0PYQKl7sTFRmUDZKKyMrw== =u9DE -----END PGP SIGNATURE----- From turner_bill at sbcglobal.net Fri Apr 30 11:29:54 2004 From: turner_bill at sbcglobal.net (Bill Turner) Date: Fri Apr 30 11:28:20 2004 Subject: Revoking Old Keys... my problem Message-ID: <40921C92.9070105@sbcglobal.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I have a problem I am not sure how to deal with. Put as plainly as I know how, there is an old key out there, I just got this information off the pgp.mit.edu keyserver: Public Key Server -- Verbose Index ``wildbill@speakeasy.net '' Type bits /keyID Date User ID pub 1024D/89F6CC2B 2002/10/18 Bill Turner sig 89F6CC2B Bill Turner I have attached my own 'new' key to this email. The reason that the old key is a problem for me is that the laptop I had it stored on, and everything else, was stolen from me. I do not remember my passphrase after almost two years. I am just not sure how to go about 'properly' revoking it. The email address that was based on is at an ISP I no longer use, not that it matters really. I am just confused about the 'correct' procedure here. There are people I was exchanging email with that have that key, and not the new one, so I just am not clear how to do this. Any help much appreciated. Bill -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (MingW32) iD8DBQFAkhyOp9drXXqFz2gRArQvAJsEtrhkL95Vt++wYt+cooSVVAkEkQCg8guK 2ru1f1AYPV0wKY4kl5a25mw= =lIZh -----END PGP SIGNATURE----- -------------- next part -------------- -----BEGIN PGP PUBLIC KEY BLOCK----- Version: GnuPG v1.2.4 (MingW32) mQGiBECPm7YRBAChISjonsRuhgBJIx3JQ3H64Ej/cXV2RpV00nTlKe8e/N6vlu6z GUgPQm1SfgcZ61+92oppgRL55wanUcY39feg9YgZmjwEtngScz+rODfENlZT8nCu ZQifM/lVEq6Oyhhv9mUD4iX6FcsMjp9eEAyLw8vHX4vVOlE5kH2ztOqaRwCg9h2o nS2sYDH74lm3Vh3PDp8TOk8EAJfSFd3ck5FpdYxGVF+5oFsX4RrtSYc6DxVBA9eh w9igrQoWC62GZrvZSZ9yNGB0TYKu/vp9D6g8c5XUiw1aHsIFcRRbQtGoNBUY4JN4 0/2056HkAv7jG4oxXWlKBOqUcc/hhouF6W37bWDEeU3QysRbuisG9NDShkff356/ FO9GA/4v8uwd1hACh//I1x3KG+Y0meuEEdjGcEZNvlFdgF/D00Km/vvETrvyA12j dEcKFb/YBpzevjZxJoDav1Ac3bdrkEm0wahD+uiXsRLM+l9VpB/DDgLdBPO0z0X/ /je1KB6YFUf12+6d9B3gkf/MM+YJAYD8ApmhTo2kZxg4Dx7OErQyQmlsbCBUdXJu ZXIgKFR1eCBSb3ghKSA8dHVybmVyX2JpbGxAc2JjZ2xvYmFsLm5ldD6IXgQTEQIA HgUCQI+btgIbAwYLCQgHAwIDFQIDAxYCAQIeAQIXgAAKCRCn12tdeoXPaCrpAJ4o 32Epku53jURzlyW3UeEGyWDzHgCgvcy0Wy045onp4oCjME7eHlO0rrm5AQ0EQI+b yBAEAMoDoyWOPMxxQFLl7ng228NM6VZszjD9l9oPNYLqtUKit/sG066ynsmYN4Ez QmJPUh8Uqtwh54dsSYoc4P0QoVIeEsjzF/NDlWLqTiRkuYbXb6Ul18D98gAl1jxd Qb2Mrtkh7FfUTRDcu5iWV5nSEQykPikniFP8yieW+d0ZcZoLAAMGA/9IgfcvCxnW YJCJVroa9chXkpzeM5HBcaOVzeADYsrYiFIvseQsbBrz2MUr8OChIcUzLyffL2d7 THpiZSYu34VouAZhVMfc5Np6aEzYWsXYxey3cGKCA3qfa98hKeXXgL40OHy9oLOZ NFYu+pYiFzqfUY6BkobVBLd+xP7xysdES4hJBBgRAgAJBQJAj5vIAhsMAAoJEKfX a116hc9oKPoAoKCmf2HRcP2aA3rWA6IinW6dXgz5AJoCK91mSOfDLA8xcwNH89w1 es8KQpkBogQ9a32vEQQA+ezBmafwTtIpNXEDdicjm3IkwFb7MydTZys1hUTulhav +f51Wo1PfnU4CA2Imfg/BOkJCr2gIp8cI/Qgqc0lNMXGKocLDaGoUGo4wdrpo1HD G/y/E78qi0YLCP3YojY2BINQjcCdEPyPpR9gvRuHzQ17Rhsj5HAoQM0jizx/3UMA oP8tZKqrQnepXICXnGQNzbRZ+HutA/9yldx9bmSlzd44G38QRS/yk6wip/NB+jkU kio3mEFjMxifYsm0aZc+ugl3LWc92B94o4dR3oVH1tnqzS7bAh9E+SK9JJrDV9N4 8otTR0wnrvB0yp1JGNCZvCHp0Lj9DHySw9td28KaADehjVcXmQR8aArvYQT5PRDf vrdiWu6vpwP/aN3bM0/tDUFmtRSBEl43xzEdloXmmjM8Xi/Pt6ImtAkegpk/x+h3 kVePCqP1IJGrladbs6sHPpJ2r/lNwqUc8FvxagyOIZrj4xhjnifrDkbn9cQIUAwM Ug+QI53QWvJMS0frcmQTs3SiNsygYM2Aw8Ngc8jjQJOnrHeB1j4RSva0I0JyZW5k YW4gS2lkd2VsbCA8YnJlbmRhbkBnbHVtcC5uZXQ+iFgEEBECABgFAj1rfa8ICwMJ CAcCAQoCGQEFGwMAAAAACgkQ4lxlBKPKA3i/QQCfeq90eipHejX5dKKRl5I3sfr6 0UcAnAmhCAtBf/tcE7usBRr6h2S+F+dBuQINBD1rfa8QCAD2Qle3CH8IF3Kiutap QvMF6PlTETlPtvFuuUs4INoBp1ajFOmPQFXz0AfGy0OplK33TGSGSfgMg71l6RfU odNQ+PVZX9x2Uk89PY3bzpnhV5JZzf24rnRPxfx2vIPFRzBhznzJZv8V+bv9kV7H AarTW56NoKVyOtQa8L9GAFgr5fSI/VhOSdvNILSd5JEHNmszbDgNRR0PfIizHHxb LY7288kjwEPwpVsYjY67VYy4XTjTNP18F1dDox0YbN4zISy1Kv884bEpQBgRjXyE pwpy1obEAxnIByl6ypUM2Zafq9AKUJsCRtMIPWakXUGfnHy9iUsiGSa6q6Jew1Xp Mgs7AAICCAC0KIKI2jb3yCNzYIerVwp7lzorZRQocph2dAzSliTSKg27WpTu9ljQ LqGnYA1XtdcpQrgCB/D0TqFmkkZitJTe5d5g4uAgdUuq9foWLPZmFhI3xINcyo5E p3uKLS9PRG9M8pehlMPT3TeiaglEr5HaswOFFadqVDSKmkG4Qql6SbC+5UKlbcoJ T4wNyktpGPhpyEqkREqO/SvsstJI9y5qmeBdnFdLbvfFJJ0PIQ5Y4OevhpaIeJxu 8qt7u348kuQ5vVU6jeeRTTiwyZEeY8Xp0HM/gZwMq2z4Hu6rfbSEyeLHPguOlBoh E27050d4Eo/iru9/PMvnG82xBcP95J/piEwEGBECAAwFAj1rfa8FGwwAAAAACgkQ 4lxlBKPKA3gM9QCeOMtfPhlNvIixDuToe//5lE4UjOUAoKRziBXMrgnzR276R9sn 9DZpXjrw =jP2v -----END PGP PUBLIC KEY BLOCK----- From avbidder at fortytwo.ch Fri Apr 30 12:46:52 2004 From: avbidder at fortytwo.ch (Adrian 'Dagurashibanipal' von Bidder) Date: Fri Apr 30 12:44:03 2004 Subject: Revoking Old Keys... my problem In-Reply-To: <40921C92.9070105@sbcglobal.net> References: <40921C92.9070105@sbcglobal.net> Message-ID: <200404301246.57772@fortytwo.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday 30 April 2004 11.29, Bill Turner wrote: > I have attached my own 'new' key to this email. The reason that the > old key is a problem for me is that the laptop I had it stored on, > and everything else, was stolen from me. I do not remember my > passphrase after almost two years. I am just not sure how to go > about 'properly' revoking it. > You'll not like that: the proper procedure is - never lose your key or forget your passphrase - never use a key when you don't have a revocation certificate prepared for such emergencies. The only thing you can do: notify all people you suspect of having the old key that it was compromised, announce your new key. Of course, you'll have to redo all the key-signing you've done on the old key and get people to verify that this is really your new key. Also, RUN and generate revocation certificates for all keys you use NOW. Store them on floppy, and print them out (seriously. You'll not be able to read that floppy when you need it.) Store this in a secure location. So if ever something like that happens again, you can still revoke that key. For the same reason, it makes sense to always set an expiration date on all keys you use (I use 10 years - which is long enough not to cause trouble for me for a long time, but I won't have old keys floating around indefinitely if for some reason the secret key and the emergency revocation cert both are killed.) As for the old key on the keyservers: tough luck, there's nothing you can do. You may have luck convincing one or two keyserver operators that they should delete your key, but since the keyservers are networked, and since anybody can just re-upload your old key, it would keep popping up again and again. So, since you've changed your email address, you can just take that lesson about key handling and then forget about it all - in a few years nobody will even remember that it was you with that key (a common name like yours has advantages, I see - I'd have the additional problem that there isn't anyone with the same name, as far as I know:-) greetings - -- vbi - -- featured link: http://www.pool.ntp.org -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: get my key from http://fortytwo.ch/gpg/92082481 iKcEARECAGcFAkCSLqBgGmh0dHA6Ly9mb3J0eXR3by5jaC9sZWdhbC9ncGcvZW1h aWwuMjAwMjA4MjI/dmVyc2lvbj0xLjUmbWQ1c3VtPTVkZmY4NjhkMTE4NDMyNzYw NzFiMjVlYjcwMDZkYTNlAAoJECqqZti935l6FtYAn0JMS5Wx6kVrukLCrTF3nCSU Ki+ZAJ9zdBQLNQCHQCeuHtfhYH2jyl9EBw== =+orq -----END PGP SIGNATURE----- From linux at codehelp.co.uk Fri Apr 30 13:22:16 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Fri Apr 30 13:18:45 2004 Subject: Revoking Old Keys... my problem In-Reply-To: <40921C92.9070105@sbcglobal.net> References: <40921C92.9070105@sbcglobal.net> Message-ID: <200404301222.22705.linux@codehelp.co.uk> On Friday 30 April 2004 10:29, Bill Turner wrote: > Hello, > > I have a problem I am not sure how to deal with. Put as plainly as I > know how, there is an old key out there, > > The reason that the old > key is a problem for me is that the laptop I had it stored on, and > everything else, was stolen from me. I do not remember my passphrase > after almost two years. I am just not sure how to go about 'properly' > revoking it. A passphrase (even if you could remember it) is useless without the secret key. If you don't have a backup of the secret key from that laptop, forget trying to remember the passphrase (if you follow) because you cannot use it to revoke this key. If you also do not have a revocation certificate (GnuPG advice is to create one immediately after generating the new key, print it out (v.small) and delete the file) then this key is doomed to hang around on keyservers for ever with no realistic possibility of being revoked or deleted. > The email address that was based on is at an ISP I no longer use, not > that it matters really. I am just confused about the 'correct' procedure Correct procedure is to always have a revocation certificate stored somewhere v.safe in case you lose the use of the secret key. The certificate can be used by anyone, it just needs to be imported into a keyring that already holds the public key to be revoked - no secret key or passphrase is required to use the revocation certificate, which is why it must be kept safe! Then, the revoked key should be sent to the keyservers and the keyserver merge the two and show the key as revoked. If you have no revocation certificate and no backup of the secret key, you haven't got a prayer. You can't really ask for the correct procedure now when you've already ignored the correct procedure that would have prevented the problem in the first place! :-) > here. There are people I was exchanging email with that have that key, > and not the new one, so I just am not clear how to do this. Tell them directly - and QUICKLY, they could be justifiably upset that you hadn't told them immediately the laptop was stolen! Explain that you were careless not to have the revocation certificate or a backup and that the old key must never be used again. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040430/7ab376ed/attachment.bin From markus_kampkoetter at t-online.de Fri Apr 30 13:19:12 2004 From: markus_kampkoetter at t-online.de (markus =?iso-8859-1?Q?kampk=F6tter?=) Date: Fri Apr 30 13:24:07 2004 Subject: Revoking Old Keys... my problem References: <40921C92.9070105@sbcglobal.net> <200404301246.57772@fortytwo.ch> Message-ID: <1BJW5K-0R0MnQ0@fwd03.sul.t-online.com> > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Friday 30 April 2004 11.29, Bill Turner wrote: > > > I have attached my own 'new' key to this email. The reason that the > > old key is a problem for me is that the laptop I had it stored on, > > and everything else, was stolen from me. I do not remember my > > passphrase after almost two years. I am just not sure how to go > > about 'properly' revoking it. > > > > You'll not like that: the proper procedure is > - never lose your key or forget your passphrase > - never use a key when you don't have a revocation certificate prepared > for such emergencies. (snipped good advice ;) you may as well add to the comment line of your new key that your old key (quote keyID of lost key) is not valid anymore. hth markus From jerry.windrel at verizon.net Fri Apr 30 16:51:06 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Fri Apr 30 16:46:47 2004 Subject: Revoking Old Keys... my problem References: <40921C92.9070105@sbcglobal.net> <200404301246.57772@fortytwo.ch> <1BJW5K-0R0MnQ0@fwd03.sul.t-online.com> Message-ID: <004301c42ec2$9164ec20$6401a8c0@Windows> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Here's some good news for you, after all that bad news. It's not perfect and I may not be remembering it perfectly either, but I think it's a big help, although it relies on exploiting a flaw in the protocol of key servers, which I read about recently. Maybe others on the list can fill in the blanks of this method. 1) Download the bad public key from a key server. 2) Export it in binary form (not ASCII armored). 3) Get a binary editor (HexEdit is a shareware product that I use... there are many others) 4) Using the binary edtior, alter the User ID, i.e. your real name "John Smith " (or 1 of them, if there are many... DO NOT change more than one!), to say something like, "Bad Key, use 0x1234ABCD". Unfortunately, you will have to be brief, as you cannot use more characters than were in the original User ID. You will be somewhat lucky if your name is long :) If you're lucky enough to have more than enough room, replace the unused characters at the end with spaces. Technically, there may be a way to make it longer, but that would require really getting into the binary format of export files. Don't forget to save the binary file. 5) Import your lobotomized binary export file. 6) You should be able to see the new "User ID" in place of the old one. 7) Upload the key. 8) Search on the key server to see the result. 9) You might expect the key server to also show the old userid replaced by the "Bad Key..." userid, but that's NOT what will happen. The old userid will remain (since you haven't done anything to delete it from the key server, and deleting it from key servers is not usually even possible), but you will have a new userid, namely "Bad Key...". Now whoever looks at the userids for your key will see that "Bad Key..." warning. It's not the ideal solution, which is to get rid of the bad key, but, key servers are quite an apt metaphor to life itself... sadly, there are mistakes that can be made that just cannot be undone. Seen in that context, having un unrevokable key, while frustrating, quite pales in comparison to the possible other irreversable mistakes that can be made in life. Chalk it up to experience. -----BEGIN PGP SIGNATURE----- Version: PGP 8.0.3 - not licensed for commercial use: www.pgp.com iQA/AwUBQJJnwYlVuABWWiqVEQIGWgCgzuQHeU0gGZ9KazRbPfuXyWqrZMIAnRxF wXnDs5HBg1lki60wF5KW/M4S =pWra -----END PGP SIGNATURE----- From jerry.windrel at verizon.net Fri Apr 30 16:58:26 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Fri Apr 30 16:54:10 2004 Subject: OT: Revoking Old Keys... my problem Message-ID: <005701c42ec3$97c99f10$6401a8c0@Windows> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At the risk of veering off topic... The problem of having a key that you cannot revoke, and the partial solution I outlined, reminds me somewhat of the situation in the Book of Esther where a king sent out a proclamation signed with his signet ring. The rule in those days that a proclamation signed with the king's signet ring could never be revoked, not even by the king himself (similar to the "non-repudiation" property of digital signatures). When the king later regretted that proclamation, the only solution was to send out another (non-revokable) proclamation that mitigated the effect of the first one. -----BEGIN PGP SIGNATURE----- Version: PGP 8.0.3 - not licensed for commercial use: www.pgp.com iQA/AwUBQJJpgYlVuABWWiqVEQIT/QCfUjyqzZrcEbYH5k0ggt+r25kY2NsAoJ4R iKPRFHg4TFqYiCTjsIBIFRqx =23Xb -----END PGP SIGNATURE----- From linux at codehelp.co.uk Fri Apr 30 17:58:43 2004 From: linux at codehelp.co.uk (Neil Williams) Date: Fri Apr 30 17:56:25 2004 Subject: Revoking Old Keys... my problem In-Reply-To: <004301c42ec2$9164ec20$6401a8c0@Windows> References: <40921C92.9070105@sbcglobal.net> <1BJW5K-0R0MnQ0@fwd03.sul.t-online.com> <004301c42ec2$9164ec20$6401a8c0@Windows> Message-ID: <200404301658.55169.linux@codehelp.co.uk> On Friday 30 April 2004 3:51, Jerry Windrel wrote: > 4) Using the binary edtior, alter the User ID, i.e. your real name > "John Smith " (or 1 of them, if there are many... DO > NOT change more than one!), to say something like, "Bad Key, use > 0x1234ABCD". ?Unfortunately, you will have to be brief, as you cannot > use more characters than were in the original User ID. ?You will be > somewhat lucky if your name is long :) If you're lucky enough to have > more than enough room, replace the unused characters at the end with > spaces. ?Technically, there may be a way to make it longer, but that > would require really getting into the binary format of export files. > Don't forget to save the binary file. > 5) Import your lobotomized binary export file. Why would GnuPG allow a forged key to be imported? It must break the self-sig on the key or it could be used to add a false UID to a genuine key! from man gpg: --allow-non-selfsigned-uid --no-allow-non-selfsigned-uid Allow the import and use of keys with user IDs which are not self-signed. This is not recommended, as a non self-signed user ID is trivial to forge. --no-allow-non-selfsigned-uid disables. > 6) You should be able to see the new "User ID" in place of the old > one. I'd hope not! -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: signature Url : /pipermail/attachments/20040430/aa0a73e2/attachment.bin From jerry.windrel at verizon.net Fri Apr 30 18:14:44 2004 From: jerry.windrel at verizon.net (Jerry Windrel) Date: Fri Apr 30 18:26:09 2004 Subject: Revoking Old Keys... my problem References: <40921C92.9070105@sbcglobal.net><1BJW5K-0R0MnQ0@fwd03.sul.t-online.com><004301c42ec2$9164ec20$6401a8c0@Windows> <200404301658.55169.linux@codehelp.co.uk> Message-ID: <007e01c42ece$408cb380$6401a8c0@Windows> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 >Why would GnuPG allow a forged key to be imported? It must break the >self-sig on the key or it could be used to add a false UID to a >genuine key! I think it would be imported as a new UID that is not signed. Rather than speculate, why doesn't the person affected by the situation try what I've suggested (perhaps waiting a few days first to see if any others in the group have improvements, contraindications, etc.) and we can all see what happens? -----BEGIN PGP SIGNATURE----- Version: PGP 8.0.3 - not licensed for commercial use: www.pgp.com iQA/AwUBQJJ7YYlVuABWWiqVEQIQIACg+MNiCPDd4B8h6iUGHt0Pq4uAz1YAoJkU K95mvQShTfL+0eJCN4FbphLV =L/W6 -----END PGP SIGNATURE-----