From schimmer at cg.cs.tu-bs.de Thu Aug 5 14:20:13 2004 From: schimmer at cg.cs.tu-bs.de (Lars Schimmer) Date: Thu Aug 5 04:20:33 2004 Subject: [SA-exim] sa-exim-stats and author-no answer? Message-ID: <411217ED.2080604@cg.cs.tu-bs.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi! I just found two "bugs/features" in sa-exim-stats that are really annoying to me. So I send two reports to the author noticed in the script, but til now there were no answer. Anyone knows something about the author is still well? Cya Lars - -- - ----------------------------------------------------------------- Technische Universit?t Braunschweig, Institut f?r Computergraphik Tel.: +49 531 391-2109 E-Mail: schimmer@cg.cs.tu-bs.de PGP-Key-ID: 0xB87A0E03 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Using GnuPG with Debian - http://enigmail.mozdev.org iD8DBQFBEhfsVguzrLh6DgMRAt5UAJwISOGIKWMTZ1rzz6B3blRgN9kQyQCgqCu6 lHKh2zrAAh0k7OyGNuPqR7M= =j7NB -----END PGP SIGNATURE----- From marc at merlins.org Sat Aug 7 17:42:14 2004 From: marc at merlins.org (Marc MERLIN) Date: Sat Aug 7 16:42:20 2004 Subject: [SA-exim] SA 3 and sa-exim 4.x In-Reply-To: <20040729191616.GZ20117@merlins.org> References: <20040729190822.GC29989@nuit.ca> <20040729191616.GZ20117@merlins.org> Message-ID: <20040807234214.GA18842@merlins.org> On Thu, Jul 29, 2004 at 12:16:16PM -0700, Marc MERLIN wrote: > On Thu, Jul 29, 2004 at 07:08:26PM +0000, simon@nuit.ca wrote: > > 1BqFlt-XXXXXX-XX SA: PANIC: SA: could not parse X-Spam-Status: to extract hits and required. Bad!. Got: 'X-Spam-Status: No, score=-2.4 required=5.5 tests=BAYES_00,NO_REAL_NAME autolearn=no version=3.0.0-pre2' (but message was accepted) > > Mmmh, it used to be hits= instead of score= > > I'll have to change that in the next sa-exim to support both. > In the meantime, feel free to change it in the source: s/hits/score/ This should be fixed in sa-exim cvs now (untested) Try http://marc.merlins.org/linux/exim/files/sa-exim-cvs.tar.gz Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems & security .... .... what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 307 bytes Desc: Digital signature Url : http://lists.merlins.org/archives/sa-exim/attachments/20040807/8d436296/attachment-0001.bin From simon at nuit.ca Sun Aug 8 01:23:07 2004 From: simon at nuit.ca (simon@nuit.ca) Date: Sat Aug 7 17:23:35 2004 Subject: [SA-exim] SA 3 and sa-exim 4.x In-Reply-To: <20040807234214.GA18842@merlins.org> References: <20040729190822.GC29989@nuit.ca> <20040729191616.GZ20117@merlins.org> <20040807234214.GA18842@merlins.org> Message-ID: <20040808002307.GH15970@nuit.ca> Ce jour Sat, 07 Aug 2004, Marc MERLIN a dit: > > Mmmh, it used to be hits= instead of score= > > > > I'll have to change that in the next sa-exim to support both. > > In the meantime, feel free to change it in the source: s/hits/score/ > > This should be fixed in sa-exim cvs now (untested) > Try http://marc.merlins.org/linux/exim/files/sa-exim-cvs.tar.gz cool, thanks > Marc > -- > "A mouse is a device used to point at the xterm you want to type in" - A.S.R. > Microsoft is to operating systems & security .... > .... what McDonalds is to gourmet cooking > Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key -- Software Patents are patently wrong: http://swpat.ffii.org/papiere/eubsa-swpat0202/ustr0309/index.en.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 652 bytes Desc: Digital signature Url : http://lists.merlins.org/archives/sa-exim/attachments/20040808/8993774c/attachment.bin From roland at astrofoto.org Fri Aug 13 23:47:03 2004 From: roland at astrofoto.org (Roland Roberts) Date: Fri Aug 13 19:47:14 2004 Subject: [SA-exim] Bypassing SA for a particular user Message-ID: -----BEGIN PGP SIGNED MESSAGE----- I'm trying to set up a local spam/nospam/white-/black-list addresses where I can forward messages to feed into sa-learn or spamassassin. I have some perl code that will filter the forward to get the original message back, and putting a little procmail recipe in place is easy. The hard part, for me anyway, it that i can't quite see how to tell exim not to run SA for a particular user. I'm still a newbie at exim, having installed it only a bit more than a week ago. Tied in with that is I'd like to allow these aliases to only accept mail from the internal network, not from the external, in order to prevent poisoning by external agents. Can someone point me to the magic config commands? A pointer to the right place in the FM would be great.... regards, roland - -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland@rlenter.com 6818 Madeline Court roland@astrofoto.org Brooklyn, NY 11220 -----BEGIN PGP SIGNATURE----- Version: 2.6.3ia Charset: noconv Comment: Processed by Mailcrypt 3.5.4, an Emacs/PGP interface iQCVAwUBQR19J+oW38lmvDvNAQHsPgQAnbtnXWfMEmqU2UG8YRc/3KqrzxYGQKCH CdNdG9sBP8cHNAx7s1/+E4sh89hUnLVZWsYHg+4Iw3aIzWUz8Ukm77l8V75ZnsoL PUAUMR2POJD66jlvhtldtusAHjG09xZ+kyi9SAEcsZBia2qym+ICKW5q6eEkSzmH OLFpmAGJhZM= =kda5 -----END PGP SIGNATURE----- From ssmeenk at freshdot.net Sun Aug 15 17:54:32 2004 From: ssmeenk at freshdot.net (Sander Smeenk) Date: Sun Aug 15 07:54:43 2004 Subject: [SA-exim] [FWD] sa-exim_4.0-3_i386.changes ACCEPTED Message-ID: <20040815145432.GJ30324@freshdot.net> Accepted: sa-exim_4.0-3.diff.gz to pool/main/s/sa-exim/sa-exim_4.0-3.diff.gz sa-exim_4.0-3.dsc to pool/main/s/sa-exim/sa-exim_4.0-3.dsc sa-exim_4.0-3_i386.deb to pool/main/s/sa-exim/sa-exim_4.0-3_i386.deb Announcing to debian-devel-changes@lists.debian.org Closing bugs: 246715 250726 251755 252341 253393 265820 -- | A box withouth hinges, key, or lid, yet golden treasure inside is hid. | 1024D/08CEC94D - 34B3 3314 B146 E13C 70C8 9BDB D463 7E41 08CE C94D From ssmeenk at freshdot.net Sun Aug 15 18:26:56 2004 From: ssmeenk at freshdot.net (Sander Smeenk) Date: Sun Aug 15 08:27:06 2004 Subject: [SA-exim] [FWD] sa-exim_4.0-3_i386.changes ACCEPTED In-Reply-To: <20040815145432.GJ30324@freshdot.net> References: <20040815145432.GJ30324@freshdot.net> Message-ID: <20040815152656.GL30324@freshdot.net> Quoting Sander Smeenk (ssmeenk@freshdot.net): > Accepted: > sa-exim_4.0-3_i386.deb > to pool/main/s/sa-exim/sa-exim_4.0-3_i386.deb I forgot the changelog :) Changes: sa-exim (4.0-3) unstable; urgency=low . + Changed hard depend on spamc, since that's all sa-exim needs. You have to choose to install spamassassin, or run spamc to another host. This entry closes: #253393 + Documentation was assed to explain about sa-exim defaulting to exim4 configured to use exim4/use_split_config. Devine intervention is needed when using monolithic configs. One day sa-exim might detect this and act on it? This entry closes: #251755, #265820 + Incorrect scoring of messages due to incorrect last-Received: header is now fixed by newer versions of the exim4 package. Dependencies have been changed. This entry closes: #250726, #246715 + Updated Japanese translations of debconf templates. A typo in the english templates has been fixed too. This entry closes: #252341 -- | Genealogy: Tracing yourself back to better people. | 1024D/08CEC94D - 34B3 3314 B146 E13C 70C8 9BDB D463 7E41 08CE C94D From marc at merlins.org Sun Aug 15 23:39:45 2004 From: marc at merlins.org (Marc MERLIN) Date: Sun Aug 15 22:39:55 2004 Subject: [SA-exim] Weird headers In-Reply-To: <20040714232035.GR22900@merlins.org> References: <40F42627.8080304@team.ar.inter.net> <20040713184611.GB22900@merlins.org> <40F43109.1080207@team.ar.inter.net> <20040713203604.GE22900@merlins.org> <40F454C2.7020004@team.ar.inter.net> <20040713213647.GF22900@merlins.org> <40F45971.8060107@team.ar.inter.net> <20040713232740.GH22900@merlins.org> <40F591CC.2030905@team.ar.inter.net> <20040714232035.GR22900@merlins.org> Message-ID: <20040816053945.GR19131@merlins.org> On Wed, Jul 14, 2004 at 04:20:35PM -0700, Marc MERLIN wrote: > Ok, so if you are really receiving a mail like this, things are really > fucked up :) > > Admittedly sa-exim should deal with it. In the meantime, you can tell exim > to refuse mails with such long headers: > header_line_maxsize = 7k > should do it The following patch, now in CVS, should fix this in sa-exim: RCS file: /cvsroot/sa-exim/sa-exim/sa-exim.c,v retrieving revision 1.64 diff -u -u -r1.64 sa-exim.c --- sa-exim.c 16 Aug 2004 05:12:16 -0000 1.64 +++ sa-exim.c 16 Aug 2004 05:39:00 -0000 @@ -418,8 +418,18 @@ { break; } - /* Slight waste of memory here, oh well... */ - *header=string_sprintf("%s\n%s", *header, buffer); + + /* Guard against humongous header lines */ + if (strlen(*header) < 8000) + { + /* Slight waste of memory here, oh well... */ + *header=string_sprintf("%s\n%s", *header, buffer); + } + else + { + log_write(0, LOG_MAIN, "SA: Warning: while parsing header %s, ignoring the following trailing line due to header size overflow: %s", headername, buffer); + + } } if (SAEximDebug > 5) { Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems & security .... .... what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key From marc at merlins.org Mon Aug 16 10:19:29 2004 From: marc at merlins.org (Marc MERLIN) Date: Mon Aug 16 09:19:30 2004 Subject: [SA-exim] SA-Exim 4.1 released In-Reply-To: <20040316175920.GE25947@merlins.org> References: <20040316175920.GE25947@merlins.org> Message-ID: <20040816161929.GA23781@merlins.org> Just when you stopped hoping, and started making bets on how many more years it would take for the next release, it's here finally http://marc.merlins.org/linux/exim/sa.html http://sourceforge.net/projects/sa-exim/ Deb package is here: http://marc.merlins.org/linux/exim/files/debian/ (compiled against unstable, you might have to rebuild for testing, or wait for official deb packages from Sander Smeenk, although I'm pretty sure they're in sync right now) The changelist is here: * Deal with being called with the number of recipients is 0 (reported by John Horne) * Made hopefully fully 64 bit clean (well, it was written that way, except for that pesky printf in C, which got in the way). Kudos go to Adam D. Barratt for finding and explaining the bug * Switched to using strchr instead of index to remove compilation warnings on some systems (reported by John Horne/Jeff Carter) * Added SAspamcHost/SAspamcPort at the request of several people * Fixed mbox From file to be standards complaint (reported by Cliff Hones) * Added SAFEMESGIDCHARS option for safer message-id based filenames * New version of the SA patch, with safe file creation (clean env from and rcpt to before using as directory names) * IMPORTANT: if you are using greylisting, and applied the previous greylisting patch, it didn't create shell safe filenames, it is therefore highly recommended that you apply the new version * Added a quick patch to sa-exim to deal with headers that are longer than 8K (and drop the remaining lines) As explained there, if you are using the old SA patch, you should really upgrade to the new one in order to get shell safe filenames (the creation was safe, but dealing with them with shell scripts to expire records was potentially unsafe depending on how you did it) Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems & security .... .... what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key From ssmeenk at freshdot.net Tue Aug 17 12:27:14 2004 From: ssmeenk at freshdot.net (Sander Smeenk) Date: Tue Aug 17 02:27:27 2004 Subject: [SA-exim] [FWD] sa-exim_4.1-1_i386.changes ACCEPTED Message-ID: <20040817092714.GA12938@freshdot.net> Accepted: sa-exim_4.1-1.diff.gz to pool/main/s/sa-exim/sa-exim_4.1-1.diff.gz sa-exim_4.1-1.dsc to pool/main/s/sa-exim/sa-exim_4.1-1.dsc sa-exim_4.1-1_i386.deb to pool/main/s/sa-exim/sa-exim_4.1-1_i386.deb sa-exim_4.1.orig.tar.gz to pool/main/s/sa-exim/sa-exim_4.1.orig.tar.gz Announcing to debian-devel-changes@lists.debian.org Closing bugs: 266257 That bug was a translation wishlistbug :) README.Debian changed to emphasize the SA-patch problem. -- |#include /*** ---- viznut --- http://www.hytti.uku.fi/~vheikkil/ */ |main(v,i,z,n,u,t){for(v=-1;;)for(n=pow(/* gcc -lm sig.c; a.out > /dev/dsp */ |1.06,"`cW`g[`cgcg[eYcb^bV^eW^be^bVecb^"[++v&31]+(v&64)/21),i=999;i;putchar( |128+((8191&u)>i?0:i/8)-((8191&(z+=n))*i-->>16)))u+=v&1?t/2:(t=v&6?t:n/4);} From smanhes at aubay.com Wed Aug 18 16:51:07 2004 From: smanhes at aubay.com (Stephane MANHES) Date: Wed Aug 18 06:51:21 2004 Subject: [SA-exim] Exim Version Message-ID: <000f01c4852a$6a286790$0500a8c0@rapanui> Morning, Excuse me, I just want to know if the sa-exim version ??2004/08/16 - v4.1? is compliant with exim 4.41 and SpamAssassin 2.64 ? I had compilation problem with Exim. Cordialement, St?phane MANHES From marc at merlins.org Wed Aug 18 09:25:10 2004 From: marc at merlins.org (Marc MERLIN) Date: Wed Aug 18 08:25:12 2004 Subject: [SA-exim] Exim Version In-Reply-To: <000f01c4852a$6a286790$0500a8c0@rapanui> References: <000f01c4852a$6a286790$0500a8c0@rapanui> Message-ID: <20040818152510.GP7914@merlins.org> On Wed, Aug 18, 2004 at 03:51:07PM +0200, Stephane MANHES wrote: > Morning, > > Excuse me, I just want to know if the sa-exim version ??2004/08/16 - v4.1? > is compliant with exim 4.41 and SpamAssassin 2.64 ? I had compilation > problem with Exim. It should be. The SA version will not matter for compilation, and exim should work too, although I haven't tested 4.41 Worst case, you can use the exim headers shipped with sa-exim (modify the makefile and comment out EXIM_SRC to only use EXIM_SRC_LOCAL). That said, I think your problem is something else. Why don't you post your compilation errors? Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems & security .... .... what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key From exipat at yahoo.fr Thu Aug 19 13:53:42 2004 From: exipat at yahoo.fr (=?iso-8859-1?q?exi=20pat?=) Date: Thu Aug 19 03:53:52 2004 Subject: [SA-exim] checking user spamlevel Message-ID: <20040819105342.83379.qmail@web25210.mail.ukl.yahoo.com> Hi, I have installed SA and sa-exim all work fine , and I find sa-exim a very good idea to check email during the smtp connection. all is ok like that ,but now , I need to reject emails based on user's spamlevel I think to do this , but don't know if it is a good idea or not, I need some help: 1: enable reject of spam when level is 20 at smtp time and not use perm & temp reject SAdevnull: 20.0 #SApermreject: 12.0 #SAtempreject: 9.0 2: if hits < 20 , the mail goes through and I want to check user spamlevel pref which is defined in ldap how can I do that ? (check the hits value) what do you think of that , and is it possible for me to make this work ? thank you very much Best Regards Patrice Vous manquez d?espace pour stocker vos mails ? Yahoo! Mail vous offre GRATUITEMENT 100 Mo ! Cr?ez votre Yahoo! Mail sur http://fr.benefits.yahoo.com/ Le nouveau Yahoo! Messenger est arriv? ! D?couvrez toutes les nouveaut?s pour dialoguer instantan?ment avec vos amis. A t?l?charger gratuitement sur http://fr.messenger.yahoo.com From exipat at yahoo.fr Thu Aug 19 13:16:15 2004 From: exipat at yahoo.fr (=?iso-8859-1?q?exi=20pat?=) Date: Thu Aug 19 04:49:05 2004 Subject: [SA-exim] checking user spamlevel Message-ID: <20040819101615.95444.qmail@web25203.mail.ukl.yahoo.com> Hi, I have installed SA and sa-exim all work fine , and I find sa-exim a very good idea to check email during the smtp connection. all is ok like that ,but now , I need to reject emails based on user's spamlevel I think to do this , but don't know if it is a good idea or not, I need some help: 1: enable reject of spam when level is 20 at smtp time and not use perm & temp reject SAdevnull: 20.0 #SApermreject: 12.0 #SAtempreject: 9.0 2: if hits < 20 , the mail goes through and I want to check user spamlevel pref which is defined in ldap how can I do that ? (check the hits value) what do you think of that , and is it possible for me to make this work ? thank you very much Best Regards Patrice Vous manquez d?espace pour stocker vos mails ? Yahoo! Mail vous offre GRATUITEMENT 100 Mo ! Cr?ez votre Yahoo! Mail sur http://fr.benefits.yahoo.com/ Le nouveau Yahoo! Messenger est arriv? ! D?couvrez toutes les nouveaut?s pour dialoguer instantan?ment avec vos amis. A t?l?charger gratuitement sur http://fr.messenger.yahoo.com From ssmeenk at freshdot.net Thu Aug 19 16:39:33 2004 From: ssmeenk at freshdot.net (Sander Smeenk) Date: Thu Aug 19 06:39:49 2004 Subject: [SA-exim] checking user spamlevel In-Reply-To: <20040819101615.95444.qmail@web25203.mail.ukl.yahoo.com> References: <20040819101615.95444.qmail@web25203.mail.ukl.yahoo.com> Message-ID: <20040819133933.GK16199@freshdot.net> Quoting exi pat (exipat@yahoo.fr): > 2: if hits < 20 , the mail goes through and I want to > check user spamlevel pref which is defined in ldap > how can I do that ? (check the hits value) sa-exim has to check this to take the correct action, and sa-exim currently does not provide support for user configurable spamlevels, let alone ldap. Your users could (or should) use Exim filters (or procmail?) to check the X-Spam-Something header and act accordingly. HTH, Sander. -- | It is better to regret something you HAVE done, | than to regret something you HAVEN'T done... | 1024D/08CEC94D - 34B3 3314 B146 E13C 70C8 9BDB D463 7E41 08CE C94D From marc at merlins.org Mon Aug 23 09:49:22 2004 From: marc at merlins.org (Marc MERLIN) Date: Mon Aug 23 08:49:24 2004 Subject: [SA-exim] Re: sa-exim v4.1 (doesn't build in standalone mode) In-Reply-To: <0746D96A-F514-11D8-A217-000A95E67602@lothlorien.net> References: <0746D96A-F514-11D8-A217-000A95E67602@lothlorien.net> Message-ID: <20040823154922.GE22509@merlins.org> On Mon, Aug 23, 2004 at 10:52:17AM -0400, Jason John Schwarz wrote: > I noticed that when I tried to move the sa-exim.c to the exim directory > and compile like I have for v4.0 that it failed. It appears the the > define used on line 550 for SAFEMESGIDCHARS is inside the #ifdef > DLOPEN_LOCAL_SCAN. I moved the #ifdef DLOPEN_LOCAL_SCAN down to after the > #ifndef block for SAFEMESGIDCHARS and everything seems to have worked. > This a bug, or just me? It's me being stupid, sorry I didn't test that This should be fixed in CVS now http://marc.merlins.org/linux/exim/files/sa-exim-cvs.tar.gz Thanks Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems & security .... .... what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key From ssmeenk at freshdot.net Mon Aug 23 19:16:11 2004 From: ssmeenk at freshdot.net (Sander Smeenk) Date: Mon Aug 23 09:16:33 2004 Subject: [SA-exim] Re: sa-exim v4.1 (doesn't build in standalone mode) In-Reply-To: <20040823154922.GE22509@merlins.org> References: <0746D96A-F514-11D8-A217-000A95E67602@lothlorien.net> <20040823154922.GE22509@merlins.org> Message-ID: <20040823161611.GU16199@freshdot.net> Quoting Marc MERLIN (marc@merlins.org): > On Mon, Aug 23, 2004 at 10:52:17AM -0400, Jason John Schwarz wrote: > > I noticed that when I tried to move the sa-exim.c to the exim directory > > and compile like I have for v4.0 that it failed. It appears the the > > define used on line 550 for SAFEMESGIDCHARS is inside the #ifdef > > DLOPEN_LOCAL_SCAN. I moved the #ifdef DLOPEN_LOCAL_SCAN down to after the > > #ifndef block for SAFEMESGIDCHARS and everything seems to have worked. > > This a bug, or just me? > It's me being stupid, sorry I didn't test that > This should be fixed in CVS now > http://marc.merlins.org/linux/exim/files/sa-exim-cvs.tar.gz With Debian coming in binary form and the freeze up ahead, i will not package this change :) http://qa.debian.org/developer.php?excuse=sa-exim Regards, Sander. -- | No man has ever been shot while doing the dishes. | 1024D/08CEC94D - 34B3 3314 B146 E13C 70C8 9BDB D463 7E41 08CE C94D From marc at merlins.org Mon Aug 23 10:42:22 2004 From: marc at merlins.org (Marc MERLIN) Date: Mon Aug 23 09:42:23 2004 Subject: [SA-exim] Re: sa-exim v4.1 (doesn't build in standalone mode) In-Reply-To: <20040823161611.GU16199@freshdot.net> References: <0746D96A-F514-11D8-A217-000A95E67602@lothlorien.net> <20040823154922.GE22509@merlins.org> <20040823161611.GU16199@freshdot.net> Message-ID: <20040823164222.GF12371@merlins.org> On Mon, Aug 23, 2004 at 06:16:11PM +0200, Sander Smeenk wrote: > Quoting Marc MERLIN (marc@merlins.org): > > On Mon, Aug 23, 2004 at 10:52:17AM -0400, Jason John Schwarz wrote: > > > I noticed that when I tried to move the sa-exim.c to the exim directory > > > and compile like I have for v4.0 that it failed. It appears the the > > > define used on line 550 for SAFEMESGIDCHARS is inside the #ifdef > > > DLOPEN_LOCAL_SCAN. I moved the #ifdef DLOPEN_LOCAL_SCAN down to after the > > > #ifndef block for SAFEMESGIDCHARS and everything seems to have worked. > > > This a bug, or just me? > > It's me being stupid, sorry I didn't test that > > This should be fixed in CVS now > > http://marc.merlins.org/linux/exim/files/sa-exim-cvs.tar.gz > > With Debian coming in binary form and the freeze up ahead, > i will not package this change :) There is no reason to anyway, this only affects people who hand build exim and sa-exim Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems & security .... .... what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key From millis at faztek.org Thu Aug 26 18:18:42 2004 From: millis at faztek.org (Millis Miller) Date: Thu Aug 26 09:18:55 2004 Subject: [SA-exim] Problems getting sa-exim/exim integration working Message-ID: <20040826161332.M45055@faztek.org> Hi, I am trying to get sa-exim/exim4 integrated, using sa-exim-4.1 and exim- 4.40, with the installation method #1 (over-writing local_scan.c) but I am getting the following error: -------- gcc -c -O -I. -o local_scan.o ../src/local_scan.c ../src/local_scan.c:511: warning: #warning you should not worry about the "might be clobbered by longjmp", see source ../src/local_scan.c: In function `local_scan': ../src/local_scan.c:551: `SAFEMESGIDCHARS' undeclared (first use in this function) ../src/local_scan.c:551: (Each undeclared identifier is reported only once ../src/local_scan.c:551: for each function it appears in.) ../src/local_scan.c:551: parse error before `static' ../src/local_scan.c:706: `SAspamcSockPath' undeclared (first use in this function) make[1]: *** [local_scan.o] Error 1 make[1]: Leaving directory `/usr/local/exim-4.40/build-SunOS5-5.7-sparc' make: *** [go] Error 2 --------- Looking at the file, I can't immediately see why the second error at least occurs: I have follwed the sa-exim instructions and placed the generated sa- exim.h file in the src directory. Any clues as to what might be going wrong? (I am using this method as opposed to #2, localscan, as I am on a Sun platform and got library symbol errors when I went for it that way). Thanks, Millis -- Open WebMail Project (http://openwebmail.org) Debian Project (http://www.debian.org) From roland at astrofoto.org Thu Aug 26 16:37:25 2004 From: roland at astrofoto.org (Roland Roberts) Date: Thu Aug 26 12:37:37 2004 Subject: [SA-exim] How to tell SA not to autolearn certain lists Message-ID: -----BEGIN PGP SIGNED MESSAGE----- I have autolearn turned on, but there are a couple of lists where I don't want autolearn to operate no matter what the score. In fact, I may not even want SA to run against them. First is the spamassassin users list, and second is this one. The spamassassin users list often has spammy bodies because that's the topic of conversation. I don't want the poison the bayes DB by having anything from there autolearned. Apart from turning off autolearn completely for ham, is there a way to do this? I'm looking at SAEximRunCond, but I'm not sure what headers are fair game at that point.... regards, roland - -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland@rlenter.com 6818 Madeline Court roland@astrofoto.org Brooklyn, NY 11220 -----BEGIN PGP SIGNATURE----- Version: 2.6.3ia Charset: noconv Comment: Processed by Mailcrypt 3.5.4, an Emacs/PGP interface iQCVAwUBQS479eoW38lmvDvNAQF/LAQAqSLt/9KaVJ55BzW5ehJd2rLHjqHpNBJG up3VVLdMBA6gJbb/Wpli7K4DFZwlrXgJHekj5O0+ah70oQsU8QRwiLlPfSPDhCaQ /C/BcvmyxKGUgQBZmSspFKbrD6QFZfjn9eifw/ZIu4BucAcrKGTXq1fm5AwjTEV0 TOWkc/HuwK4= =LtaD -----END PGP SIGNATURE----- From marc at merlins.org Thu Aug 26 14:31:54 2004 From: marc at merlins.org (Marc MERLIN) Date: Thu Aug 26 13:31:56 2004 Subject: [SA-exim] Problems getting sa-exim/exim integration working In-Reply-To: <20040826161332.M45055@faztek.org> References: <20040826161332.M45055@faztek.org> Message-ID: <20040826203154.GP28999@merlins.org> On Thu, Aug 26, 2004 at 05:18:42PM +0100, Millis Miller wrote: > Hi, > > I am trying to get sa-exim/exim4 integrated, using sa-exim-4.1 and exim- > 4.40, with the installation method #1 (over-writing local_scan.c) but I am > getting the following error: > > -------- > gcc -c -O -I. -o local_scan.o ../src/local_scan.c > ../src/local_scan.c:511: warning: #warning you should not worry about > the "might be clobbered by longjmp", see source > ../src/local_scan.c: In function `local_scan': > ../src/local_scan.c:551: `SAFEMESGIDCHARS' undeclared (first use in this > function) The first error is known, it's my screwup. It's fixed in http://marc.merlins.org/linux/exim/files/sa-exim-cvs.tar.gz > ../src/local_scan.c:551: (Each undeclared identifier is reported only once > ../src/local_scan.c:551: for each function it appears in.) > ../src/local_scan.c:551: parse error before `static' > ../src/local_scan.c:706: `SAspamcSockPath' undeclared (first use in this This shouldn't be an error, it's probably just triggered by the previous line Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems & security .... .... what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key From ssmeenk at freshdot.net Thu Aug 26 23:37:29 2004 From: ssmeenk at freshdot.net (Sander Smeenk) Date: Thu Aug 26 13:37:43 2004 Subject: [SA-exim] How to tell SA not to autolearn certain lists Message-ID: <20040826203729.GC13129@freshdot.net> Quoting Roland Roberts (roland@astrofoto.org): > Apart from turning off autolearn completely for ham, is there a way to > do this? I'm looking at SAEximRunCond, but I'm not sure what headers > are fair game at that point.... I can only show how I switch sa-exim on and off for certain domains by using acl_* variables in Exim, and in the SAEximRunCond expansion. It's not a plug-in solution for your problem, but it might get you started: In my acl_check_rcpt, I have: | warn condition = <.. condition ..> | set acl_m0 = run-sa And in SAEximRunCond I added: | {eq{$acl_m0}{run-sa}} to the string. This gives me the possibility to check headers and values while still "inside exim namespace", that may not be available to the sa-exim namespace when running localscan. The acl_m* variables *are* in fact available for checking in the SAEximRunCond expansion. Note that this solution will not work for 'enabling or disabling sa-exim for certain users', since the RCPT TO list of addresses will not be handled separately. HTH, Sander. -- | My mind not only wanders, it sometimes leaves completely. | 1024D/08CEC94D - 34B3 3314 B146 E13C 70C8 9BDB D463 7E41 08CE C94D From millis at faztek.org Fri Aug 27 00:04:31 2004 From: millis at faztek.org (Millis Miller) Date: Thu Aug 26 15:04:43 2004 Subject: [SA-exim] Problems getting sa-exim/exim integration working In-Reply-To: <20040826203154.GP28999@merlins.org> References: <20040826161332.M45055@faztek.org> <20040826203154.GP28999@merlins.org> Message-ID: <20040826220346.M79463@faztek.org> Indeed, that was the problem: both compilation errors are now fixed and, what's more, it's even working! :) Many thanks, Millis -- Open WebMail Project (http://openwebmail.org) Debian Project (http://www.debian.org) ---------- Original Message ----------- From: Marc MERLIN To: Millis Miller Cc: sa-exim@lists.merlins.org Sent: Thu, 26 Aug 2004 13:31:54 -0700 Subject: Re: [SA-exim] Problems getting sa-exim/exim integration working > On Thu, Aug 26, 2004 at 05:18:42PM +0100, Millis Miller wrote: > > Hi, > > > > I am trying to get sa-exim/exim4 integrated, using sa-exim-4.1 and exim- > > 4.40, with the installation method #1 (over-writing local_scan.c) but I am > > getting the following error: > > > > -------- > > gcc -c -O -I. -o local_scan.o ../src/local_scan.c > > ../src/local_scan.c:511: warning: #warning you should not worry about > > the "might be clobbered by longjmp", see source > > ../src/local_scan.c: In function `local_scan': > > ../src/local_scan.c:551: `SAFEMESGIDCHARS' undeclared (first use in this > > function) > > The first error is known, it's my screwup. It's fixed in > http://marc.merlins.org/linux/exim/files/sa-exim-cvs.tar.gz > > > ../src/local_scan.c:551: (Each undeclared identifier is reported only once > > ../src/local_scan.c:551: for each function it appears in.) > > ../src/local_scan.c:551: parse error before `static' > > ../src/local_scan.c:706: `SAspamcSockPath' undeclared (first use in this > > This shouldn't be an error, it's probably just triggered by the previous > line > > Marc > -- > "A mouse is a device used to point at the xterm you want to type in" > - A.S.R. Microsoft is to operating systems & security .... > .... what McDonalds is to gourmet cooking > Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org > for PGP key ------- End of Original Message ------- From roland at astrofoto.org Fri Aug 27 15:32:40 2004 From: roland at astrofoto.org (Roland Roberts) Date: Fri Aug 27 11:32:51 2004 Subject: [SA-exim] How to tell SA not to autolearn certain lists In-Reply-To: <20040826203729.GC13129@freshdot.net> (Sander Smeenk's message of "Thu, 26 Aug 2004 22:37:29 +0200") References: <20040826203729.GC13129@freshdot.net> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- >>>>> "Sander" == Sander Smeenk writes: Sander> Note that this solution will not work for 'enabling or Sander> disabling sa-exim for certain users', since the RCPT TO Sander> list of addresses will not be handled separately. Hmmm, I forgot about how RCPT TO complicates it and was thinking of message to a single sender. But I think your solution will work fine for me since I'm more interested in the per-list issue. I just don't want anything to the spamassassin users mailing list autolearned, and I think it is reasonable to not bother scanning that particular list either. Autolearn seems to be a difficult thing to get right. I just had stuff come in on one list, where a member apparently has been virus infected, and it got autolearned as ham :-( This is one of those cases where having it show up in my inbox is not *too* bad, but I really would like to avoid the autolearn regardless of the scan under "certain circumstances" which are easier to figure out in hindsight. E.g., if there is a MS binary attached, do not autolearn.... Thanks, roland - -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland@rlenter.com 6818 Madeline Court roland@astrofoto.org Brooklyn, NY 11220 -----BEGIN PGP SIGNATURE----- Version: 2.6.3ia Charset: noconv Comment: Processed by Mailcrypt 3.5.4, an Emacs/PGP interface iQCVAwUBQS9+SOoW38lmvDvNAQFGkAQAsCZomkqDAO4doRl2xa7KaoRdS3+fBEI5 X989SJXePcan7DLDYya5RATzSSrVLq9uzv03zQtotjmwbUckoK9ApbTXBz+yJkQb 2Ysq9eQtQ9nLWB/Ig609hYd02Rw2E8OwjnbvAu6Nhr4dtQtoHA8VX7bSTJ9SvL1f T3RyGNwLzaI= =ZqTC -----END PGP SIGNATURE-----