From: The M. P. <sy...@an...> - 2008-06-13 22:21:45
|
... ciao: "Milter insert (1): header: Authentication-Results: aniota.com; sender-id=none header.from=us...@co...; spf=none smtp.mfrom=us...@co..." ".cox.net" listed in '-a /etc/mail/sid.accept'. with '-r 3' under 0.6, that domain ignored. a question. if 'none' treated like a 'fail', is there any way other than '-a', to deal with domains that don't publish spf records ... as an aside: you had asked on sourceforge about the os, well, it linux 2.2.26, with gcc version egcs-2.91.66. i suspect, the "in_addr_t" problem , mostly mine ... |
From: Jeff A. <an...@he...> - 2008-06-14 04:23:27
|
I want to echo this problem with -a local-host-names.... I'm getting... Jun 13 03:02:09 paris sm-mta[83527]: m5DA24KT083527: from=<ro...@pa...>, size=1723, class=0, nrcpts=1, msgid=<200...@pa...>, proto=ESMTP, daemon=IPv4, relay=localhost.hesiod.org [127.0.0.1] Jun 13 03:02:09 paris sm-mta[83527]: m5DA24KT083527: Milter insert (1): header: Authentication-Results: paris.hesiod.org; sender-id=fail (NotPermitted) header.from=ro...@pa...; spf=fail (NotPermitted) smtp.mfrom=ro...@pa... But my mail local-host-names file specified with -a has localhost localhost.hesiod.org paris.hesiod.org and many others. It looks like -a doesn't work with v1.0 sid-filter paris.hesiod.org:anton[5]: uname -a FreeBSD paris.hesiod.org 6.3-RELEASE-p1 FreeBSD 6.3-RELEASE-p1 #2: Mon Mar 17 17:57:25 PDT 2008 ro...@at...:/usr/src/sys/i386/compile/PARIS i386 paris.hesiod.org:anton[6]: gcc -v Using built-in specs. Configured with: FreeBSD/i386 system compiler Thread model: posix gcc version 3.4.6 [FreeBSD] 20060305 Anton The Mighty Phlabaud wrote: > ... ciao: > > "Milter insert (1): header: Authentication-Results: aniota.com; > sender-id=none header.from=us...@co...; spf=none > smtp.mfrom=us...@co..." > > ".cox.net" listed in '-a /etc/mail/sid.accept'. > > with '-r 3' under 0.6, that domain ignored. > > > a question. if 'none' treated like a 'fail', is there any way other > than '-a', to deal with domains that don't publish spf records ... > > |
From: Murray S. K. <ms...@se...> - 2008-06-14 23:24:50
|
OK, I'll see if I can simulate this in the coming week. If there's a bug, I'll work on getting a patch out. |
From: The M. P. <sy...@an...> - 2008-06-19 01:44:39
|
... ciao: > On Wed, 18 Jun 2008, Murray S. Kucherawy wrote: > I've opened bug #1997251 on SourceForge, and attached a patch that appears > to fix it. Let me know if it works for you. patch took, with no compile errors. "Jun 18 17:59:42 yossarian sid-filter[28339]: smfi_register() failed" when the sendmail script tried to invoke the filter ... |
From: Murray S. K. <ms...@se...> - 2008-06-19 16:13:47
|
On Wed, 18 Jun 2008, The Mighty Phlabaud wrote: > "Jun 18 17:59:42 yossarian sid-filter[28339]: smfi_register() failed" > > when the sendmail script tried to invoke the filter ... That's pretty bizarre. The patch I gave you doesn't touch the registration code (in sid-filter.c), only something in the list lookup code (in util.c). smfi_register() fails if: - it couldn't allocate enough space for an internal data structure (unlikely, unless your system is already extremely taxed) - the registration structure passed in had no name defined (it does in this case; in fact it's hard-coded) - the version of milter requested by the filter and the version supported by libmilter were incompatible At a guess, I'd say it's a compilation problem. Did you try building from scratch ("sh Build -c")? |
From: The M. P. <sy...@an...> - 2008-06-19 22:00:41
|
... ciao: > On Thu, 19 Jun 2008, Murray S. Kucherawy wrote: > At a guess, I'd say it's a compilation problem. Did you try building from > scratch ("sh Build -c")? just tried that, same result. like you, i'm sure it local to me, on this system. i am going to build it on a 'more modern' system, 2.4.x and 3.2 compiler, and see what happens ... |
From: SM <sm...@re...> - 2008-06-20 15:01:57
|
At 15:00 19-06-2008, The Mighty Phlabaud wrote: > just tried that, same result. like you, i'm sure it local to me, on >this system. i am going to build it on a 'more modern' system, 2.4.x and >3.2 compiler, and see what happens ... Which version of sendmail are you using? Regards, -sm |
From: The M. P. <sy...@an...> - 2008-06-20 23:42:10
|
... ciao: > On Fri, 20 Jun 2008, SM wrote: > just tried that, same result. like you, i'm sure it local to me, on > >this system. i am going to build it on a 'more modern' system, 2.4.x and > >3.2 compiler, and see what happens ... > Which version of sendmail are you using? "Sendmail 8.14.0/8.14.0" ... |
From: SM <sm...@re...> - 2008-06-21 17:18:06
|
At 12:57 20-06-2008, The Mighty Phlabaud wrote: > "Sendmail 8.14.0/8.14.0" ... That version of sendmail uses a newer version of the milter library. Verify which version of the milter library is installed on the system and see whether you have compiled sid-milter against it. Regards, -sm |
From: The M. P. <sy...@an...> - 2008-07-02 13:47:38
|
... ciao: > On Fri, 20 Jun 2008, SM wrote: > Date: Fri, 20 Jun 2008 21:20:17 -0700 >> "Sendmail 8.14.0/8.14.0" ... > That version of sendmail uses a newer version of the milter > library. Verify which version of the milter library is installed on > the system and see whether you have compiled sid-milter against it. first, let me apolologize for the delay in getting back to you. i installed "Sendmail 8.14.3/8.14.3", and built 'sid-filter' against it. the 'PATCH' you supplied took care of the problem. thanks you ... |
From: Murray S. K. <ms...@se...> - 2008-06-14 23:14:55
|
On Fri, 13 Jun 2008, The Mighty Phlabaud wrote: > ... ciao: > > "Milter insert (1): header: Authentication-Results: aniota.com; > sender-id=none header.from=us...@co...; spf=none > smtp.mfrom=us...@co..." > > ".cox.net" listed in '-a /etc/mail/sid.accept'. "-a" ignores based on connecting IP address, not the envelope sender. > a question. if 'none' treated like a 'fail', is there any way other > than '-a', to deal with domains that don't publish spf records ... "none" isn't treated like "fail", unless the rule demands a "pass". > as an aside: > > you had asked on sourceforge about the os, well, it linux 2.2.26, > with gcc version egcs-2.91.66. i suspect, the "in_addr_t" problem , > mostly mine ... I'll work on a patch to send you which fixes this for older systems. You'll have to tweak Makefile.m4 to get it to work. |
From: The M. P. <sy...@an...> - 2008-06-16 11:18:03
|
... ciao: > On Sat, 14 Jun 2008, Murray S. Kucherawy wrote: > > ".cox.net" listed in '-a /etc/mail/sid.accept'. > "-a" ignores based on connecting IP address, not the envelope sender. "Search ARIN WHOIS for: 68.230.240.46 Cox Communications NETBLK-AT-INT-68-230-240-0 (NET-68-230-240-0-1) 68.230.240.0 - 68.230.241.255" > > linux 2.2.26, gcc version egcs-2.91.66 "in_addr_t" > I'll work on a patch to send you which fixes this for older systems. actually, i think it might be easier for us both if you don't bother. as it stands, it looks like i'm the only one with the problem, i know what it looks like, and you've given me the solution .. |
From: Murray S. K. <ms...@se...> - 2008-06-16 16:41:20
|
On Mon, 16 Jun 2008, The Mighty Phlabaud wrote: >>> ".cox.net" listed in '-a /etc/mail/sid.accept'. >> "-a" ignores based on connecting IP address, not the envelope sender. > > "Search ARIN WHOIS for: 68.230.240.46 > Cox Communications NETBLK-AT-INT-68-230-240-0 (NET-68-230-240-0-1) > 68.230.240.0 - 68.230.241.255" What you showed me was an envelope sender. Anyone could claim to be us...@co... from any IP address. |
From: The M. P. <sy...@an...> - 2008-06-16 19:31:08
|
... ciao: > On Mon, 16 Jun 2008, Murray S. Kucherawy wrote: > Date: Mon, 16 Jun 2008 09:41:15 -0700 (PDT) > > "Search ARIN WHOIS for: 68.230.240.46 > > Cox Communications NETBLK-AT-INT-68-230-240-0 (NET-68-230-240-0-1) > > 68.230.240.0 - 68.230.241.255" > > What you showed me was an envelope sender. Anyone could claim to be > us...@co... from any IP address. my mistake then; from the maillog (usernames changed) "Jun 13 06:25:28 yossarian sm-mta[573]: m5DDPRg0000573: from=<us...@co...>, size=6239, class=0, nrcpts=1, msgid=<015a01c8cd58$eeca4a70$6501a8c0@MOM>, bodytype=7BIT, proto=ESMTP, daemon=MTA, relay=eastrmmtao104.cox.net [68.230.240.46]" ... |
From: Murray S. K. <ms...@se...> - 2008-06-18 22:48:39
|
On Mon, 16 Jun 2008, The Mighty Phlabaud wrote: > my mistake then; from the maillog (usernames changed) > > "Jun 13 06:25:28 yossarian sm-mta[573]: m5DDPRg0000573: > from=<us...@co...>, size=6239, class=0, nrcpts=1, > msgid=<015a01c8cd58$eeca4a70$6501a8c0@MOM>, bodytype=7BIT, proto=ESMTP, > daemon=MTA, relay=eastrmmtao104.cox.net [68.230.240.46]" ... I've opened bug #1997251 on SourceForge, and attached a patch that appears to fix it. Let me know if it works for you. |
From: Jeff A. <an...@he...> - 2008-06-21 16:52:51
|
Patch worked fine for me. Thanks very much, Jeff Murray S. Kucherawy wrote: > On Mon, 16 Jun 2008, The Mighty Phlabaud wrote: >> my mistake then; from the maillog (usernames changed) >> >> "Jun 13 06:25:28 yossarian sm-mta[573]: m5DDPRg0000573: >> from=<us...@co...>, size=6239, class=0, nrcpts=1, >> msgid=<015a01c8cd58$eeca4a70$6501a8c0@MOM>, bodytype=7BIT, proto=ESMTP, >> daemon=MTA, relay=eastrmmtao104.cox.net [68.230.240.46]" ... > > I've opened bug #1997251 on SourceForge, and attached a patch that appears > to fix it. Let me know if it works for you. |