Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
David, I have a core file for you. I'll send it to you off-list. Bart ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
David F. Skoll wrote: Well, if you can *get* a core file, I can help you out off-list. Thank you very much for you offer, David. I really appreciate it! I will try my best to figure out how to make the program generate a core file. This is unknown territory, so I will have to dig around in Google or maybe ask for help about this in a FreeBSD forum/list/group. In the mean while, I have made a script that restarts Mimedefang when either the multiplexor or slaves die. It is not a neat solution, but it will keep the mail flowing as I try to sort things out. Bart ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
On Tuesday 02 May 2006 15:51, Bjarte Husebø wrote: David F. Skoll wrote: Doh... that's a segfault. You'll need to convince your system to generate a core file to have any hope of diagnosing this. Thanks for your fast response. Unfortunately, analyzing core dumps is probably far beyond my expertise. :-\ Cormack, Ken wrote: If this is a linux machine, I know of two circumstances under which you'll see this... 1.) Bad memory (try swapping out the memory modules) 2.) If you've over-clocked the system, you're pushing it too fast. I'm running FreeBSD 6.0 on a Pentium III that is about 3-4 years old. No piece is overclocked. Bad memory might be the reason for my problems - although I did run Memtest86+ on it before I installed FreeBSD. I tried to keep quiet ... Just couldn't do it. Signal 11 (Segfault) is nearly always a Program bug (Accessing memory that is not allocated to the current process, frequently address zero, usually related to a pointer being used that has not been set correctly). It is, nearly, impossible that it points to a hardware (memory, overclocking, etc.) problem. I'd recommend trying to produce a core file and accept Davids' help. That will, very likely, lead to a real diagnosis, and correction, of the problem. It might be worth the while to run another test. Thanks, Bart ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang -- Larry G. Starr - [EMAIL PROTECTED] or [EMAIL PROTECTED] Software Engineer: Full Compass Systems LTD. Phone: 608-831-7330 x 1347 FAX: 608-831-6330 === There are only three sports: bullfighting, mountaineering and motor racing, all the rest are merely games! - Ernest Hemmingway ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
Larry Starr wrote: Signal 11 (Segfault) is nearly always a Program bug (Accessing memory that is not allocated to the current process, frequently address zero, usually related to a pointer being used that has not been set correctly). It is, nearly, impossible that it points to a hardware (memory, overclocking, etc.) problem. While the most likely cause of a segfault is a software bug, hardware problems can and do cause segfaults. See for example http://www.bitwizard.nl/sig11/ If you're talking about MIMEDefang, my money's on a hardware problem or a bug in a system library, of course! :-) (But let's get the core file just in case...) Regards, David. ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
Hi, To get a mimedefang corefile you may have to execute: # sysctl kern.sugid_coredump=1 Martin ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
Martin Blapp wrote: To get a mimedefang corefile you may have to execute: # sysctl kern.sugid_coredump=1 Thank you, Martin. That seems to be exactly what I need. :-) Waiting for the next segfault, Bart ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
[Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
I have been using MIMEDefang successfully for several years now. After I installed it on a new server (FreeBSD 6.0), a rather nasty problem has come to my attention... After copying sendmail and mimedefang configurations files from the old server along with user accounts, mail and so forth, everything looked great. Maybe for a couple of hours or so. Then I spotted the following error in /var/log/messages: May 2 18:06:02 amanda kernel: pid 19878 (mimedefang-multiple), uid 26: exited on signal 11 And here is an excerpt from /var/log/maillog: May 2 18:06:13 amanda sm-mta[23003]: NOQUEUE: connect from cm56-138-42.liwest.at [86.56.138.42] May 2 18:06:13 amanda sm-mta[23003]: AUTH: available mech=NTLM LOGIN ANONYMOUS PLAIN GSSAPI OTP DIGEST-MD5 CRAM-MD5, allowed mech=GSSAPI DIGEST-MD5 CRAM-MD5 LOGIN May 2 18:06:13 amanda sm-mta[23003]: k42G6D3a023003: Milter (mimedefang): init success to negotiate May 2 18:06:13 amanda sm-mta[23003]: k42G6D3a023003: Milter (greylist): init success to negotiate May 2 18:06:13 amanda mimedefang[19893]: MXCommand: connect: Connection refused: Is multiplexor running? May 2 18:06:13 amanda sm-mta[23003]: k42G6D3a023003: Milter: connect to filters May 2 18:06:13 amanda mimedefang[19893]: mfconnect: Error communicating with multiplexor May 2 18:06:13 amanda sm-mta[23003]: k42G6D3a023003: milter=mimedefang, action=connect, tempfail May 2 18:06:13 amanda sm-mta[23003]: k42G6D3a023003: Milter: connect, ending May 2 18:06:13 amanda sm-mta[23003]: k42G6D3a023003: Milter: connect: host=cm56-138-42.liwest.at, addr=86.56.138.42, temp failing commands ..and so it kept temp-failing for about an hour, before I issued the following: # /usr/local/etc/rc.d/mimedefang.sh restart ..and it came back running just smoothly. But only for a random time, and it all happened again. Sendmail (8.13.6), mimedefang (2.56) and all required perl modules are installed from the FreeBSD ports collection. Everything is up to date, as far as I can tell. I also use milter-greylist (2.0.2) for greylisting, spamassassin (3.1.1) and f-prot (4.6.6). I have Googled for help on this, but have come up with nothing so far. Any advice on how to solve this problem will be greatly appreciated. Thank you in advance! Bart # uname -a FreeBSD amanda 6.0-RELEASE FreeBSD 6.0-RELEASE #0: Thu Nov 3 09:36:13 UTC 2005 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC i386 ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
RE: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
May 2 18:06:02 amanda kernel: pid 19878 (mimedefang-multiple), uid 26: exited on signal 11 Doh... that's a segfault. You'll need to convince your system to generate a core file to have any hope of diagnosing this. If this is a linux machine, I know of two circumstances under which you'll see this... 1.) Bad memory (try swapping out the memory modules) 2.) If you've over-clocked the system, you're pushing it too fast. Good luck. ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
David F. Skoll wrote: Doh... that's a segfault. You'll need to convince your system to generate a core file to have any hope of diagnosing this. Thanks for your fast response. Unfortunately, analyzing core dumps is probably far beyond my expertise. :-\ Cormack, Ken wrote: If this is a linux machine, I know of two circumstances under which you'll see this... 1.) Bad memory (try swapping out the memory modules) 2.) If you've over-clocked the system, you're pushing it too fast. I'm running FreeBSD 6.0 on a Pentium III that is about 3-4 years old. No piece is overclocked. Bad memory might be the reason for my problems - although I did run Memtest86+ on it before I installed FreeBSD. It might be worth the while to run another test. Thanks, Bart ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
Bjarte Husebø wrote: Bad memory might be the reason for my problems - although I did run Memtest86+ on it before I installed FreeBSD. It might be worth the while to run another test. If you can afford the time (i.e. if you've got another machine that can handle its function for a while), let it run multiple passes. Sometimes if the hardware is borderline, it can take a while to trigger a problem. When I've got a system acting suspiciously, I'll just let Memtest86 run overnight. Or occasionally even over the weekend. -- Kelson Vibber SpeedGate Communications www.speed.net ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Re: [Mimedefang] Multiplexor dies, leaving its slaves clueless behind...
Bjarte Husebø wrote: Thanks for your fast response. Unfortunately, analyzing core dumps is probably far beyond my expertise. :-\ Well, if you can *get* a core file, I can help you out off-list. Regards, David. ___ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list MIMEDefang@lists.roaringpenguin.com http://lists.roaringpenguin.com/mailman/listinfo/mimedefang