Atle Samuelsen wrote:

Hi nick

Try to do the following :

ulimit -c unlimited

that will make it coredump instead. then, next time you start it, it should make a core file ser.core or something like that.

When that´s done, do :

gdb �-c ser.core(orwhatthecorefileiscalled)ser

this will open gdb, so you can run a backtrace.

type : bt

and you will get the backtrace. Past it here, and maybe somebody can see what´s wrong :-)

Best regards,

Atle

• nick nick@mobilia.it [060623 12:29]:

...

To explain my situation, I've got a dual opteron 246 system with 1gig of ram and 2 80gig disks in raid1 (software raid), I'm running CentOS 4.3 x86_64 with kernel 2.6.9-34.ELsmp I'm running ser 0.9.6-6 (the -6 may just be an artifact from the fedora core rpm.. I dunno) (compiled from ser-0.9.6_src.tar.gz)

Since no one seems to makes rpms for redhat or centOS (the same thing obviously) I went and made my own. I grabbed a fedora core 6 src rpm, and modified the specfile to my own liking (basically I took out the stuff related to postgres), the compile went fine, as far as I can tell, no errors were big enough to stop the compilation at least, and the rpms installed just fine. All the modules are in the right place, and my ser.cfg seems to be correct (I don't get any more errors during startup), however, I do get a segfault :(

here's the attached output from /log/messages

Jun 23 12:20:18 sipserver ser: Listening on Jun 23 12:20:18 sipserver ser: udp: 192.168.1.93 [192.168.1.93]:5060 Jun 23 12:20:18 sipserver ser: tcp: 192.168.1.93 [192.168.1.93]:5060 Jun 23 12:20:18 sipserver ser: Aliases: Jun 23 12:20:18 sipserver ser: tcp: pc-00093:5060 Jun 23 12:20:18 sipserver ser: udp: pc-00093:5060 Jun 23 12:20:18 sipserver ser: Jun 23 12:20:18 sipserver kernel: ser[3682]: segfault at 0000000000000000 rip 0000002a9589ef92 rsp 0000007fbffff850 error 4 Jun 23 12:20:18 sipserver ser: ser startup succeeded

Obviously ser isn't running, does anyone have any idea where I can start looking to find out what dies/died/went wrong with the compile?

Thanks,

Nick _______________________________________________ Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

OK, got the core dumps working did a gdb -c on the first coredump file

here's the output after typing bt:

[root@sipserver /]# gdb -c core.3658 GNU gdb Red Hat Linux (6.3.0.0-1.96rh) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "x86_64-redhat-linux-gnu". Core was generated by `/usr/sbin/ser'. Program terminated with signal 11, Segmentation fault. #0 0x0000002a9589ef92 in ?? () (gdb) bt #0 0x0000002a9589ef92 in ?? () #1 0x0000000000000010 in ?? () #2 0x0000002a95681686 in ?? () #3 0x0000002a95782c40 in ?? () #4 0x0000000000000af0 in ?? () #5 0x000000000000000c in ?? () #6 0x000000000000000c in ?? () #7 0x0000000000000000 in ?? ()

Nick, We're running ser-0.9.7-pre (at onsip.org) under CentOs and kernel 2.6.9-34.EL. No problems whatsoever. It's Intel, though. g-)

nick wrote:

...

...

...

To explain my situation, I've got a dual opteron 246 system with 1gig of ram and 2 80gig disks in raid1 (software raid), I'm running CentOS 4.3 x86_64 with kernel 2.6.9-34.ELsmp I'm running ser 0.9.6-6 (the -6 may just be an artifact from the fedora core rpm.. I dunno) (compiled from ser-0.9.6_src.tar.gz)

Since no one seems to makes rpms for redhat or centOS (the same thing obviously) I went and made my own. I grabbed a fedora core 6 src rpm, and modified the specfile to my own liking (basically I took out the stuff related to postgres), the compile went fine, as far as I can tell, no errors were big enough to stop the compilation at least, and the rpms installed just fine. All the modules are in the right place, and my ser.cfg seems to be correct (I don't get any more errors during startup), however, I do get a segfault :(

here's the attached output from /log/messages

Jun 23 12:20:18 sipserver ser: Listening on Jun 23 12:20:18 sipserver ser: udp: 192.168.1.93 [192.168.1.93]:5060 Jun 23 12:20:18 sipserver ser: tcp: 192.168.1.93 [192.168.1.93]:5060 Jun 23 12:20:18 sipserver ser: Aliases: Jun 23 12:20:18 sipserver ser: tcp: pc-00093:5060 Jun 23 12:20:18 sipserver ser: udp: pc-00093:5060 Jun 23 12:20:18 sipserver ser: Jun 23 12:20:18 sipserver kernel: ser[3682]: segfault at 0000000000000000 rip 0000002a9589ef92 rsp 0000007fbffff850 error 4 Jun 23 12:20:18 sipserver ser: ser startup succeeded

Obviously ser isn't running, does anyone have any idea where I can start looking to find out what dies/died/went wrong with the compile?

Thanks,

Nick _______________________________________________ Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

OK, got the core dumps working did a gdb -c on the first coredump file

here's the output after typing bt:

[root@sipserver /]# gdb -c core.3658 GNU gdb Red Hat Linux (6.3.0.0-1.96rh) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "x86_64-redhat-linux-gnu". Core was generated by `/usr/sbin/ser'. Program terminated with signal 11, Segmentation fault. #0 0x0000002a9589ef92 in ?? () (gdb) bt #0 0x0000002a9589ef92 in ?? () #1 0x0000000000000010 in ?? () #2 0x0000002a95681686 in ?? () #3 0x0000002a95782c40 in ?? () #4 0x0000000000000af0 in ?? () #5 0x000000000000000c in ?? () #6 0x000000000000000c in ?? () #7 0x0000000000000000 in ?? ()

Ok, to see if it might be a problem with the kernel, I just grabbed the latest from the servers, it's now 2.6.9-34.0.1.ELsmp, but it segfaults anyways, although the backtrace is a bit different:

[root@sipserver /]# gdb -c core.3095 Core was generated by `/usr/sbin/ser'. Program terminated with signal 11, Segmentation fault. #0 0x0000002a95785f92 in ?? () (gdb) bt #0 0x0000002a95785f92 in ?? () #1 0x0000000000000400 in ?? () #2 0x0000002a95568686 in ?? () #3 0x0000002a95669c40 in ?? () #4 0x0000000000000000 in ?? ()

---

Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

Hello Nick,

To me it seems that gdb does not find the debugging symbols, needed to output an easy to read call stack. Was ser compiled with debugging enabled (make mode=debug ...)?

WL.

nick wrote:

Greger V. Teigre wrote:

...

Nick, We're running ser-0.9.7-pre (at onsip.org) under CentOs and kernel 2.6.9-34.EL. No problems whatsoever. It's Intel, though. g-)

nick wrote:

...

...

...

...

To explain my situation, I've got a dual opteron 246 system with 1gig of ram and 2 80gig disks in raid1 (software raid), I'm running CentOS 4.3 x86_64 with kernel 2.6.9-34.ELsmp I'm running ser 0.9.6-6 (the -6 may just be an artifact from the fedora core rpm.. I dunno) (compiled from ser-0.9.6_src.tar.gz)

Since no one seems to makes rpms for redhat or centOS (the same thing obviously) I went and made my own. I grabbed a fedora core 6 src rpm, and modified the specfile to my own liking (basically I took out the stuff related to postgres), the compile went fine, as far as I can tell, no errors were big enough to stop the compilation at least, and the rpms installed just fine. All the modules are in the right place, and my ser.cfg seems to be correct (I don't get any more errors during startup), however, I do get a segfault :(

here's the attached output from /log/messages

Jun 23 12:20:18 sipserver ser: Listening on Jun 23 12:20:18 sipserver ser: udp: 192.168.1.93 [192.168.1.93]:5060 Jun 23 12:20:18 sipserver ser: tcp: 192.168.1.93 [192.168.1.93]:5060 Jun 23 12:20:18 sipserver ser: Aliases: Jun 23 12:20:18 sipserver ser: tcp: pc-00093:5060 Jun 23 12:20:18 sipserver ser: udp: pc-00093:5060 Jun 23 12:20:18 sipserver ser: Jun 23 12:20:18 sipserver kernel: ser[3682]: segfault at 0000000000000000 rip 0000002a9589ef92 rsp 0000007fbffff850 error 4 Jun 23 12:20:18 sipserver ser: ser startup succeeded

Obviously ser isn't running, does anyone have any idea where I can start looking to find out what dies/died/went wrong with the compile?

Thanks,

Nick _______________________________________________ Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

OK, got the core dumps working did a gdb -c on the first coredump file

here's the output after typing bt:

[root@sipserver /]# gdb -c core.3658 GNU gdb Red Hat Linux (6.3.0.0-1.96rh) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "x86_64-redhat-linux-gnu". Core was generated by `/usr/sbin/ser'. Program terminated with signal 11, Segmentation fault. #0 0x0000002a9589ef92 in ?? () (gdb) bt #0 0x0000002a9589ef92 in ?? () #1 0x0000000000000010 in ?? () #2 0x0000002a95681686 in ?? () #3 0x0000002a95782c40 in ?? () #4 0x0000000000000af0 in ?? () #5 0x000000000000000c in ?? () #6 0x000000000000000c in ?? () #7 0x0000000000000000 in ?? ()

Ok, to see if it might be a problem with the kernel, I just grabbed the latest from the servers, it's now 2.6.9-34.0.1.ELsmp, but it segfaults anyways, although the backtrace is a bit different:

[root@sipserver /]# gdb -c core.3095 Core was generated by `/usr/sbin/ser'. Program terminated with signal 11, Segmentation fault. #0 0x0000002a95785f92 in ?? () (gdb) bt #0 0x0000002a95785f92 in ?? () #1 0x0000000000000400 in ?? () #2 0x0000002a95568686 in ?? () #3 0x0000002a95669c40 in ?? () #4 0x0000000000000000 in ?? ()

---

Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

So this is all the help I can expect??

Sigh, off to build myself an earlier 9, and possibly an 8 series rpm :(

---

Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

nick wrote:

Weiter Leiter wrote:

...

Hello Nick,

To me it seems that gdb does not find the debugging symbols, needed to output an easy to read call stack. Was ser compiled with debugging enabled (make mode=debug ...)?

WL.

nick wrote:

...

Greger V. Teigre wrote:

...

Nick, We're running ser-0.9.7-pre (at onsip.org) under CentOs and kernel 2.6.9-34.EL. No problems whatsoever. It's Intel, though. g-)

nick wrote:

...

...

>> To explain my situation, I've got a dual opteron 246 system with >> 1gig of ram and 2 80gig disks in raid1 (software raid), I'm >> running CentOS 4.3 x86_64 with kernel 2.6.9-34.ELsmp I'm running >> ser 0.9.6-6 (the -6 may just be an artifact from the fedora core >> rpm.. I dunno) (compiled from ser-0.9.6_src.tar.gz) >> >> >> Since no one seems to makes rpms for redhat or centOS (the same >> thing obviously) I went and made my own. I grabbed a fedora core >> 6 src rpm, and modified the specfile to my own liking (basically >> I took out the stuff related to postgres), the compile went fine, >> as far as I can tell, no errors were big enough to stop the >> compilation at least, and the rpms installed just fine. All the >> modules are in the right place, and my ser.cfg seems to be >> correct (I don't get any more errors during startup), however, I >> do get a segfault :( >> >> here's the attached output from /log/messages >> >> Jun 23 12:20:18 sipserver ser: Listening on >> Jun 23 12:20:18 sipserver ser: udp: 192.168.1.93 >> [192.168.1.93]:5060 >> Jun 23 12:20:18 sipserver ser: tcp: 192.168.1.93 >> [192.168.1.93]:5060 >> Jun 23 12:20:18 sipserver ser: Aliases: >> Jun 23 12:20:18 sipserver ser: tcp: pc-00093:5060 >> Jun 23 12:20:18 sipserver ser: udp: pc-00093:5060 >> Jun 23 12:20:18 sipserver ser: >> Jun 23 12:20:18 sipserver kernel: ser[3682]: segfault at >> 0000000000000000 rip 0000002a9589ef92 rsp 0000007fbffff850 error 4 >> Jun 23 12:20:18 sipserver ser: ser startup succeeded >> >> Obviously ser isn't running, does anyone have any idea where I >> can start looking to find out what dies/died/went wrong with the >> compile? >> >> Thanks, >> >> Nick >> _______________________________________________ >> Serusers mailing list >> Serusers@lists.iptel.org >> http://lists.iptel.org/mailman/listinfo/serusers >> > > OK, got the core dumps working did a gdb -c on the first coredump file

here's the output after typing bt:

[root@sipserver /]# gdb -c core.3658 GNU gdb Red Hat Linux (6.3.0.0-1.96rh) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "x86_64-redhat-linux-gnu". Core was generated by `/usr/sbin/ser'. Program terminated with signal 11, Segmentation fault. #0 0x0000002a9589ef92 in ?? () (gdb) bt #0 0x0000002a9589ef92 in ?? () #1 0x0000000000000010 in ?? () #2 0x0000002a95681686 in ?? () #3 0x0000002a95782c40 in ?? () #4 0x0000000000000af0 in ?? () #5 0x000000000000000c in ?? () #6 0x000000000000000c in ?? () #7 0x0000000000000000 in ?? ()

Ok, to see if it might be a problem with the kernel, I just grabbed the latest from the servers, it's now 2.6.9-34.0.1.ELsmp, but it segfaults anyways, although the backtrace is a bit different:

[root@sipserver /]# gdb -c core.3095 Core was generated by `/usr/sbin/ser'. Program terminated with signal 11, Segmentation fault. #0 0x0000002a95785f92 in ?? () (gdb) bt #0 0x0000002a95785f92 in ?? () #1 0x0000000000000400 in ?? () #2 0x0000002a95568686 in ?? () #3 0x0000002a95669c40 in ?? () #4 0x0000000000000000 in ?? ()

---

Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

So this is all the help I can expect??

Sigh, off to build myself an earlier 9, and possibly an 8 series rpm :(

---

Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

I used both the official compiled 0.9.6 x86_64 binary downloaded from iptel.org, as well as my own build (which happened to make an rpm called ser-debuginfo) ... I installed all of the packages overwriting everything in the /sbin directory with the files from the official compile of 0.9.6 (I did the same with all the modules also), I can add the "make mode=debug" into the spec file for recompilation, if that will help... should I add it at the end of the:

%{__make} all skip_modules="%{special_modules}" cfg-target=%{_sysconfdir}/ser/ modules-dir=%{_lib}/ser/modules

or put it before skip_modules?

I'd like to get this install up and running, and I'm willing to do what's necessary to get rid of this problem..

Nick _______________________________________________ Serusers mailing list Serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers

I've done the make mode=debug all ..

but I still have nothing back from gdb :(

what a life..

I'm going to try and compile it by hand now..