Hey list,
There's an utility shipped with systemd, /usr/lib/systemd/systemd-coredump, which, being set up as kernel.core_pattern handler, writes core dumps to journal. A companion utility, systemd-coredumpctl, can list core dumps, export them and run gdb against them. I'm aware openSUSE disables this feature, but I'm consdering it very useful for my purposes, so I was actually able to make use of it in openSUSE 13.1 by installing systemd-logger and creating /etc/sysctl.d/50-coredump.conf pretty identical to bundled one — http://cgit.freedesktop.org/systemd/systemd/tree/sysctl.d/50-coredump.conf.i....
But trying to do the same on Factory I've faced an issue — rsyslog is gone, core_pattern is set, but systemd-coredumpctl shows nothing, despite segfaults happen pretty often.
So I've got following questions regarding this situation:
* What is blessed way to automatically collect coredumps in openSUSE? I've noticed /var/log/core directory, but it's empty and nobody seem to touch it. * If systemd-coredump is okay, what else would you suggest to try to track down root cause of an issue?
P.S. Some snippets from shell:
$ /usr/sbin/sysctl kernel.core_pattern kernel.core_pattern = /usr/lib/systemd/systemd-coredump %p %u %g %s %t %e
$ cat /tmp/sigsegv.c int main(void) { char *s = "hello world"; *s = 'H'; }
$ /tmp/a.out fish: Job 1, “/tmp/a.out ” terminated by signal SIGSEGV (Address boundary error) [trust me, a.out is compiled from sigsegv.c]
$ journalctl -e …skip… Aug 28 00:00:46 linux-z0lo.site kernel: a.out[432]: segfault at 4005a4 ip 000000000040050d sp 00007fffe4fe8120 error 7 in a.out[400000+1000]
$ systemd-coredumpctl list | wc -l 0
Thank you in advance.