This is on our test cluster so while it is a high serverity bug in that it causes a crash, it is not on our production servers so its not currently causing us problems. When restarting slurmctld in 16.05.2 the ctld crashes on start. This may be due to a specific job that is faulty, but slurm shouldn't crash on that. I've included the log and I have a core dump. The ctld is running CentOS 7. Let me know what other info you need from the core dump. Thanks. -Paul Edmon-
Created attachment 3292 [details] crash log
Can you get full backtrace from all thread? thread apply all bt full Dominik
Created attachment 3293 [details] slurmctld backtrace
This bug was fixed in commit 65b4f283ef2 (https://github.com/SchedMD/slurm/commit/65b4f283ef2a908b6e3e8921acf62dad73528f00.patch). This patch will be included also in 16.05.3 release. Dominik
Thanks! -Paul Edmon- On 07/11/2016 09:54 AM, bugs@schedmd.com wrote: > > *Comment # 6 <https://bugs.schedmd.com/show_bug.cgi?id=2885#c6> on bug > 2885 <https://bugs.schedmd.com/show_bug.cgi?id=2885> from Dominik > Bartkiewicz <mailto:bart@schedmd.com> * > This bug was fixed in commit 65b4f283ef2 > (https://github.com/SchedMD/slurm/commit/65b4f283ef2a908b6e3e8921acf62dad73528f00.patch). > This patch will be included also in 16.05.3 release. > > Dominik > ------------------------------------------------------------------------ > You are receiving this mail because: > > * You reported the bug. >
Please reopen if the problem still exists after this commit. Dominik