Comment # 9 on bug 1096401 from Stefan Schubert

(In reply to Dominique Leuenberger from comment #0)
> While checking some logs on openQA, trying to find why so many tests are
> failing, I stumbled over a segfault of snapper:
> 
> Jun 07 02:18:14 susetest org.opensuse.Snapper[719]: terminate called after
> throwing an instance of
> 'boost::exception_detail::clone_impl<boost::exception_detail::
> error_info_injector<boost::lock_error> >'
> Jun 07 02:18:14 susetest org.opensuse.Snapper[719]:   what():  boost: mutex
> lock failed in pthread_mutex_lock: Invalid argument
> Jun 07 02:18:14 susetest systemd[1]: Started Process Core Dump (PID
> 20684/UID 0).
> Jun 07 02:18:15 susetest systemd-coredump[20685]: Process 18707 (snapperd)
> of user 0 dumped core.
> 
> TIME                            PID   UID   GID SIG COREFILE  EXE
> Thu 2018-06-07 02:09:35 EDT    2722     0     0   6 present   /bin/login
> Thu 2018-06-07 02:14:34 EDT   14648     0     0   6 present  
> /usr/sbin/snapperd
> Thu 2018-06-07 02:15:58 EDT   16462     0     0   6 present  
> /usr/sbin/snapperd
> Thu 2018-06-07 02:18:15 EDT   18707     0     0   6 present  
> /usr/sbin/snapperd
> 
> This can explain why openQA has a hard time passing the tests, as the system
> is pre-occupied writing coredumps
> 
> The three snaperd coredumps are available in the openQA provided tarball
>  https://openqa.opensuse.org/tests/688332/file/gimp-problem_detection_logs.
> tar.xz
Hm, this file does not exist anymore. Could you please point us to it again ?
Additional it would be nice if you point us to the point where it happens in
this test. (The test is quite huge :-))
> 
> (directory coredumps)