The only debug symbols I can find on the repos is libicapapi0-dbg... Do You think it could be useful running c-icap under gdb using that?
Regards Luca Il giorno lun, 01/12/2014 alle 06.17 +0100, Mathieu Parent ha scritto: > Control: tag -1 + moreinfo unreproducible > > Hello luskin, > > 2014-11-30 2:09 GMT+01:00 luskin <lucalanar...@gmail.com>: > [...] > > > After last system update (2 days ago) I've this problem after each daily > > logrotate: > > > > Nov 28 07:36:14 andromeda kernel: [375111.408129] c-icap[5693]: > > segfault at 20 ip 00007f936dda29ca sp 00007fff305d33e0 error 4 in > > libLLVM-3.0.so.1[7f936d3d0000+1301000] > > Nov 29 07:36:17 andromeda kernel: [461352.882110] c-icap[32150]: > > segfault at 20 ip 00007fca64edc9ca sp 00007fff56132ca0 error 4 in > > libLLVM-3.0.so.1[7fca6450a000+1301000] > > Nov 30 00:49:34 andromeda kernel: [523233.843037] c-icap[30766]: > > segfault at 20 ip 00007f4504aa09ca sp 00007fff430ad210 error 4 in > > libLLVM-3.0.so.1[7f45040ce000+1301000] > > Nov 30 01:24:32 andromeda kernel: [525327.612150] c-icap[18178]: > > segfault at 7f708766001c ip 0000000000413bfd sp 00007fffdc654830 error 4 in > > c-icap[400000+1e000] > > > > It seems to be the "reconfigure" message to c-icap.ctl to cause the issue > > (I've invoked it manually and got the seg fault to syslog). > > Could not reproduce here. > > > Any help? > > Can you reproduce this bug with c-icap running under gdb, and attach > the backtrace here? > > Howto is here: https://wiki.debian.org/HowToGetABacktrace > > > For now I've bash-scripted a cron watchdog that checks c-icap running at > > every minute and, if needed, restarts it. > > Anyway any hint is welcome. > > OK > > Regards > > Mathieu Parent
