On 2/18/19 1:26 PM, Martin Simmons wrote: >>>>>> On Mon, 18 Feb 2019 12:57:55 -0500, Phil Stracchino said: >> >> On 2/18/19 12:17 PM, Martin Simmons wrote: >>>>>>>> On Mon, 18 Feb 2019 08:19:34 -0500, Phil Stracchino said: >>>> >>>> Twice now in the four days since updating to 9.4.2, my Director has >>>> silently died without logging anything. The last thing it logged was a >>>> dbcheck run that completed at 09:37 yesterday morning. I don't know >>>> (yet) precisely when it died. >>>> >>>> Debugging suggestions...? >>> >>> I suggest leaving the Director running under gdb (or attaching gdb) to see >>> if >>> that catches any problems before it exits. >> >> Hm. I'd have to build a non-stripped binary. For the moment, I was >> more thinking along the lines of is there any particular debugging level >> likely to be useful here? > > If it is dying with a SIGSEGV (or some such signal) then debugging levels are > unlikely to help because they don't narrow it down sufficiently. > > Running with gdb on the stripped binary will at least give some info.
I tried strace, but it seems not to follow bacula-dir through daemonization. Possibly my invocation is wrong. I did also just notice in Bacula's jobs-run list that it thinks dbcheck terminated with a fatal error. This is not apparent from bacula-dir.log. I'm running a manual dbcheck now to see whether it sheds any light. -- Phil Stracchino Babylon Communications ph...@caerllewys.net p...@co.ordinate.org Landline: +1.603.293.8485 Mobile: +1.603.998.6958 _______________________________________________ Bacula-devel mailing list Bacula-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-devel
