Hello, pon., 22 cze 2020 o 16:23 Phil Stracchino <ph...@caerllewys.net> napisał(a):
> On 2020-06-22 03:07, Radosław Korzeniewski wrote: > > Well, I have a different point of view. In this case SIGABRT is > > intentionally raised by the library telling us - I have no idea what is > > going wrong and I'm unable to recover from the situation and it is soooo > > bad that I have to abort! > > It is not a situation where a client executed the library function with > > invalid parameters (invalid values, syntax errors or null pointers). In > > that case the library should return with error code or SIGSEGV. > > > > Something goes wrong on the database side or it is a bug in library code. > > > Radoslaw, for clarification, this comment is concerning the Director > crashing when it fails to cancel a hung job, correct? > I'm referring to this crash: Thread 5 (Thread 0x7f74c67fc700 (LWP 23810)): #0 0x00007f74d89439ae in waitpid () from /lib64/libpthread.so.0 #1 0x00007f74d899a74c in signal_handler (sig=6) at signal.c:233 #2 <signal handler called> #3 0x00007f74d8537621 in raise () from /lib64/libc.so.6 #4 0x00007f74d852155b in abort () from /lib64/libc.so.6 #5 0x00007f74d7e88c01 in ?? () from /usr/lib64/libmysqlclient.so.21 #6 0x00007f74d7e8b5c7 in ?? () from /usr/lib64/libmysqlclient.so.21 #7 0x00007f74d7e398d0 in ?? () from /usr/lib64/libmysqlclient.so.21 #8 0x00007f74d7e39e1d in ?? () from /usr/lib64/libmysqlclient.so.21 #9 0x00007f74d7e35146 in mysql_send_query () from /usr/lib64/libmysqlclient.so.21 #10 0x00007f74d7e35385 in mysql_real_query () from /usr/lib64/libmysqlclient.so.21 #11 0x00007f74d89e0138 in BDB_MYSQL::sql_query (this=0x558706673108, query=0x7f74c801b320 "UPDATE Job SET JobStatus='f',EndTime='2020-06-17 08:24:13',ClientId=31,JobBytes=15644859,ReadBytes=0,JobFiles=19,JobErrors=1,VolSessionId=33,VolSessionTime=1592158122,PoolId=5,FileSetId=31,JobTDate=15"..., flags=0) at mysql.c:537 The crash is caused by sig abort raised from /usr/lib64/libmysqlclient.so.21. I do not refer to any other cause of the crash or working condition. best regards -- Radosław Korzeniewski rados...@korzeniewski.net
_______________________________________________ Bacula-devel mailing list Bacula-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-devel
