Folks,
We're using an old (1.4.1p1) version of Poco, and would like to make minimal changes to fix the following bug,
assuming it is indeed a problem with Poco. We're asking for help to nail down exactly
what the problem is and exactly what we need to change to fix it.
The bug is intermittent, and so it's hard for us to debug or verify a fix.
Any help is appreciated. We're not that familiar with Poco internals, certainly,
and inherited most of our Poco code. As I've said we're not sure we want to upgrade
to a newer version -- we'd rather fix this specific problem, since the rest of the system
works ok.
Thanks in advance -- Steve
Here's what's on the stack when we experience this, as a segfault (on a unix system):
[Fri Aug 03 21:41:31 2012][Fatal][] Processing aborted via SEGV.
(Error:800002)
[Fri Aug 03 21:41:31 2012][Trace][FaultHandler] 1 /lib64/libc.so.6()
[0x3b07277cf9]
[Fri Aug 03 21:41:31 2012][Trace][FaultHandler] 2 /lib64/libpthread.so.0()
[0x3b0760f4c0]
[Fri Aug 03 21:41:31 2012][Trace][FaultHandler] 3 /lib64/libc.so.6()
[0x3b07277cf9]
[Fri Aug 03 21:41:31 2012][Trace][FaultHandler] 4
/home/shared/software/454/2.8/apps/gsRunProcessor/bin/gsRunProcessor,Poco::NotificationQueue::waitDequeueNotification(long)
[Fri Aug 03 21:41:31 2012][Trace][FaultHandler] 5
/home/shared/software/454/2.8/apps/gsRunProcessor/bin/gsRunProcessor,Poco::Net::SyslogParser::run()
[Fri Aug 03 21:41:31 2012][Trace][FaultHandler] 6
/home/shared/software/454/2.8/apps/gsRunProcessor/bin/gsRunProcessor,Poco::ThreadImpl::runnableEntry(void*)
[Fri Aug 03 21:41:31 2012][Trace][FaultHandler] 7 /lib64/libpthread.so.0()
[0x3b076077e1]
[Fri Aug 03 21:41:31 2012][Trace][FaultHandler] 8 /lib64/libc.so.6,clone()
O
