Robin Stevens wrote:
> > As this is a fatal error, you should start by reading "Squid FAQ 11.19
> > Sending in Squid bug reports"
> > <http://www.squid-cache.org/Doc/FAQ/FAQ-11.html#ss11.19>.
>
> Which I looked at and saw that core dumps and async-io don't mix, and I
> couldn't see that we could run squid through ghdb for days at a time
> waiting for this to happen again. I was hoping someone could suggest a way
> I could get more information next time in order to submit a bug
> report with some more helpful information.
Did you also see what is said last in that FAQ entry? (the
--enable-stacktraces or catchsegv alternatives)
> However I see I somehow missed bugid 94 when I looked at the bugzilla
> database - my apologies...
You should get a stack trace to verify that your problem is the same
before assuming it is.
For this, the --enable-stacktraces option is more than sufficient. If
the two calls leading up to xassert (below it in the listing) is the
same functions then the problem is with 99.9% the exact same problem.
Furthermore, there is no penalty of --enable-stacktraces until Squid
crashes.
If they are different, the problem may still quite likely be the same,
and we have some additional data helping us to pinpoint where the
problem is.
Regards
Henrik Nordström
Received on Sun Oct 28 2001 - 06:16:33 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:03:10 MST