Re: [squid-users] 2.4.20-aa and LARGE Squid process -> SIGSEGV

On Sun, 2002-12-22 at 01:03, Ralf Hildebrandt wrote:
> * Robert Collins <robertc@squid-cache.org>:
>
> >
> > Which squid children?
>
> Dunno. I'm just telling you what squid writes to /var/log/messages
> (see below) - it clearly talks of "Squid Parent: child process"!

Ok, thats the single child, crashing sequentially.

> > Squid should trap the crash and log it to cache.log.
>
> It doesn't. It's simply logging the restart, but not the crash!

Ok, so follow the FAQ and get us a stacktrace.
 
> > As to your problem, may I suggest the following:
> > 1) Use diskd. A nasty aufs bug was found recently that could impact you.
> > 2) Check cache.log as your primary source for squid diagnostics.
>
> See attached log - not entirely illuminating (for the unitiated?)

Nothing useful there, but we did need to check :}.

> > 3) Check squid memory use via cache manager for statistics.
>
> We're using http://wessels.squid-cache.org/squid-rrd/ to monitor &
> graph all kinds of data. I'll attach some pics

Uhm, you may find the general runtime statistics page of some use. It's
nearly all there via snmp as well for rrd logging :].

> > 4) Use squid 2.5-stable-cvs if you really want aufs support, the bug is
> > fixed there.
>
> The BUG? Which BUG? :) Note that our problem occurs with both ufs and
> aufs.

The 'recently found aufs bug that could impact you'.

Rob