Re: [squid-users] Re: squid 3.2.0.5 even slower than squid 3.1 and doesn't work with workers >1

From: Amos Jeffries <squid3_at_treenet.co.nz>
Date: Tue, 22 Mar 2011 17:10:58 +1300

On 22/03/11 14:49, david_at_lang.hm wrote:
> On Mon, 21 Mar 2011, david_at_lang.hm wrote:
>
>> one thing that I've found is that even with --disable-ipv6 squid will
>> still use IPv6 on a system that has it configured (next I'll try and
>> see if that's what's going wrong on the systems that don't have it
>> configured, but those systems don't have strace on them, so I'll have
>> to build a throw-away system instead of using one of my standard build
>> test systems)
>
> if the kernel doesn't support IPv6 squid doesn't try to use it, but it
> does try to use it even if configured with --disable-ipv6 if the kernel
> does support it.
>
> the problem turns out to be that with more than one worker, the workers
> try to write to /var/run when they start (to write their pid file) and
> if that fails they won't start.
>
> on my debian systems, /var/run is 755 root root and that seems like a
> very sane setting to use, what is it set for on your systems where you
> have been testing this feature?

Yes. /var/run owned by root. Squid files should end up in
/var/run/squid/ which has 755 to whatever low-privilege user/group Squid
operates as (usually 'nobody:nobody' or 'proxy:proxy').

This can vary if you have overridden the FHS standard location using
pid_filename directive. Only the master process or coordinator process
writes the .pid file.

Amos

-- 
Please be using
   Current Stable Squid 2.7.STABLE9 or 3.1.11
   Beta testers wanted for 3.2.0.5
Received on Tue Mar 22 2011 - 04:11:21 MDT

This archive was generated by hypermail 2.2.0 : Tue Mar 22 2011 - 12:00:02 MDT