Re: [squid-users] Problems with ICAP and parent-child configuration

From: Amos Jeffries <squid3_at_treenet.co.nz>
Date: Sun, 27 Jul 2008 13:09:20 +1200

Christos Tsantilas wrote:
> Amos Jeffries wrote:
>> Christos Tsantilas wrote:
>>>>> I think you can safely ignore this error message.
>>>>>
>>>> Yes but maybe I might have some problems to convince my customer to
>>>> ignore
>>>> it! :)
>>>>
>>> This message can appeared when the squid 3.0 shutdown while the ICAP
>>> subsystem shutdown the ICAP services and in this case this message can
>>> ignored.
>>>
>>> If you see this message while squid3 is running or while is starting
>>> then
>>> it means that the ICAP service did not respond to the ICAP OPTIONS
>>> request, and squid3 will stop use the service and in this case it is
>>> problem.
>
> I make a mistake, sorry.
> This message appeared (1) when squid3.0 shutdown and (1) when an
> icap_service config line for a reason fails (parse error etc..), but in
> the second case there are other messages too which inform the user
> about the configuration error.
>
>>
>> Christos. You should have the global flag shutting_down to test and
>> silence this at the proper times.
>
> In any case this message has a debug level 2, I think it is OK to leave
> it as is.
> Do you think that it is a problem? It is not difficult to fix it, I can
> make a patch which separate the cases....
>

Okay. Should not be a problem at level-2 . I just got the impression it
was at level-1 or -0. What I was hinking with the shutdown'silencing was
to make it "debugs(N, (shuting_down?2:1), ...".

But that is not really needed if its already below level-1.

Daniele;
  Since you're doing this for a customer. You may find its better to run
the production squid debug logging at level-0 (CRITICAL) or level-1
(IMPORTANT) debugging. The other levels are generally for informative
purposes on tracking a known issue down, not finding new ones, and can
quickly fill the cache.log.

Amos

-- 
Please use Squid 2.7.STABLE3 or 3.0.STABLE8
Received on Sun Jul 27 2008 - 01:09:20 MDT

This archive was generated by hypermail 2.2.0 : Sun Jul 27 2008 - 12:00:04 MDT