Re[2]: [squid-users] SQUID2.5.D/lx2.4.4 - Peer failover - how to config squid to fail over to direct queries if upstream is offline? (NON-ICP PEER)

Thanks for your help Robert, and the quick reply.

==
A few further questions for the list, anyone
who can reel off some info on the below queries:-

Are these all the timeouts relevant to a non-ICP upstream peer?
#squid.conf
connect_timeout (for direct http gets?)
peer_connect_timeout (when this times out, does squid go direct?)
read_timeout

From the docos it's not clear which parameters apply to which
situation; is connect_timeout a tcp timeout relevant to
 squid->webserver connections or user->squid connections?
Should I check this with the dev's or duane via the faq?

One piece of important info I couldn't discern;
How would one set the post-parent-failure retry interval?
==
thx

ben

> Thats the default behaviour if you simply configure a parent cache. When
> the parent fails, squid goes direct.
> Rob
>
>> Subject: [squid-users] SQUID2.5.D/lx2.4.4 - Peer failover - how to
>> config squid to fail over to direct queries if upstream is offline?
>> (NON-ICP PEER)
>>
>>
>> Hi all, hope someone can help me out on this one...
>>
>> Have a slackware(lx2.2.4) box with squid 2.5.D installed to default
>> path.
>>
>> I have to insert the squid box to replace the MS ISA
>> server (it needs some kinetic maintenance).
>> This squid box will take over the primary handling of web caching for
>> the users. It will then chain through to the MS ISA machine.
>>
>> What I'm trying to do is to use the squid box to put a layer of fault
>> tolerance between the unstable ISA server and the users - the ISA box
>> falls over regularly - to divert traffic through the ISA server when
>> the ISA box is up, and to detect when it's down and route
>> queries directly out.
>>
>> I have given the crawlers/faq's a good workout but can't seem
>> to find out
>> exactly how to configure squid for this particular configuration.
>> I basically want squid to route all requests through the ISA box, and
>> if the queries timeout (eg BSOD) then route them directly for a while.
>>
>> The reason the ISA box needs to stay in the loop is to keep traffic
>> loads on it so I can do proper debugging, but I don't want to have to
>> expose the users to service outages.
>>
>> Can somebody help me, if not, point me in the right direction?
>>
>> thanks in advance
>>
>>
>>
>>
>> ben
>>
>>
>>
Received on Tue May 22 2001 - 21:42:59 MDT