Re: [SQU] two parallel parent-caches : does not work

From: Peter Pilsl <pilsl@dont-contact.us>
Date: Tue, 23 Jan 2001 23:06:45 +0100

>
> Which error do you get?
>
> (squid is supposed to fall back on direct when it cannot get to the
> selected parent..)
>

Thanx for your response:

I just shut the nic down where the route to proxy1 is crossing and got this error:
-------
ERROR
The requested URL could not be retrieved
While trying to retrieve the URL: http://somewhere/somesub
The following error was encountered:
Connection Failed
The system returned:
    (101) Network is unreachable
The remote host or network may be down. Please try the request again.
Your cache administrator is me@myhost
Generated Tue, 23 Jan 2001 21:23:38 GMT by squidhost (Squid/2.3.STABLE4)
------

while proxy2 was still fully reachable. When I just transpose the two
cache_peer lines in squid.conf and put
'cache_peer proxy2 parent 8080 0 no-query'
as first line and make a 'killall -HUP squid' I will get the requested page.

But I discovered that when I try to refresh after getting the above
errormessage about 3 to 5 times, squid will finally switch to proxy2
and give me the requested page.

When implementing the round-robin-options squid is cycling between the
two parents very fine and also delivers the page when one proxy is not
reachable, but is very slow then (seems, it tries to connect the other
proxy very often)

I dont think that this is how squid should behave at all.

peter

ps:
To Henrik: sorry for the previous half-finished mail to your private
address. It 'escaped' mutt while writing (hit send-key instead of
postpone-key)

-- 
mag. peter pilsl
phone: +43 676 3574035
fax  : +43 676 3546512
email: pilsl@goldfisch.at
sms  : pilsl@max.mail.at
pgp-key available
--
To unsubscribe, see http://www.squid-cache.org/mailing-lists.html
Received on Tue Jan 23 2001 - 15:09:43 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:57:32 MST