weight is currently only used for ICP peerings.
For non-ICP peerings the order is always the same as you have in
squid.conf. Once Squid detects that the first have failed it will
switch to the second.
I'd recommend you to tune down the peer_connect_timeout.
Also note the difference between "never_direct allow all" and
"always_direct deny all".
Regards
Henrik Nordström
On Thursday 24 January 2002 00.56, --==[bMan]==-- wrote:
> O.K. I thought I had it but it's still not working properly.
>
> Here are my two lines from squid.conf:
>
> cache_peer parent cache1.foo.org 8080 0 weight=1
> always_direct deny all
> cache_peer parent cache2.foo.org 8080 0 weight=2
> always_direct deny all
>
> My thinking was that if the first proxy goes down, my cache could
> automatically flip to the second (that's why I used 'weight').
> However, from what I have observed, 'weight' tag has absolutely no
> difference in the way squid grabs its parent cache. It's ALWAYS
> the first in the line. If the first is down, it never goes down to
> the second. Henrik N., please, explain to me what I am doing wrong
> here or maybe it's something that cannot be done with Squid?
>
> Thanks.
-- MARA Systems AB, Giving you basic free Squid support Customized solutions, packaged solutions and priority support available on requestReceived on Tue Feb 12 2002 - 21:53:28 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:06:15 MST