Jeff Peng wrote:
> I have tested squid-3.0.stable13 for round-robin loadbalancing.
> I found it behaves as not the same as the before version of 3.0 branch.
> For example, there are two backend servers, server A and server B.
> From squid to server A the connection is fast, but to server B the
> connection is slow.
> So under this version of Squid it will deliver more requests to server
> A than server B, even it's using round-robin.
> But in before version of Squid it seems most likely to deliver the
> traffic averagely.
>
> Is it the right fact? Thanks.
>
> Jeff.
No changes were made that affect peering selection between STABLE9 and
STABLE13. Only one between STABLE11 and STABLE12 that should have had no
behavior affect.
This is unexpected behavior. Check your configuration carefully and that
server B is not being declared dead.
Amos
-- Please be using Current Stable Squid 2.7.STABLE6 or 3.0.STABLE13 Current Beta Squid 3.1.0.5Received on Fri Feb 06 2009 - 12:29:09 MST
This archive was generated by hypermail 2.2.0 : Fri Feb 06 2009 - 12:00:02 MST