1. mod_jk doesn't balance the load on the base of
 packets.

2. mod_jk works with sticky sessions so only new sessions
 are balanced. I belief but am not shure that it's just
 round robin.

3. Bill Barker claims that the load balancing is broken
 as the instances of mod_jk don't know the load of each
 other. So mod_jk will balance to some extend but not as
 good as it could/should.

> -----Original Message-----
> From: tomcat@(protected)]
> Sent: Friday, October 10, 2003 1:37 PM
> To: tomcat-user@(protected)
> Subject: mod_jk lbfactor strangeness
>
> I've an apache servers with 4 backend app servers and using mod_jk to
> balance the load over them.
>
> Two of the machines are a fair bit quicker than the other
> two, so I've adjusted the weighting with lbfactor
>
> app1 (slow) = lbfactor=100
> app2 (slow) = lbfactor=100
> app3 (fast) = lbfactor=150
> app4 (fast) = lbfactor=150
>
> Yet what I see is that app2 and app3 get most of the load?
>
> I've checked this with snoop(tcpdump) and counted the packets to the
> various app servers. And app2 and app3 defiantly seems to be getting
> more work. I've checked my host file and workers.properties and all
> seems right.
>

---------------------------------------------------------------------
To unsubscribe, e-mail: tomcat-user-unsubscribe@(protected)
For additional commands, e-mail: tomcat-user-help@(protected)