The system i'm currently testing with is only a little test system,
it is not inteded to go productive in that configuration ;-)
But thanks for the tip you anyway. The jk.log has no error messages,
and here is the log grepped for "found best worker":

[Tue Oct 25 15:52:23 2005] [28304:49172] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi1) using 
by request method
[Tue Oct 25 15:52:23 2005] [28315:16410] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi2) using 
by request method
[Tue Oct 25 15:52:23 2005] [28311:49176] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi1) using 
by request method
[Tue Oct 25 15:52:23 2005] [28307:0021] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi2) using 
by request method
[Tue Oct 25 15:52:23 2005] [28269:32771] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi1) using 
by request method
[Tue Oct 25 15:52:25 2005] [28302:32787] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi2) using 
by request method
[Tue Oct 25 15:52:28 2005] [28312:0025] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi1) using 
by request method
[Tue Oct 25 15:52:32 2005] [28308:16406] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi2) using 
by request method
[Tue Oct 25 15:52:38 2005] [28317:32795] [debug] 
get_most_suitable_worker::jk_lb_worker.c (539): found best worker (debi1) using 
by request method


I don't know how the mod_jk wants to figure out which parameters it should use
to identify requests belonging to one session. Do you have an idea? I'm not 
a c expert, but I think I will taje a look at the sources, maybe I find 
something
about that.

bye Tom


> IMO having 10 connections limit is way too low for any serious usage.
> I would suggest that you rise maxProcessor to at least 100.
>
> Also do you observe any error messages in mod_jk.log?
>
> Regards,
> Mladen.

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to