While WLS did install out off the box, its HA cluster
> wasn't able to handle a completely saturated (no think time) request queue.
> Once requests were in queue, all WLS test clients received errors.
I'm evaluating WLS, WLE and other rich featured HA appservers and want
to know more about your test/feature setup you describe.
"saturated request queue" ??
How are measuring saturation?
What's the use case? I.E.: N clients doing home.create(); myObj.method() or
just myObj.method() ??
The clients getting exceptions brings up the question I've been meaning to
ask:
Appservers support pools of beans to process client's
home.create()'s. Does this mean there's a 1:1 between
client's EJBObject's and members in the pool?
What happens to the client that does home.create() and
there's no more room in the pool? Is it queued?
Is there a DD option to cause this create to be queued?
(I.E. robustness and simplicity on the client side).
Thanks,
curt
--
Curt Smith
[EMAIL PROTECTED]
(w) 404-237-1166 x182
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
