From: "Peter Dobszai"
Sender: "Peter Dobszai"
Reply-To: "Peter Dobszai"
Subject: Cluster pool using WTC
Newsgroups: weblogic.developer.interest.clustering
X-User-Info: 193.225.155.254
NNTP-Posting-Host: 193.225.155.254
X-Original-NNTP-Posting-Host: 193.225.155.254
Message-ID: <3f0951fd@newsgroups.bea.com>
Date: 7 Jul 2003 03:57:01 -0700
X-Trace: newsgroups.bea.com 1057575421 193.225.155.254 (7 Jul 2003 03:57:01 -0700)
X-Original-Trace: 7 Jul 2003 03:57:01 -0700, 193.225.155.254
Organization: BEA NEWS SITE
Lines: 25
XPident: Unknown
Path: newsgroups.bea.com!not-for-mail
Xref: newsgroups.bea.com weblogic.developer.interest.clustering:9427


Hi!

I'm having the following problem with WLS 7.0 SP2 on Win2K.
In the configuration there're 2 managed servers running in a cluster,
an EJB deployed to the cluster, a jdbc pool deployed to the cluster and
a WTCServer deployed to a server (either admin or managed).
The pool holds connections using the WTCServer to a Tuxedo domain,
the bean calls a Tuxedo service using the pool.

My client calls the bean in the cluster continously. Everything is OK, until
the client runs only in 1 instance, but if I start another instance, after some
time the call is stuck. If I kill the running client, the other one throws an
exception,
but the next call is OK.
If only one managed server is running, the clients run fine.

I tried to set up 2 WTCServers, two pools and a multipool. The pools and WTC servers
are deployed only to one server (pool1 and wtcserver1 to server1), and the multipool
to the cluster. Each pool reads its own wtc server instance. But the result is
the same.

Please help me, I have no further ideas...

Thank you,Peter