|
Hi all,
This might be a bit off topic as it is weblogic
specific, but I have already tried all that I can.
I encountering the following problem when running cluster servlets. Hope that someone out there can enlighten me: Server setup: 2 application servers (Win NT) running it's own cluster servlet, supported by a load balancer to route the http request. The EJBs are also running at each of the servers. We are using Weblogic 5.1 with the latest patch. 1. Intermitently, one of the weblogic instance will go down unknowingly, SOMETIMES reported by a 'Dr Watson'. 2. This will cause both cluster servlets to be hanged. The other weblogic instance at the other server is alright as we were able to access directly to the instance, but not via the loadbalancer IP address. 3. We are not able to bring the 2 clusterservlets down as it reports server not found. 4. The only solution is th reboot both servers. Below is the error msg in the failed weblogic instance (though we are not sure if it is the cause): ------------- Thu Sep 27 14:37:41 GMT+08:00 2001:<W> <ListenThread> Connection rejected: 'Login timed out after 5000 msec. The socket came from [host=xxx,port=xxx,localport=xxx] See property weblogic.login.readTimeoutMillis to increase of decrease timeout. See property weblogic.login.logReadTimeoutsProp to turn off these log messages.' ------------- Pls help as we were not able to simulate this to BEA sinces it is ad hoc, and they can't seem to know the answer. Thanks!! Gina |
