|
| We have a pre-production 3 node WAS 6.0.2.14 cluster hosting WPS 5.1.0.4. When starting the nodes through the DMGR if you try to start them all at once (using the Application Servers page or the Cluster page) one of them always has issues. The server re
ports that its started, but you can't reach the WPS URL. Which member fails seems to be random. Sometimes it?s the primary node, and sometimes its one of the secondary nodes. If you start just two of them at once they both start fine, then you can star
t the third fine after they start. Combing through the logs I find that there is an issue the wpsDS datasource on the failing node each time it happens - obviously pointing at a DB2 issue since it will work when started 'alone'. I tried to use the DB2 t
ools (Activity Monitor, Health Monitor, etc) to see if I could see any funkyness, but none of them report any problems. I am not much of a DB2 guy so I'm not sure what else I can do to see what the problem is.
I am concerned that this problem during startup will manifest itself in an ugly way when the servers start to take on a user load.
Has anyone seen anything like this before? Any troubleshooting tips on the DB2 side that you can toss out to me to use?
Thanks.
|
|