Hi All,

I am getting following threads in logs.

ThreadMonitor W WSVR0606W: Thread "WebContainer : 577" (000002c0) was previously reported to be hung but has completed. It was active for approximately 895119 milliseconds. There is/are 28 thread(s) in total in the server that still may be hung

ThreadMonitor W WSVR0605W: Thread "WebContainer : 686" (00000330) has been active for 753074 milliseconds and may be hung. There is/are 29 thread(s) in total in the server that may be hung

Please help how can i resolve.

we deployed application in cluster four nodes, in AIX environment. Before this application is working fine in LINUX. In AIX servers disk IO is getting 90 %, logs are not writting into disk. We disabled logs for application still also same problem.

Please help ,me. 

looks like your server is resouce constrained .  check cpu of each jvm and see if anyone is using a lot .

topas is the command which can help and which OS you are running WAS.

I would also ask you to take the head dump , thread dump and enable gc verbose logging

As per the information you provided above - logs shows that application is releasing the threads but it is taking long time. It can be due to slow disk or application is writing huge amount of data. Discuss with your developer, how much amount of data they expect the application to write to that disk. That gives you an idea on how intense your application is on I/O. Still, we cannot just say whats the reason behind the hung threads. 

Next Steps:
When you notice hung threads on the server ... issue "kill -3 PID'. This will generate thread dump, which is a snap shot of thread activity at that time. 

Then use the following article and analyse the thead dump. www.websphereusergroup.org/josephamrithr...

Look for threads in HUNG state and see where the java stack is pointing to.