XADisk in WebSphere causes thread hung

[pille...@gmail.com]

Hi,

deploying XADisk 1.2.2 as a JCA resource adapter under WebSphere 8.5.5.9, the logs are showing dozens of blocks like this:

[3/31/16 13:17:56:202 CEST] 0000009c SystemOut     O 13:17:56.202 [Deferrable Alarm : 3] WARN  c.i.w.r.component.ThreadMonitorImpl - WSVR0605W: Thread "Default : 3" (00000067) has been active for 717,984 milliseconds and may be hung.  There is/are 1 thread(s) in total in the server that may be hung.
        at sun.misc.Unsafe.park(Native Method)
        at java.util.concurrent.locks.LockSupport.parkNanos(LockSupport.java:238)
        at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.awaitNanos(AbstractQueuedSynchronizer.java:2094)
        at java.util.concurrent.LinkedBlockingQueue.poll(LinkedBlockingQueue.java:479)
        at org.xadisk.filesystem.workers.FileSystemEventDelegator.run(FileSystemEventDelegator.java:77)
        at com.ibm.ejs.j2c.work.WorkProxy.run(WorkProxy.java:668)
        at com.ibm.ws.util.ThreadPool$Worker.run(ThreadPool.java:1881)

... which seems obviously related to XA Disk. 

So far, I didn't see a real problem in my application with that trace, but I'd like to know how critical it is. I didn't do much testing yet, but it scare me a bit...

Please note the exact file and line of XA Disk involved in this stacktrace change over time. Sometimes ObjectPoolReliever appears, sometimgs this is DeadLockDetector, sometimes other files. 

Thanks

[Nitin Verma]

Hi,

When xadisk is deployed (whether on standalone jvm, or in a java-ee server as JCA RA), it starts some worker threads for various purposes. Many times these workers are just waiting on some events to happen etc. So your server is only giving a warning thinking that these threads may actually be stuck. But for these xadisk worker threads, the warnings can be ignored (or suppressed if the server allows such configuration).

Thanks,
Nitin

[pille...@gmail.com]

Ok thank you for your feedback. I will ignore these statements.