Hi All,

I'm getting the below exception

"ResourceException thrown in J2CMethodBindingImpl.invoke() javax.resource.ResourceException: DSRA9110E: Connection is closed." after the source system got abnormally restarted. Post that we purged the Datasource(Purge Policy: Entire Pool) and Adapter-Activation Spec(Under: J2C connection factories > JCA lifecycle management)  but we were not able to get back the connection. we were still getting the connection closed exception.  (not able to  purge the fatal connections). The issue was later resolved by restarting the Server.(with 2Nodes).

if they have encounterd similar issue Could some please share some pointers on what needs to be done or anything else needs to be taken care for such issues so that we need not restart the server and the connection issues get resolved automatically after the source system is up and running. 

Note: we have a connection to same DB from other server (Other Cluster).  

Hi Anupam,

  Please enable traces (*=info:WAS.j2c=all:RRA=all:Transaction=all) and reproduce the issue.
 
  Seems that you are having the next problem.
 
  How to resolve "DSRA9110E: Connection is closed." exceptions that occur in a own implemented J2EE Application?
  https://developer.ibm.com/answers/questions/233044/how-to-resolve-dsra9110e-connection-is-closed-exce.html
 
  Hope this helps. Tell us if you need more support
 
Regards

Do you see StaleConnectionException in the systemOut.log? 

Yes, Please find the below details.  

[11/15/16 3:00:08:706 CET] 00000152 OEBSRA001     E com.ibm.j2ca.oracleebs.runtime.OracleManagedConnection getWBIConnection CWYDB0201E: The adapter could not establish a connection to the database for the following reason: The Network Adapter could not establish the connection DSRA0010E: SQL State = 61000, Error Code = 20.  Explanation=The problem might be in the driver, the database, or the user credentials provided. The reason stated in the error message can help you identify the problem.  UserAction=Analyze the reason for the failure to establish a connection, and fix the related setup problems.

[11/15/16 3:00:08:765 CET] 00000152 MCWrapper     E   J2CA0021E: An exception occurred while trying to get a Connection from the Managed Connection resource  : javax.resource.ResourceException: The Network Adapter could not establish the connection DSRA0010E: SQL State = 61000, Error Code = 20

Caused by: com.ibm.websphere.ce.cm.StaleConnectionException: The Network Adapter could not establish the connection DSRA0010E: SQL State = 61000, Error Code = 20

11/15/16 3:00:31:993 CET] 00000156 WSJdbcConnect W   DSRA9400E: Fatal error occurred during Connection reassociation: javax.resource.spi.ResourceAllocationException: DSRA8100E: Unable to get a XAConnection from the DataSource.

[11/15/16 3:42:26:046 CET] 000006a8 MCWrapper     E   J2CA0081E: Method destroy failed while trying to execute method destroy on ManagedConnection com.ibm.j2ca.oracleebs.runtime.OracleManagedConnection@d7f7bfd8 from resource No longer available. Caught exception: javax.resource.ResourceException: com.ibm.websphere.ce.cm.ObjectClosedException: DSRA9110E: Connection is closed.

Above are few noticable lines from Sysoutlog.

Purge policy is "Entire Pool". Post this error we were not able to release the stale connections automatically. we even tried purging manually and later had to do restart server.

As per IBM Documentation i have checked few key configurations in DS, Minumum Connection = 0 , Unused Reap time = 1800 seconds .

You need to verify whether your application code catches the StaleConnectionException, so that it can get a new connection from the pool.

Demystifying the WebSphere StaleConnectionException
http://www-01.ibm.com/support/docview.wss?uid=swg21063645