11.9.5. Limitations of the XA Recovery Process

If the JBoss EAP server crashes after an XAResource commit method successfully completes and commits the transaction, but before the coordinator can update the log, you may see the following warning message in the log when you restart the server: This is because upon recovery, the JBoss Transaction Manager sees the transaction participants in the log and attempts to retry the commit. Eventually the JBoss Transaction Manager assumes the resources are committed and no longer retries the commit. In this situation, can safely ignore this warning as the transaction is committed and there is no loss of data.

To prevent the warning, set the com.arjuna.ats.jta.xaAssumeRecoveryComplete property value to true . This property is checked whenever a new XAResource instance cannot be located from any registered XAResourceRecovery instance. When set to true, the recovery assumes that a previous commit attempt succeeded and the instance can be removed from the log with no further recovery attempts. This property must be used with care because it is global and when used incorrectly could result in XAResource instances remaining in an uncommitted state.

If the JBoss EAP server crashes after the completion of an XAResource prepare() method call, all of the participating XAResources are locked in the prepared state and remain that way upon server restart, The transaction is not rolled back and the resources remain locked until the transaction times out or a DBA manually rolls back the resources and clears the transaction log.

When the server is under excessive load, the server log may contain the following warning message, followed by a stacktrace:

Under heavy load, the processing time taken by a transaction can overlap with the timing of the periodic recovery process’s activity. The periodic recovery process detects the transaction still in progress and attempts to initiate a rollback but in fact the transaction continues to completion. At the time the periodic recovery attempts but fails the rollback, it records the rollback failure in the server log. The underlying cause of this issue will be addressed in a future release, but in the meantime a workaround is available.

Increase the interval between the two phases of the recovery process by setting the com.arjuna.ats.jta.orphanSafetyInterval property to a value higher than the default value of 10000 milliseconds. A value of 40000 milliseconds is recommended. Please note that this does not solve the issue, instead it decreases the probability that it will occur and that the warning message will be shown in the log.