[JBoss JIRA] Updated: (JBTM-81) Delay start of recovery manager
by Mark Little (JIRA)
[ http://jira.jboss.com/jira/browse/JBTM-81?page=all ]
Mark Little updated JBTM-81:
----------------------------
Fix Version/s: 4.2.2
> Delay start of recovery manager
> -------------------------------
>
> Key: JBTM-81
> URL: http://jira.jboss.com/jira/browse/JBTM-81
> Project: JBoss Transaction Manager
> Issue Type: Feature Request
> Security Level: Public(Everyone can see)
> Components: JTA Implementation, JTS Implementation
> Affects Versions: 4.2
> Reporter: Mark Little
> Assigned To: Kevin Conner
> Priority: Minor
> Fix For: 4.2.2
>
> Time Spent: 1 day
> Remaining Estimate: 0 minutes
>
> Currently the Recovery Manager begins working as soon as JBossAS starts. When the local JTA implementation is used and there is a transaction to recover, we get the following stack trace:
> 10:52:35,529 WARN [loggerI18N] [com.arjuna.ats.internal.jta.resources.arjunacore.restorestate] [com.arjuna.ats.interna\
> l.jta.resources.arjunacore.restorestate] Exception on attempting to restore XAResource
> java.io.StreamCorruptedException: unexpected end of block data
> at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1321)
> at java.io.ObjectInputStream.defaultReadFields(ObjectInputStream.java:1912)
> at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:1836)
> at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1713)
> at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1299)
> at java.io.ObjectInputStream.readObject(ObjectInputStream.java:339)
> at com.arjuna.ats.internal.jta.resources.arjunacore.XAResourceRecord.restore_state(XAResourceRecord.java:879)
> at com.arjuna.ats.arjuna.coordinator.BasicAction.restore_state(BasicAction.java:1410)
> at com.arjuna.ats.arjuna.coordinator.BasicAction.activate(BasicAction.java:711)
> at com.arjuna.ats.arjuna.coordinator.BasicAction.activate(BasicAction.java:673)
> at com.arjuna.ats.arjuna.recovery.RecoverAtomicAction.<init>(RecoverAtomicAction.java:60)
> at com.arjuna.ats.internal.arjuna.recovery.AtomicActionRecoveryModule.doRecoverTransaction(AtomicActionRecovery\
> Module.java:178)
> This is because the rest of the system isn't yet fully initialised. The next periodic run of recovery does not suffer this problem and recovery completes successfully. We should look at delaying the start of the Recovery Manager until JBossAS is fully initialised.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
19 years, 10 months
[JBoss JIRA] Assigned: (JBTM-66) "Already marked for rollback" Exception could include original cause
by Mark Little (JIRA)
[ http://jira.jboss.com/jira/browse/JBTM-66?page=all ]
Mark Little reassigned JBTM-66:
-------------------------------
Assignee: Jonathan Halliday (was: Mark Little)
> "Already marked for rollback" Exception could include original cause
> --------------------------------------------------------------------
>
> Key: JBTM-66
> URL: http://jira.jboss.com/jira/browse/JBTM-66
> Project: JBoss Transaction Manager
> Issue Type: Feature Request
> Security Level: Public(Everyone can see)
> Components: JTA Implementation
> Affects Versions: 4.2
> Environment: JBoss 4.0.4.CR2
> Reporter: Richard Kennard
> Assigned To: Jonathan Halliday
> Priority: Optional
> Fix For: 4.6
>
> Original Estimate: 1 week
> Remaining Estimate: 1 week
>
> First up, thank you for providing such a wondefully stable implementation! I have a small feature request:
> At present whenever a, say, SFSB throws a RuntimeException, JTA goes into 'rollback' mode. From then on, anyone who tries to do any further work (including when the SessionContext tries to remove the rolled back SFSB) will trigger a different RuntimeException saying, in short, 'Already marked for rollback'.
> It would be very helpful if the 'Already marked for rollback' Exception could include, as its getCause (which is standard as of JDK 1.4), the original exception that started off the rollback? At present we are often left with the situation where all the original caller gets back is a 'Already marked for rollback' Exception, and there is no way to 'drill down' to find what the original cause was.
> Of course, you could also argue that the SessionContext should be smarter about trying to remove rolled back EJBs.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
19 years, 10 months
[JBoss JIRA] Updated: (JBTM-66) "Already marked for rollback" Exception could include original cause
by Mark Little (JIRA)
[ http://jira.jboss.com/jira/browse/JBTM-66?page=all ]
Mark Little updated JBTM-66:
----------------------------
Fix Version/s: 4.6
Affects Version/s: 4.2
(was: 4.4)
> "Already marked for rollback" Exception could include original cause
> --------------------------------------------------------------------
>
> Key: JBTM-66
> URL: http://jira.jboss.com/jira/browse/JBTM-66
> Project: JBoss Transaction Manager
> Issue Type: Feature Request
> Security Level: Public(Everyone can see)
> Components: JTA Implementation
> Affects Versions: 4.2
> Environment: JBoss 4.0.4.CR2
> Reporter: Richard Kennard
> Assigned To: Jonathan Halliday
> Priority: Optional
> Fix For: 4.6
>
> Original Estimate: 1 week
> Remaining Estimate: 1 week
>
> First up, thank you for providing such a wondefully stable implementation! I have a small feature request:
> At present whenever a, say, SFSB throws a RuntimeException, JTA goes into 'rollback' mode. From then on, anyone who tries to do any further work (including when the SessionContext tries to remove the rolled back SFSB) will trigger a different RuntimeException saying, in short, 'Already marked for rollback'.
> It would be very helpful if the 'Already marked for rollback' Exception could include, as its getCause (which is standard as of JDK 1.4), the original exception that started off the rollback? At present we are often left with the situation where all the original caller gets back is a 'Already marked for rollback' Exception, and there is no way to 'drill down' to find what the original cause was.
> Of course, you could also argue that the SessionContext should be smarter about trying to remove rolled back EJBs.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
19 years, 10 months