XAResource problem was Re: Naming test timeout was Re: [jboss-dev] PUSH

[Adrian Brock]

The first mention of this "broken" XAResource looks to be after the
ha invoker tests, can you have a look at the log to see
if there are any other clues?

This is about 13 minutes before the naming test hangs.

http://hudson.jboss.org/hudson/job/JBoss-AS-5.0.x-TestSuite-sun15/lastBuild/artifact/trunk/build/output/jboss-5.0.0.CR1/server/all/log/server.log

2008-06-26 03:52:16,940 DEBUG
[org.jboss.deployers.plugins.deployers.DeployersImpl] (RMI TCP
Connection(349)-10.18.94.175) Fully Undeployed
vfszip:/qa/services/hudson/
hudson_workspace/workspace/JBoss-AS-5.0.x-TestSuite-sun15/trunk/testsuite/output/lib/ha-invoker.sar
2008-06-26 03:52:18,312 DEBUG [com.arjuna.ats.arjuna.logging.arjLogger]
(Thread-19) Periodic recovery - second pass <Thu, 26 Jun 2008 03:52:18>
2008-06-26 03:52:18,312 DEBUG [com.arjuna.ats.arjuna.logging.arjLogger]
(Thread-19) AtomicActionRecoveryModule: Second pass 
2008-06-26 03:52:18,319 DEBUG
[com.arjuna.ats.arjuna.logging.arjLoggerI18N] (Thread-19)
[com.arjuna.ats.internal.arjuna.recovery.TransactionStatusConnector_4] -
Transa
ctionStatusManager process for uid bcb0 is ALIVE. connected to host:
10.18.0.134, port: 34714 on socket:
Socket[addr=/10.18.0.134,port=34714,localport=34373]
2008-06-26 03:52:18,321 DEBUG
[com.arjuna.ats.arjuna.logging.arjLoggerI18N]
(Server.Connection:10.18.0.134:34373)
[com.arjuna.ats.arjuna.recovery.ActionStatusService_1
] -
transactionType: /StateManager/BasicAction/TwoPhaseCoordinator/AtomicAction uid: a120086:bcb0:486344fa:3e90 
 Status is 7
2008-06-26 03:52:18,335 WARN  [com.arjuna.ats.jta.logging.loggerI18N]
(Thread-19)
[com.arjuna.ats.internal.jta.resources.arjunacore.norecoveryxa]
[com.arjuna.ats.inter
nal.jta.resources.arjunacore.norecoveryxa] Could not find new XAResource
to use for recovering non-serializable XAResource < 131075, 28, 26,
49459749504848565458989998
48585256545152521029758511015748974950484856545898999848585256545152521029758511015753 >
2008-06-26 03:52:18,336 WARN
[com.arjuna.ats.arjuna.logging.arjLoggerI18N] (Thread-19)
[com.arjuna.ats.internal.arjuna.gandiva.inventory.StaticInventory_1] -
cannot 
find null implementation.


On Thu, 2008-06-26 at 16:10 +0100, Andrew Dinn wrote:
> Adrian Brock wrote:
>   > This test passes for me locally, so there's obviously something
> > happening different on the test machine.
> > 
> > . . .
> > 2) TM Warnings (these are repeated)
> > 
> > 2008-06-26 04:05:18,631 WARN  [com.arjuna.ats.jta.logging.loggerI18N]
> > (Thread-19)
> > [com.arjuna.ats.internal.jta.resources.arjunacore.norecoveryxa]
> > [com.arjuna.ats.inter
> > nal.jta.resources.arjunacore.norecoveryxa] Could not find new XAResource
> > to use for recovering non-serializable XAResource < 131075, 28, 26,
> > 49459749504848565458989998
> > 48585256545152521029758511015748974950484856545898999848585256545152521029758511015753 >
> 
> Hmm, this normally only happens after a crash and restart. Could a 
> previous test have crashed and left entries in the tx object store?
> 
> > 2008-06-26 04:05:18,631 WARN
> > [com.arjuna.ats.arjuna.logging.arjLoggerI18N] (Thread-19)
> > [com.arjuna.ats.internal.arjuna.gandiva.inventory.StaticInventory_1] -
> > cannot 
> > find null implementation.
> 
> Oh boy, null is meant to be a classname -- the log was obviously hosed! 
> This accounts for the rest of the messages -- garbage out garbage in. If 
> the TX store is unreadable like this then there must have been a crash 
> in the middle of writing it on a previous run or else something worse 
> like running out of disk space or a disk error during the previous or 
> current run. My money is on it not being a bug in the TX code :-)
> 
> regards,
> 
> 
> Andrew Dinn
> -----------
> 
> 
> 
> _______________________________________________
> jboss-development mailing list
> jboss-development at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/jboss-development
-- 
xxxxxxxxxxxxxxxxxxxxxxxxxxxx
Adrian Brock
Chief Scientist
JBoss, a division of Red Hat
xxxxxxxxxxxxxxxxxxxxxxxxxxxx