atom feed15 messages in org.codehaus.btm.dev[btm-dev] [jira] Commented: (BTM-63) ...
FromSent OnAttachments
Ben Daici (JIRA)Jan 7, 2010 6:43 am 
Ben Daici (JIRA)Jan 7, 2010 6:48 am 
Ben Daici (JIRA)Jan 7, 2010 7:29 am 
Ludovic Orban (JIRA)Jan 7, 2010 7:33 am 
Ludovic Orban (JIRA)Jan 7, 2010 8:09 am 
Ludovic Orban (JIRA)Jan 7, 2010 8:09 am 
Ludovic Orban (JIRA)Jan 7, 2010 8:29 am 
Ludovic Orban (JIRA)Jan 7, 2010 8:36 am 
Ben Daici (JIRA)Jan 7, 2010 9:02 am 
Ludovic Orban (JIRA)Jan 7, 2010 10:00 am 
Ludovic Orban (JIRA)Jan 7, 2010 10:02 am 
Ben Daici (JIRA)Jan 7, 2010 10:18 am 
Ludovic Orban (JIRA)Jan 14, 2010 3:59 am 
Ben Daici (JIRA)Jan 14, 2010 4:38 am 
Ludovic Orban (JIRA)Jan 14, 2010 4:44 am 
Subject:[btm-dev] [jira] Commented: (BTM-63) ACCESSIBLE connection closed due to Recoverer failure (nothing prevents recoverer from running twice in parallel)
From:Ben Daici (JIRA) (ji@codehaus.org)
Date:Jan 7, 2010 10:18:35 am
List:org.codehaus.btm.dev

[
http://jira.codehaus.org/browse/BTM-63?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=205596#action_205596
]

Ben Daici commented on BTM-63:

Not sure about the DB locks but it's possible, at least in theory. We do have
long running transactions that lock resources. However, by looking at the
recovery durations, it looks as though the recovery takes about a minute or so
so I am not sure if locking plays any role. Thanks for the patch, we will test it and let you know how it went.

ACCESSIBLE connection closed due to Recoverer failure (nothing prevents
recoverer from running twice in parallel)

-----------------------------------------------------------------------------------------------------------------

Key: BTM-63 URL: http://jira.codehaus.org/browse/BTM-63 Project: BTM Issue Type: Bug Affects Versions: 1.3.3 Environment: SUSE Linux Enterprise Server 11 (x86_64) Reporter: Ben Daici Assignee: Ludovic Orban Fix For: 1.3.4

Attachments: btm.tar.bz2, emergency_fox_for_BTM-63.patch

By the look of things it seems the following sequence of events happens: - Recoverer kicks in and attempt to process data source oracle-asm
(aimprod88_Other.log.1 - 05 Jan 2010 23:22:53,285) - an attempt is made to change a AbstractXAStatefulHolder's state from IN_POOL
to IN_POOL (aimprod88_Other.log.1 - 05 Jan 2010 23:22:53,299) - an exception is thrown from AbstractXAStatefulHolder and the recovery of
oracle-asm fails; oracle-asm iss marked as "failed" (aimprod88_Other.log.1 - 05
Jan 2010 23:22:53,300) - DspThread-11-ContractWorkUnit attempts to acquire a connection from oracle-asm
and notices oracle-asm was marked as "failed" (aimprod88_Other.log.1 - 05 Jan
2010 23:22:53,355) - DspThread-11-ContractWorkUnit closes the XAPool which includes connection
T4CXAConnection@5917afa8 which is currently in use (state ACCESSIBLE)
(aimprod88_Other.log.1 - 05 Jan 2010 23:22:53,371) - DspThread-17-ContractWorkUnit05 gets SQLException when trying to use closed
connection T4CXAConnection@5917afa8 (aimprod88_ContractWorkUnit.log - Jan 2010
23:22:59,579)