[JBoss JIRA] (JBTM-3034) CMR recovery wrongly handles commit and rollback
by Tom Jenkinson (Jira)
[ https://issues.jboss.org/browse/JBTM-3034?page=com.atlassian.jira.plugin.... ]
Tom Jenkinson closed JBTM-3034.
-------------------------------
Resolution: Done
> CMR recovery wrongly handles commit and rollback
> -------------------------------------------------
>
> Key: JBTM-3034
> URL: https://issues.jboss.org/browse/JBTM-3034
> Project: JBoss Transaction Manager
> Issue Type: Bug
> Components: JTA
> Affects Versions: 5.8.1.Final
> Reporter: Ondra Chaloupka
> Assignee: Ondra Chaloupka
> Priority: Major
> Fix For: 5.9.0.Final, 5.5.33.Final
>
> Attachments: JBTM-3034-xarm-to-query-cmrrm.diff
>
>
> The recovery of CMR works wrongly.
> For scenario I currently investigate there is issue the second resource beging committed and rolled-back too.
> # cmr resource prepare (no real action on the local transction)
> # xa resource prepare (prepared in real as XA)
> # cmr resource commit (commiting the local transaction)
> # JVM crash
> # expecting the xa resource being committed, but it's committed and immediatelly rolled-back. fortunatelly it seems it does not causes data consistency issue.
> This is similar to what was seen in issue https://issues.jboss.org/browse/JBEAP-6326 but not the same. The seems could be connected with fix for https://issues.jboss.org/browse/JBTM-2734. More investigation is needed.
> This is *regression* against EAP 7.0.0. The same scenario works in 7.0.0 smoothly.
--
This message was sent by Atlassian Jira
(v7.12.1#712002)
5 years, 11 months
[JBoss JIRA] (JBTM-3034) CMR recovery wrongly handles commit and rollback
by Tom Jenkinson (Jira)
[ https://issues.jboss.org/browse/JBTM-3034?page=com.atlassian.jira.plugin.... ]
Tom Jenkinson updated JBTM-3034:
--------------------------------
Fix Version/s: 5.5.33.Final
> CMR recovery wrongly handles commit and rollback
> -------------------------------------------------
>
> Key: JBTM-3034
> URL: https://issues.jboss.org/browse/JBTM-3034
> Project: JBoss Transaction Manager
> Issue Type: Bug
> Components: JTA
> Affects Versions: 5.8.1.Final
> Reporter: Ondra Chaloupka
> Assignee: Ondra Chaloupka
> Priority: Major
> Fix For: 5.5.33.Final, 5.9.0.Final
>
> Attachments: JBTM-3034-xarm-to-query-cmrrm.diff
>
>
> The recovery of CMR works wrongly.
> For scenario I currently investigate there is issue the second resource beging committed and rolled-back too.
> # cmr resource prepare (no real action on the local transction)
> # xa resource prepare (prepared in real as XA)
> # cmr resource commit (commiting the local transaction)
> # JVM crash
> # expecting the xa resource being committed, but it's committed and immediatelly rolled-back. fortunatelly it seems it does not causes data consistency issue.
> This is similar to what was seen in issue https://issues.jboss.org/browse/JBEAP-6326 but not the same. The seems could be connected with fix for https://issues.jboss.org/browse/JBTM-2734. More investigation is needed.
> This is *regression* against EAP 7.0.0. The same scenario works in 7.0.0 smoothly.
--
This message was sent by Atlassian Jira
(v7.12.1#712002)
5 years, 11 months
[JBoss JIRA] (JBTM-3034) CMR recovery wrongly handles commit and rollback
by Tom Jenkinson (Jira)
[ https://issues.jboss.org/browse/JBTM-3034?page=com.atlassian.jira.plugin.... ]
Tom Jenkinson reopened JBTM-3034:
---------------------------------
> CMR recovery wrongly handles commit and rollback
> -------------------------------------------------
>
> Key: JBTM-3034
> URL: https://issues.jboss.org/browse/JBTM-3034
> Project: JBoss Transaction Manager
> Issue Type: Bug
> Components: JTA
> Affects Versions: 5.8.1.Final
> Reporter: Ondra Chaloupka
> Assignee: Ondra Chaloupka
> Priority: Major
> Fix For: 5.5.33.Final, 5.9.0.Final
>
> Attachments: JBTM-3034-xarm-to-query-cmrrm.diff
>
>
> The recovery of CMR works wrongly.
> For scenario I currently investigate there is issue the second resource beging committed and rolled-back too.
> # cmr resource prepare (no real action on the local transction)
> # xa resource prepare (prepared in real as XA)
> # cmr resource commit (commiting the local transaction)
> # JVM crash
> # expecting the xa resource being committed, but it's committed and immediatelly rolled-back. fortunatelly it seems it does not causes data consistency issue.
> This is similar to what was seen in issue https://issues.jboss.org/browse/JBEAP-6326 but not the same. The seems could be connected with fix for https://issues.jboss.org/browse/JBTM-2734. More investigation is needed.
> This is *regression* against EAP 7.0.0. The same scenario works in 7.0.0 smoothly.
--
This message was sent by Atlassian Jira
(v7.12.1#712002)
5 years, 11 months
[JBoss JIRA] (JBTM-3079) InboundBridge recovery aborts live transactions
by Tom Jenkinson (Jira)
[ https://issues.jboss.org/browse/JBTM-3079?page=com.atlassian.jira.plugin.... ]
Tom Jenkinson updated JBTM-3079:
--------------------------------
Fix Version/s: 5.next
5.5.34.Final
> InboundBridge recovery aborts live transactions
> -----------------------------------------------
>
> Key: JBTM-3079
> URL: https://issues.jboss.org/browse/JBTM-3079
> Project: JBoss Transaction Manager
> Issue Type: Bug
> Components: TxBridge
> Affects Versions: 5.5.32.Final, 5.9.0.Final
> Environment: EAP 7.1.5 / 5.5.32.Final
> Reporter: Jonathan Halliday
> Assignee: Tom Jenkinson
> Priority: Critical
> Fix For: 5.5.34.Final, 5.next
>
> Attachments: jbtm3079.patch
>
>
> During a recovery pass, the InboundBridgeRecoveryManager scans for subordinate XA branches that may need cleanup. The filtering process applied by checkXid correctly excludes tx that are not owned by the bridge, but fails to ignore those that are owned but also still live. Therefore, a race condition exists such that the recovery process may incorrectly abort a branch if invoked between the prepare and commit steps, resulting in data corruption relative to other committed branches from the parent i.e. Heuristic outcomes.
> TRACE [org.jboss.jbossts.txbridge] (TaskWorker-3) BridgeDurableParticipant.prepare(Xid=< 131080, 35, 64, ... >)
> TRACE [org.jboss.jbossts.txbridge] (Periodic Recovery) rolling back orphaned subordinate tx < 131080, 35, 64, ... >
> ERROR [org.jboss.jbossts.txbridge] (TaskWorker-8) ARJUNA033004: commit on Xid=< 131080, 35, 64, ... > failed: javax.transaction.xa.XAException
> It is necessary to enhance checkXid to validate against known live tx. The easiest way would seem to be to have the InboundBridgeManager singleton hold a lookup table of live tx, effectively a secondary index into its existing collection of InboundBridges, against which the recovery system can validate the in-doubt Xid.
--
This message was sent by Atlassian Jira
(v7.12.1#712002)
5 years, 11 months
[JBoss JIRA] (JBTM-2967) Some quickstarts are not tested in CI
by Tom Jenkinson (Jira)
[ https://issues.jboss.org/browse/JBTM-2967?page=com.atlassian.jira.plugin.... ]
Tom Jenkinson updated JBTM-2967:
--------------------------------
Fix Version/s: 5.next
(was: 5.9.1.Final)
> Some quickstarts are not tested in CI
> -------------------------------------
>
> Key: JBTM-2967
> URL: https://issues.jboss.org/browse/JBTM-2967
> Project: JBoss Transaction Manager
> Issue Type: Bug
> Components: Demonstrator
> Affects Versions: 5.7.1.Final
> Reporter: Michael Musgrove
> Assignee: Michael Musgrove
> Priority: Major
> Fix For: 5.next
>
>
> Most quickstarts used to ship with a run.[sh|bat] script for running the quickstart and we used to test them in CI by executing the run script. At some point we changed the CI script (scripts/hudson/quickstart.sh) to just run the pom instead (./build.sh -B clean install). Since not all poms execute their run script we aren't getting full CI coverage.
> This JIRA is to go through each quickstart and ensure the the pom does indeed exercise the quickstart.
> The following poms do execute a run script
> transactionaldriver-jpa-and-tomcat/pom.xml
> rts/at/simple/pom.xml
> rts/lra/lra-test/pom.xml
> ArjunaJTS/interop/glassfish/pom.xml
> transactionaldriver-and-tomcat/pom.xml
> spring/camel-with-narayana-spring-boot/pom.xml
> The following quickstarts contain run scripts:
> transactionaldriver-jpa-and-tomcat/run.sh
> ArjunaCore/txoj/run.sh
> jboss-as/build/target/wildfly-9.0.0.CR1-SNAPSHOT/bin/run.sh
> ArjunaJTA/object_store/run.sh
> ArjunaJTA/maven/run.sh
> ArjunaJTA/recovery/run.sh
> ArjunaJTA/javax_transaction/run.sh
> rts/at/undertow/run.sh
> rts/at/recovery/recovery2/run.sh
> rts/at/recovery/recovery1/run.sh
> rts/at/simple/run.sh
> rts/at/service/service2b/run.sh
> rts/at/service/service2/run.sh
> rts/at/service/service1b/run.sh
> rts/at/service/service1/run.sh
> rts/at/demo/run.sh
> rts/lra/run.sh
> ArjunaJTS/standalone/run.sh
> ArjunaJTS/interop/glassfish/run.sh
> ArjunaJTS/recovery/run.sh
--
This message was sent by Atlassian Jira
(v7.12.1#712002)
5 years, 12 months