[JBoss JIRA] (JBTM-1217) WSTX11-interop tests fail on pure IPv6 nodes
by Paul Robinson (JIRA)
Paul Robinson created JBTM-1217:
-----------------------------------
Summary: WSTX11-interop tests fail on pure IPv6 nodes
Key: JBTM-1217
URL: https://issues.jboss.org/browse/JBTM-1217
Project: JBoss Transaction Manager
Issue Type: Bug
Security Level: Public (Everyone can see)
Components: XTS
Reporter: Paul Robinson
Assignee: Paul Robinson
Fix For: 4.17.0, 5.0.0.M2
You can see the failure for the job here: https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/narayana-java6-ipv6/...
Although, this is now the job configured to test just the WSTX11-interop tests: https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/narayana-java6-ipv6-...
This can be reproduced by taking a test node, such as vmg36, offline. These are the commands I ran to reproduce. However, you may need to change the wget command to clone an appropriate workspace. Notice the comment that tells you to remove a dependency from a pom.xml. The build will fail if you miss this step out.
{code}
ssh probinso(a)dev89.mw.lab.eng.bos.redhat.com
ssh probinso(a)vmg36.mw.lab.eng.bos.redhat.com
wget http://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/narayana-java6-ipv6/w...
unzip narayana-java6-ipv6.zip
cd narayana-java6-ipv6
export MAVEN_OPTS="-DproxySet=true -DproxyHost=proxy-01-ipv6.mw.lab.eng.bos.redhat.com -DproxyPort=3128"
export IPV6_OPTS="-Djava.net.preferIPv4Stack=false -Djava.net.preferIPv6Addresses=true -Djboss.bind.address=[::1] -Djboss.bind.address.management=[::1] -Djboss.bind.address.unsecure=[::1]"
export IPV6_OPTS="$IPV6_OPTS $MAVEN_OPTS"
export WSTX_MODULES="WSTX11-interop"
export ARQ_PROF=arqIPv6
export PATH=/usr/lib/jvm/java-1.6.0/bin/:$PATH
export JAVA_HOME=/usr/lib/jvm/java-1.6.0/
export JBOSS_HOME=~/narayana-java6-ipv6/jboss-as/build/target/jboss-as-7.2.0.Alpha1-SNAPSHOT/
#Now Remove rest-at javadoc dep from narayana-full/pom.xml
./build.sh install -DskipTests=true $IPV6_OPTS -Dmaven.javadoc.skip=true
cd jboss-as
./build.sh install -DskipTests -Dts.smoke=false $IPV6_OPTS
cp ./build/target/jboss-as-7.2.0.Alpha1-SNAPSHOT/docs/examples/configs/standalone-xts.xml ./build/target/jboss-as-7.2.0.Alpha1-SNAPSHOT/standalone/configuration/
cd ..
./build.sh -f XTS/localjunit/pom.xml --projects "$WSTX_MODULES" -P$ARQ_PROF "$@" $IPV6_OPTS clean install
{code}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JBTM-1292) Add delay for all XTS ParticipantCompletion tests
by Paul Robinson (JIRA)
Paul Robinson created JBTM-1292:
-----------------------------------
Summary: Add delay for all XTS ParticipantCompletion tests
Key: JBTM-1292
URL: https://issues.jboss.org/browse/JBTM-1292
Project: JBoss Transaction Manager
Issue Type: Task
Security Level: Public (Everyone can see)
Components: XTS
Reporter: Paul Robinson
Assignee: Paul Robinson
Fix For: 4.17.1, 5.0.0.M2
When a WSBA participant employs the ParticipantCompletion protocol, it is the responsibility of the participant to notify the coordinator when it has completed its work. This notification is asynchronous.
In the 'WSBAParticipantCompletionTestCase' test, the client invokes the participant's web service who notifies completion just before returning from the invocation. The client then sends a message to the coordinator requesting to close (complete) the activity.
As the "complete" message (from the participant to coordinator) is asynchronous, we now have a race. If the Client's close message is processed by the coordinator before the participant's "complete" message, then the coordinator cancels the BA as not all participants have completed. This results in the client receiving a TransactionRolledBackException and the completed participant is (eventually) compensated. The outcome is atomic, but a BA that would have otherwise succeeded, is unsuccessful.
In reality we only expect this scenario to happen in the rather artificial scenario where all parties (client, coordinator and participants) are on the same server. It also only seems to happen on very slow machines. Therefore, it's fine to fix the test to prevent this scenario from arising, rather than to somehow change the protocol (without breaking the WS-BA standard) to prevent it.
We have two options, that I can see to fix the test:
1) Byteman Rendezvous. Here we would introduce a dependency on Byteman and write a script that delays the client's close message until all participants' 'complete' messages have been acknowledged by the coordinator. This is probably an over-engineered solution as we would be introducing Byteman, to these tests, for this single case.
2) We add a Thread.sleep(5000) to the test, just before the client sends the 'close' message to the coordinator. This is what we did in many of the XTS tests do far as a stop-gap until we decided how to do it "properly".
I suggest we go with 2) as it is the simplest solution.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JBTM-1267) JBossAS7ServerKillProcessor should wait for defunct java process to disappear
by Paul Robinson (JIRA)
Paul Robinson created JBTM-1267:
-----------------------------------
Summary: JBossAS7ServerKillProcessor should wait for defunct java process to disappear
Key: JBTM-1267
URL: https://issues.jboss.org/browse/JBTM-1267
Project: JBoss Transaction Manager
Issue Type: Sub-task
Security Level: Public (Everyone can see)
Components: TxBridge, XTS
Reporter: Paul Robinson
Assignee: Paul Robinson
Fix For: 4.16.5, 4.17.0, 5.0.0.M2
It's possible that this is what is causing JBTM-1259 as Arquillian may be detecting the defunct process.
Remember to apply the fix to the Kill Procesor in TXBridge too.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JBTM-1268) Test timeout for com.jboss.transaction.wstf.interop.Sc007Test#test3_10
by Paul Robinson (JIRA)
Paul Robinson created JBTM-1268:
-----------------------------------
Summary: Test timeout for com.jboss.transaction.wstf.interop.Sc007Test#test3_10
Key: JBTM-1268
URL: https://issues.jboss.org/browse/JBTM-1268
Project: JBoss Transaction Manager
Issue Type: Bug
Security Level: Public (Everyone can see)
Components: XTS
Reporter: Paul Robinson
Assignee: Paul Robinson
Fix For: 4.16.5, 4.17.0, 5.0.0.M2
The test took longer than the 120s default of the jmx Arquillian runner. The interop tests can be quite intensive, running the same operation many times. It's possible that on slow machines that this would take longer than 120s.
It could also be caused by a legitimate bug, but it's hard to tell from the log as these tests produce a lot of expected exceptions as they also test correct behavior under erroneous conditions.
We should increase the test timeout to 5mins to resolve this issue. We can re-open if we see it happen again.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months
[JBoss JIRA] (JBTM-1258) QA build does not archive JTS remote test output for both ORBs
by Michael Musgrove (JIRA)
Michael Musgrove created JBTM-1258:
--------------------------------------
Summary: QA build does not archive JTS remote test output for both ORBs
Key: JBTM-1258
URL: https://issues.jboss.org/browse/JBTM-1258
Project: JBoss Transaction Manager
Issue Type: Bug
Security Level: Public (Everyone can see)
Components: Build System
Affects Versions: 5.0.0.M1
Reporter: Michael Musgrove
Assignee: Tom Jenkinson
Fix For: 5.0.0.M1
The QA test suite runs the remote JTS tests against JacORB and the JDK orb but only the last run is archived.
The fix should save the first run in an archive:
testoutput/tests-jtsremote.zip
The results from the second run will be in testoutput/jtsremote as normal
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
12 years, 4 months