You ran the lockmgr and query.timeout tests under
openjpa-persistence-jdbc with the derby-locking profile below? Most of
the lockmgr and query timeout tests failed when I tried using
waitTimeout=6 while working on OPENJPA-1071....
-Donald
Michael Dick (JIRA) wrote:
[
https://issues.apache.org/jira/browse/OPENJPA-1210?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Michael Dick updated OPENJPA-1210:
----------------------------------
Remaining Estimate: 0h
Original Estimate: 0h
Currently trunk uses the default derby.locks.waitTimeout (60) and a derby.locks.deadlockTimeout of 5. I suspect that these can be tuned for the developer's hardware and provide quicker test execution.
<anecdotal evidence>
setting waitTimeout = 10 saved me 5 - 10 minutes on a build last night.
</anecdotal evidence>
The properties can be specified on the command line :
$ mvn -Dderby.locks.deadlockTimeout=10 -Dderby.locks.waitTimeout=15 clean test
or in settings.xml :
<profile>
<id>derby-locking</id>
<properties>
<derby.locks.deadlockTimeout>3</derby.locks.deadlockTimeout>
<derby.locks.waitTimeout>5</derby.locks.waitTimeout>
</properties>
</profile>
The profile (as shown) needs to be activated on the command line :
$ mvn -Pderby-locking clean test
Specify derby timeouts via properties.
--------------------------------------
Key: OPENJPA-1210
URL: https://issues.apache.org/jira/browse/OPENJPA-1210
Project: OpenJPA
Issue Type: Sub-task
Reporter: Michael Dick
Assignee: Michael Dick
Original Estimate: 0h
Remaining Estimate: 0h
