[
https://jira.jboss.org/jira/browse/JBAS-6990?page=com.atlassian.jira.plug...
]
David Lloyd resolved JBAS-6990.
-------------------------------
Resolution: Done
This was caused by the fact that the naming service fires off long-lived tasks that
quickly starve the thread pool of core threads; since the queue never gets filled up, any
subsequent tasks are never run, causing later tests to time out.
The fix was to introduce a second thread pool - "LongRunningTasksThreadPool" -
which will always run tasks in a new or pooled thread and never queue, and has a very high
thread limit. The naming service is then pointed at this thread pool instead. Any other
service which uses an Executor to start up long-living tasks like acceptor threads or
connection handlers for blocking I/O servers, etc., should use this thread pool instead of
the "regular" one.
PooledInvokerUnitTestCase failure
----------------------------------
Key: JBAS-6990
URL:
https://jira.jboss.org/jira/browse/JBAS-6990
Project: JBoss Application Server
Issue Type: Bug
Security Level: Public(Everyone can see)
Components: ProfileService
Affects Versions: JBossAS-6.0.0.Alpha1
Reporter: Scott Marlow
Assignee: David Lloyd
Fix For: JBossAS-6.0.0.Alpha1
To see the failure, do this:
1. cd testsuite
2. ./build.sh -Dtest=org.jboss.test.naming.test.PooledInvokerUnitTestCase one-test
You will see:
[junit] Running org.jboss.test.naming.test.PooledInvokerUnitTestCase
[junit] Tests run: 1, Failures: 0, Errors: 1, Time elapsed: 0 sec
[junit] Test org.jboss.test.naming.test.PooledInvokerUnitTestCase FAILED (timeout)
Also, if you try to access the jndi naming service via the jmx console, you will also see
faliures (after running the PooledInvokerUnitTestCase).
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira