"ron.sigal(a)jboss.com" wrote : I looked at some old logs, and it looks like the
problem is associated with tests that deliberately kill the server
|
Yes, I noticed this during crash and failover tests.
anonymous wrote : The NullPointerException seems to occur because the client invoker has
closed before the ControlMonitorTimerTask has stopped running. I'm not sure how this
is happening, but it could be that the TimerTask was in the middle of running when the
client invoker closed.
|
When using a timer, you can ensure the task has finished running by doing the following:
(from java.util.Timer javadoc - cancel method)
anonymous wrote :
| Note that calling this method from within the run method of a timer task that was
invoked by this timer absolutely guarantees that the ongoing task execution is the last
task execution that will ever be performed by this timer.
|
So you just need to add a sychronized stop() method to the TimerTask which calls cancel()
and make sure you call this from your code - take a look at
org.jboss.messaging.core.plugin.postoffice.cluster.StatsSender which does something very
similar.
anonymous wrote :
| In any case, I've modified the code to avoid the NPE and print a warning message.
|
Is a warning message really appropriate? A warning would imply something wrong or strange
- but this is to be expected.
anonymous wrote :
| I ran the JBM functional tests, and they all pass except for occasional failures in
FailoverValveTest at line 287
|
|
| | assertNotNull(o);
| |
|
| I reran FailoverValveTest with the remoting jar from the SVN repository and the same
thing happened, so I don't think I broke anything new.
FailoverValve is not currently used so I wouldn't worry too much about this. We need
to address this in the test suite.
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4027942#...
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...