Since I have had no luck in determining the cause of the exception, I am
trying to at least handle it gracefully. What I am considering is applying
the following patch to the InactivityMonitor in order to trigger a fail over
on the transport...
try
{
WRITE_CHECK_TIMER.scheduleAtFixedRate(writeCheckerTask,
initialDelayTime,writeCheckTime);
READ_CHECK_TIMER.scheduleAtFixedRate(readCheckerTask,
initialDelayTime,readCheckTime);
}
catch(Exception e)
{
LOG.error("Logging exception in
InactivityMonitor.startMonitorThreads() and trying to restart transport: ");
LOG.error(getStackTraceString(e));
ASYNC_TASKS.execute(new Runnable() {
public void run() {
onException(new IOException("A previously
uncaught exception in InactivityMonitor.startMonitorThreads() is now being
used to interrupt (and hopefully restart) the transport"));
};
});
}
Does this seem reasonable?
--
View this message in context:
http://www.nabble.com/Exception-in-thread-ActiveMQ-Transport-tp18570264p18598452.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
