Gurunath, are you using NIO like Ishitori is? Or is that an unrelated red herring?
Ishitori, did you ever submit a JIRA bug report for this? If you haven't, please do, since it sounds like there's a bug here and without a JIRA entry it's less likely to get investigated/fixed. I looked at the code ( http://grepcode.com/file/repo1.maven.org/maven2/org.apache.activemq/activemq-all/5.10.0/org/apache/activemq/network/MBeanBridgeDestination.java) and it looks to me like we're attempting to create a new MBean in both onInboundMessage() and onOutBoundMessage(). In both cases, we check for non-existence before doing the work, but the check is on a different map in each method (inboundDestinationViewMap and outboundDestinationViewMap, respectively). So we can only add one entry for inbound messages and one for outbound ones, but that still seems like it would allow double registration in the situation where we had both inbound and outbound messages. Am I looking at that right? Tim On Fri, May 22, 2015 at 12:05 PM, Ishitori <[email protected]> wrote: > Nope :( It still happens from time to time. Restart of the services helped > us, so we do it from time to time. > > > > -- > View this message in context: > http://activemq.2283324.n4.nabble.com/Broker-log-full-of-Failed-to-register-queue-messages-tp4685241p4696875.html > Sent from the ActiveMQ - User mailing list archive at Nabble.com. >
