Thanks for the detailed answer, the bug you linked seems interesting but I'm
not sure it's relevant in this situation.
Unfortunately it happened again (the whole system collapsed), so I've got
new information.
I think there might be something wrong with DLQ handling. One of the clients
log is now full of the following messages:
WARN 2015.01.26 09:03:42,001
org.apache.activemq.store.kahadb.MessageDatabase - Duplicate message add
attempt rejected. Destination:
QUEUE://DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.ActiveMQ.DLQ.Topic.Consumer.HUNYADI.VirtualTopic.SFOperationShare,
Message id: ID:c-micro-1-46647-1421372713039-6:1:3:1:5592
WARN 2015.01.26 09:03:42,002 org.apache.activemq.broker.region.Queue -
duplicate message from store ID:c-micro-1-46647-1421372713039-6:1:3:1:5592,
redirecting for dlq processing
WARN 2015.01.26 09:03:42,017
org.apache.activemq.broker.region.cursors.AbstractStoreCursor -
org.apache.activemq.broker.region.cursors.QueueStorePrefetch@4300a1f8:DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.ActiveMQ.DLQ.Topic.Consumer.CLOUD.VirtualTopic.SFOperationShare,batchResetNeeded=false,storeHasMessages=true,size=62,cacheEnabled=true,maxBatchSize:200,hasSpace:true
- cursor got duplicate: ID:c-micro-1-46647-1421372713039-6:1:3:1:5592, 4
WARN 2015.01.26 09:03:42,018 org.apache.activemq.broker.region.Queue -
duplicate message from store ID:c-micro-1-46647-1421372713039-6:1:3:1:5592,
redirecting for dlq processing
WARN 2015.01.26 09:03:42,018
org.apache.activemq.broker.region.cursors.AbstractStoreCursor -
org.apache.activemq.broker.region.cursors.QueueStorePrefetch@7eb8d136:DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.ActiveMQ.DLQ.Topic.Consumer.CLOUD.VirtualTopic.SFOperationShare,batchResetNeeded=false,storeHasMessages=true,size=18,cacheEnabled=true,maxBatchSize:200,hasSpace:true
- cursor got duplicate: ID:c-micro-1-46647-1421372713039-6:1:3:1:5592, 4
The interesting part is how can a queue named
"DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.DLQ.ActiveMQ.DLQ.Topic.Consumer.HUNYADI.VirtualTopic.SFOperationShare"
can be created?
I tried to google it, but I haven't found anything like this.
Here is the broker configuration:
BrokerService brokerService = new BrokerService();
//
http://activemq.apache.org/message-redelivery-and-dlq-handling.html
brokerService.setSchedulerSupport(true);
// set dead letter startegy
IndividualDeadLetterStrategy idls = new
IndividualDeadLetterStrategy();
idls.setQueuePrefix("DLQ.");
idls.setUseQueueForQueueMessages(true);
idls.setUseQueueForTopicMessages(true);
idls.setProcessNonPersistent(true);
idls.setProcessExpired(false);
idls.setEnableAudit(false);
brokerService.setDestinationPolicy(new PolicyMap());
PolicyEntry pe = new PolicyEntry();
pe.setQueue(">");
pe.setDeadLetterStrategy(idls);
brokerService.getDestinationPolicy().setDefaultEntry(pe);
// redelivery policy
RedeliveryPlugin rdplugin = new RedeliveryPlugin();
rdplugin.setFallbackToDeadLetter(true);
rdplugin.setSendToDlqIfMaxRetriesExceeded(true);
RedeliveryPolicy rdp = new RedeliveryPolicy();
rdp.setInitialRedeliveryDelay(1000);
rdp.setRedeliveryDelay(1000);
rdp.setUseExponentialBackOff(false);
rdp.setMaximumRedeliveries(2);
rdplugin.getRedeliveryPolicyMap().setDefaultEntry(rdp); // for
all cases
brokerService.setPlugins(new BrokerPlugin[] { rdplugin });
Do you have any ideas?
Thanks,
Zoltan Nyari
--
View this message in context:
http://activemq.2283324.n4.nabble.com/ActiveMQ-embedded-broker-100-cpu-usage-tp4689765p4690398.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
