CQ5.3: Replication agents fail due to maximum job queue limit

Issue

You get the following error in your log file when you create more replication agents than the number of available job queues in CQ5.3
23.08.2010 13:26:53.400 *WARN* [pool-3-thread-3] org.apache.sling.event.impl.JobEventHandler Unable to create new job queue named com/day/cq/replication/job/publish11 as there are already 10 job queues. Try to increase the maximum number of job queues!

Cause

The org.apache.sling.event.impl.JobEventHandler has a default limit of 10 on the number of job queues that the system can create.

Resolution

Increase the maximum number of Job queues as needed (the number of queues should be higher than the number of configured replication agents + number of workflow models)

1. Log into the Felix Console (http://host:port/system/console/configMgr)
2. Go to the Configuration tab in Felix console
3. Select "Apache Sling Job Event Handler (org.apache.sling.event.impl.JobEventHandler)" from the dropdown menu
4. Change "Max Job Queues" value
5. Click save

Note:

• The number of queues should be higher than the number of configured replication agents + total number of workflow models.
• If a queue is running then the system is either processing a workflow or replication, the overhead associated with this is an additional thread.
• Configuring 'too many queues' will not cause a problem - it does not occupy any resources. The configuration is in place to avoid and detect misuse of queues in 3rd party application code.

Applies To

CQ5.3 with Feature Pack 30015 applied