Associated case: 316351 The CheckForSuspectedAgentsJob appears to be consuming an excessive amount of memory, mostly in hibernates StatefulPersistenceContext. Heap dump analysis org.hibernate.engine.StatefulPersistenceContext (208mb) is being referenced from a thread executing org.rhq.enterprise.server.scheduler.jobs.CheckForSuspectedAgentsJob Breaking down the 208mb, there is 16mb in the entityEntries member variable, 6mb in the entitiesByKey member variable and 5mb in the proxiesByKey member variable. That leaves around 180mb of what looks like just org.rhq.core.domain.resource.Resource_$$_javassist_50 (and possibly other object) instances. There are ~177,000 objects referenced in the heap dump as being associated with the StatefulPersistenceContext instance, I've only checked the class of the first 250 of which they are all Resource_$$_javassist_50.
Push to 1.4
This bug was previously known as http://jira.rhq-project.org/browse/RHQ-2271
Temporarily adding the keyword "SubBug" so we can be sure we have accounted for all the bugs. keyword: new = Tracking + FutureFeature + SubBug
making sure we're not missing any bugs in rhq_triage
It's possible this is related to the SLSB leak issue that was fixed recently (https://bugzilla.redhat.com/show_bug.cgi?id=693232).
This whole mechanism has been reworked since this bug was reported.