JON agent remoting retention in Java heap
Issue
- JON agent does "soft restart" by restarting Plugin Container due to
VMHealthCheckThreadmonitoring seeing heap retention crossing the threshold (90% by default). - Heap dump analyis shows the following list of 20 biggest objects:
*******************************************************
org.jboss.remoting3.remote.RemoteConnectionChannel#4 253,449,509
org.jboss.remoting3.remote.IntIndexHashMap#15 253,440,648
org.jboss.remoting3.remote.IntIndexHashMap$Table#16 253,440,600
java.util.concurrent.atomic.AtomicReferenceArray#66 253,440,560
java.lang.Object[]#1100223 253,440,536
org.jboss.remoting3.remote.RemoteConnectionChannel#1 251,819,045
org.jboss.remoting3.remote.IntIndexHashMap#2 251,810,184
org.jboss.remoting3.remote.IntIndexHashMap$Table#8 251,810,136
java.util.concurrent.atomic.AtomicReferenceArray#18 251,810,096
java.lang.Object[]#1029684 251,810,072
org.jboss.remoting3.remote.RemoteConnectionChannel#3 245,609,789
org.jboss.remoting3.remote.IntIndexHashMap#9 245,600,904
org.jboss.remoting3.remote.IntIndexHashMap$Table#12 245,600,856
java.util.concurrent.atomic.AtomicReferenceArray#33 245,600,816
java.lang.Object[]#1059980 245,600,792
org.jboss.remoting3.remote.RemoteConnectionChannel#6 159,431,973
org.jboss.remoting3.remote.IntIndexHashMap#23 159,423,112
org.jboss.remoting3.remote.IntIndexHashMap$Table#24 159,423,064
java.util.concurrent.atomic.AtomicReferenceArray#84 159,423,024
java.lang.Object[]#1149196 159,423,000
Environment
- JON 3.x
- JBoss Enterprise Application Platform (EAP) 6.0.1
- Oracle JDK 1.7.0_75
- Collecting 2040 metrics per minute (and have 20674 metrics enabled) with 2GB heap
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
