New ActiveMQ installation goes out of memory after 30 minutes

We have a fresh installation of ActiveMQ 5.9.1 running on Red Hat Linux. Without external connections, without queues and only by default in the system, the process ends from memory (1 GB allocated when starting with "-Xms1G -Xmx1G") after about 30 minutes, even with absolutely no activity. I first ran into this issue with version 5.10.0 and downgraded it to 5.9.1 to see if it might have popped up in a new build.

Literally all I did was:

tar xzf apache-activemq-5.9.1-bin.tar.gz
mv apache-activemq-5.9.1-bin activemq
cd activemq
bin/activemq start

      

Using the "top" I noted that it started out using about 150MB of real memory and it continued to crawl up. Once the top showed it with 1.1GB, there were several heapdump, core, javacore and trace files in the base directory. In javacore file all state:

Dump Event "systhrow" (00040000) Detail "java/lang/OutOfMemoryError" "Java heap space" received

      

Has anyone else encountered this? How did you fix this?

UPDATE 2014-08-22

"java -version" yeilds:

java version "1.7.0"
Java(TM) SE Runtime Environment (build pxa6470sr6fp1-20140108_01(SR6 FP1))
IBM J9 VM (build 2.6, JRE 1.7.0 Linux amd64-64 Compressed References 20140106_181350 (JIT enabled, AOT enabled)
J9VM - R26_Java726_SR6_20140106_1601_B181350
JIT  - r11.b05_20131003_47443.02
GC   - R26_Java726_SR6_20140106_1601_B181350_CMPRSS
J9CL - 20140106_181350)
JCL - 20140103_01 based on Oracle 7u51-b11

      

I am starting to think that the IBM JVM issue might be the problem.

EDIT 2014-08-29 Replaced the IBM JVM with the standard Oracle JVM and updated ActiveMQ to 5.10.0, but still has a problem. No connections to the server, one queue with no messages. Using the Eclipse memory analyzer, the suspicion report shows 197 instances of org.apache.activemq.broker.jmx.ManagedTransportConnection consuming approx. 500 MB of memory out of 529 MB in total. Not sure what this means or how to fix it.

+3


source to share





All Articles