Repeated, rapid connections of authenticated clients to the Fuse message broker causes increasingly long connection times and high garbage collection overheads

Solution Verified - Updated -

Issue

When a JMS client makes rapid, authenticated connections to an Active MQ message broker running within Fuse 6.1, an increasingly long connection time may be noted over a period of time (hours-days), accompanied by frequent, lengthy garbage collection operations. A JVM heap histogram (obtained, for example, using jmap -histo) will typically show a large and increasing number of objects of class org.apache.karaf.jaas.modules.properties.PropertiesInstaller and/or org.apache.felix.framework.ServiceRegistrationImpl.

This problem does not usually arise in systems that use connection pooling for JMS connections, because the rate of connection creation is insufficient to stress the JVM heap. It should not arise at all when running ActiveMQ outside of the Fuse 6.1 OSGi container.

Environment

  • Red Hat JBoss Fuse
    • 6.1.0 GA

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.

Current Customers and Partners

Log in for full access

Log In