Issue: The ActiveMQ Disk is Full

Issue

If you're seeing errors in your log file related to a queue, it's likely that the storage of your ActiveMQ database is causing the problem. You might observe any of the following in your contrast.log file:

• "Failed to page in more queue" messages
• queue://queue.activity.app
• Queue.java
• A link to http://activemq.apache.org/producer-flow-control.html
• Any reference to ActiveMQ with ERROR log level
• "Failed to page in more queue messages" java.lang.NegativeArraySizeException thrown by activemq.broker
• "Failed to fill batch" caused by NegativeArraySizeException.
• "Problem retrieving message for browse" thrown by Queue.java:1117

When the disk first fills up, you might start to see error messages in contrast.log every 30 seconds. Example error message:

Main:store:queue://queue/activity.app:store) ... percentUsage=102% ... Persistent store is Full, 100% of SOME#. Stopping producer

Cause

At this point, you should check to see why Contrast thinks that your disk is full. It's likely that ActiveMQ surpassed its allotted storage; but, in some cases, your disk could be full.

Resolution

Before clearing any files, stop the Contrast process, and then begin the process of clearing out unnecessary files. If you see these ActiveMQ errors, you can resolve this in just a few steps:

• Confirm the Contrast isn't running.
• Navigate to $CONTRAST_HOME/data/activemq
• Delete everything in this folder (both the data and index folders).
• Start Contrast.