I just recently started seeing issues with our Bamboo Remote Agent and plans being queued for a few minutes before running, even though no other plans are running and the agent is clear to build the queued plan. In the log for the remote agent, the following is printed almost every minute until eventually the agent picks up the plan and runs it like normal.
2016-04-25 09:14:10,987 WARN [ActiveMQ InactivityMonitor Worker] [FailoverTransport] Transport (https://xxx.atlassian.net:443/builds/jmsServlet) failed, attempting to automatically reconnect
org.apache.activemq.transport.InactivityIOException: Channel was inactive (no connection attempt made) for too (>30000) long: https://xxx.atlassian.net:443/builds/jmsServlet
at org.apache.activemq.transport.AbstractInactivityMonitor$1$1.run(AbstractInactivityMonitor.java:91)
at java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
at java.lang.Thread.run(Unknown Source
2016-04-25 09:14:26,624 INFO [ActiveMQ Task-48] [HttpClientTransport] Broker Servlet supports GZip compression.
I have tried to completely uninstall/re-install the remote agent, Java, etc. All plans seem to get run from the queue eventually, but they are not instantaneous like they used to be, and all of the plans being queued creates a backlog. Any suggestions are welcome.
Thanks.