Workload Automation agent failed to start. Error message included unable to create new native thread

Document ID : KB000044297
Last Modified Date : 14/02/2018
Show Technical Document Details


The Workload Automation AE agent fails to start.

The following message is seen in the logs:


06/21/2016 09:28:08.895 EDT-0400 1 main.MainThread.CybFilebrowserPluginDriver.runApplication[:95] - CybFilebrowserPluginDriver terminated:

 java.lang.OutOfMemoryError: unable to create new native thread

 at java.lang.Thread.start0(Unknown Source)

 at java.lang.Thread.start(Unknown Source)

 at cybermation.plugins.filebrowser.CybFilebrowserPluginDriver.runApplication(

 at sun.reflect.NativeMethodAccessorImpl.invoke0(Unknown Source)

 at sun.reflect.NativeMethodAccessorImpl.invoke(Unknown Source)

 at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown Source)

 at java.lang.reflect.Method.invoke(Unknown Source)

 at cybermation.agent.plugins.manager.CybPluginManager.invokePluginMethod(

 at cybermation.agent.plugins.manager.CybPluginManager.loadPlugins(

 at cybermation.agent.CybAgentDriver.runApplication(

 at cybermation.agent.CybAgentDriver.main(



Workload Automation AE 11.3.x

Platform - ANY




Examine the system resources to see if the message is accurate.  Available memory or process space might be exhausted. Additionally look at the other agent log files.  In some cases the agent is attempting to restart a large amount of jobs. Make sure that too is accurate.  If the agent is attempting to restart jobs it should not be then attempt a cold start of the agent.


To perform a cold start of the agent remove the contents of the agent's database directory and start the agent.

 The database directory contains a list of the jobs / work the agent is doing. 

 These files are in a binary type database.

 When the agent restarts it would read this to see what it needs to restart or make sure is still running.

Additionally manual action will need to be taken on the jobs that were impacted by the cold start.

For a listing of the jobs that AE had running on the host run the following command: autorep -m <machine> -d