KB-1070 Some engines are unable to start with a FATAL error in the checkengine script

Nick Vigilante — Wed, 22 Feb 2017 20:17:14 GMT

Current Revision posted to Appian Knowledge Base by Nick Vigilante on 2/22/2017 8:17:14 PM

Symptoms

One or more of the following symptoms may be present.

When attempting to start the engines, one or more engines fail to start. Running checkengine.bat|sh shows FATAL for these engines.

In the db_*.log files corresponding to the engines that failed to start, the following message is seen:

FATAL .a.db "bind: Only one usage of each socket address (protocol/network address/port) is normally permitted."

After stopping Appian, k processes are still running. This can be seen in Windows in the Task Manager and in Linux by running ps -ef | grep /k.
After running stop-suite, the command netstat -vatn shows that one or more of the ports needed by the engines are in use.

Cause

There are other processes running on the ports needed by the engines. The ports used by the engines are assigned in the appian-topology.xml file, which is located in <APPIAN_HOME>/ear/suite.ear/conf. For Appian versions 7.8 and later, this file will also be located in <APPIAN_HOME>/search-server/conf. Changes to the topology should be made in both locations. If there are other processes running on the ports specified by the topology, the engines will be unable to start.

Action

Stop Appian (see Starting and Stopping Appian).
Stop or kill any processes that are running on the ports reserved for the engines.
Start the engines.
Run checkengine.bat|sh to confirm that the engines have successfully started.

Affected Versions

This article applies to all versions of Appian.

Last Reviewed: February 2017

Tags: administration, engines