How to purge the Integration Agent queue

Problem

We identified a backlog in the Integration Agent: a high number of events in the ACTIVE queue for a particular integration service. 

Environment

  • Integration Agent version 5.1.8.2
  • BMC ProactiveNet Performance Manager integration

Resolution 

To resolve this issue, we took the following steps:

  1. Increased the logging, and then allowed the system to run for a bit under new logging to collect data:
    • Increased logging levels (APXML, HTTP functions)
    • Right-sized the number and size of the logs kept for each appender: 10 backup files, and 10MB each for Integration Agent and integration data logs
  2. Purged the Integration Agent queue:
    • Stopped Integration Agent services (<IAHOME>/bin/shutdown.sh)
    • Moved queue-related data out of the way.
    • Renamed .mule to old-mule
    • Renamed .activemq-data to old-activemq-data
    • Started Integration Agetn services.
    • Verified new events were being processed correctly.

Cause

This issue was caused by a break in communications between the Integration Agent and the xMatters On-Demand service, or possibly between the Integration Agent and the source system (if callbacks/annotation were enabled for the integration). The underlying cause of the communications break could be one of a number of separate issues.  

 

xMatters internal reference: DOC-7955

Have more questions? Submit a request

0 Comments

Please sign in to leave a comment.
Powered by Zendesk