SA-18368

Summary

Agents stop running after a database disconnect. Agents scheduled for running intervals do not start again on the next interval.

The remainder of the system recovered after the disconnect - approximately 30 seconds, but Agents that encountered an error during the timeframe are reporting "Terminated Manually"

Error Messages

2015-12-02 08:12:46,757 [   PegaRULES-Batch-5] [  STANDARD] [                    ] [     PegaRULES:07.10] (     internal.async.AgentQueue) ERROR   - Agent [Test] activity [PegaSample-Task.CreateRequest] queue number [0] disabled due to execution errors ErrorMsg: Internal state is compromised; cannot write out application hierarchy and/or rulesets
 

Steps to Reproduce

Disconnect or failover database.

Root Cause

A defect in Pegasystems’ code or rules which is raised as BUG-224027.

Database connectivity issue should not affect the agent cycle, even though many application functionality might not be available.

Expected behavior is that agent would be still able to run without any manual intervention. 

 

Resolution

Perform the following local-change: 

Restart the application server, whenever there is a database connectivity or database restart. Coordinate application server restart along with database restart.