External Monitoring Interface (EMI) stuck in reconnection loop after Database restart

Products

• CA Automic Workload Automation - Automation Engine

Issue/Introduction

On an AE system connected to a DB Cluster where EMI is running, as soon as the Database is failover to another member of the Cluster the EMI does not reconnect properly and starts opening a lot of connections to the JCP (until reaching huge amount of connections such as 2000 or more).

The kind of messages that appear in the emi log are the following:

20240226/111016.977 - U00009005 MBean collection 'Agents' cannot be created, reason: 'Automic:type=AUTOMIC,name=Agents'.
20240226/111016.977 - U00009005 MBean collection 'AgentGroups' cannot be created, reason: 'Automic:type=AUTOMIC,name=AgentGroups'.
20240226/111016.977 - U00009005 MBean collection 'Queues' cannot be created, reason: 'Automic:type=AUTOMIC,name=Queues'.
20240226/111016.977 - U00009005 MBean collection 'Usage' cannot be created, reason: 'Automic:type=AUTOMIC,name=Usage'.
20240226/111016.977 - U00009005 MBean collection 'Servers' cannot be created, reason: 'Automic:type=AUTOMIC,name=Servers'.
20240226/111016.977 - U00009005 MBean collection 'Cache' cannot be created, reason: 'Automic:type=AUTOMIC,name=Cache'.
20240226/111016.977 - U00009005 MBean collection 'Client' cannot be created, reason: 'Automic:type=AUTOMIC,name=Client'.
20240226/111016.977 - U00009005 MBean collection 'Users' cannot be created, reason: 'Automic:type=AUTOMIC,name=Users'.
20240226/111016.977 - U00009005 MBean collection 'Database' cannot be created, reason: 'Automic:type=AUTOMIC,name=MQ'.
20240226/111016.977 - U00009005 MBean collection 'Tasks' cannot be created, reason: 'Automic:type=AUTOMIC,name=Tasks'.

Environment

Automic Automation 21.x and 24.x where EMI has been connected to.

Cause

Defect on the Automation Engine.

Resolution

Update to a fix version listed below or a newer version if available.

Fix version:
Component(s): Automation Engine
Automation.Engine 21.0.10 - Available
Automation.Engine 24.0.0 - Available

Additional Information

Defect ID: AE-35800

Public Description: A problem has been fixed where the EMI got stuck in a loop trying to reconnect to the JCP over and over again resulting in a drastic load increase on the JCP Process.