Restarts for doi-topology-processor pod

Products

• DX Operational Intelligence

Issue/Introduction

The doi-topology-processor pod has restarted more then 100 times in the past two days.

Environment

Release : 21.3

Component : CA DOI Foundations - DEFAULT

Resolution

This is related to the pod configuration issue.

Here is the pod memory configuration:

    Limits:
      cpu:     1
      memory:  4Gi
    Requests:
      cpu:     200m
      memory:  2Gi

 

But the JVM settings of the doi-topology-processor has following settings :

      ACN_HEAP_MIN:          2g
      ACN_HEAP_MAX:         6g

The heap max is 6G which means the JVM memory may grow up to 6G which exceeds the pod limit of 4G. That will cause the OOMKilled.

We would suggest to increase the pod memory limit to 8G, and request to 7G so that the pod has enough memory to run the doi-topology-processor.