VCF Automation upgrade from 9.0.x to 9.1: precheck failed due to logging-operator-filelogs-fluentbit pod in crash loop.

Products

• VCF Automation

Issue/Introduction

 Using this command, the status of logging-operator-filelogs-fluentbit pod(s) goes from Running to OOMKilled to CrashLoopBackOff:

kubectl get pods -A | grep logging-operator-filelogs-fluentbit

Environment

VCF Automation 9.0.x upgrading to 9.1

Resolution

Steps for VCFA 9.0.x to persist a memory increase throughout the upgrade:

1. Login to VCFA 9.0.x node on SSH
2. kubectl -n vmsp-platform edit rt logging-operator-4.8.0-35
3. In the logging: section, directly under fluentbit: (above annotations:), add this text, being sure to use correct indentation with spaces, no tabs. resources:  being one level of indentation below fluentbit:
   • resources:
  limits:
    memory: 256Mi
4. Then return to the upgrade and click RETRY