Loading...

Type: Bug
Resolution: Done
Priority: Critical
Fix Version/s: 1.14.1
Affects Version/s: 1.14.0
Component/s: Operator
Labels:
- customer-facing
- qe-manual

Story Points:
5
Blocked:
False
Blocked Reason:
None
Ready:
False
Release Note Text:

Hide
Before this update, resource limits were set on the operator container as a recommended practice. However, these limits caused functional issues on clusters with a large number of secrets and configmaps due to the operator requiring more memory than allowed by these limits. This update removes the resource limits for the manager container to minimize the impact on functionality. Ongoing efforts to optimize memory consumption are planned for future releases.

Show
Before this update, resource limits were set on the operator container as a recommended practice. However, these limits caused functional issues on clusters with a large number of secrets and configmaps due to the operator requiring more memory than allowed by these limits. This update removes the resource limits for the manager container to minimize the impact on functionality. Ongoing efforts to optimize memory consumption are planned for future releases.
Git Pull Request:
https://github.com/redhat-developer/gitops-operator/pull/783
Intelligence Requested:
Market:
Target Version:

1.14.1

Sprint:
GitOps Crimson - Sprint 3264

SFDC Cases Counter:
SFDC Cases Open:
SFDC Cases Links:

Description of Problem:

- High memory utilization of `openshift-gitops-operator-controller-manager` pod has been observed post upgrade to operator v 1.14.0

Additional Info:

The "manager" container in the openshift-gitops-operator-controller-manager pod goes into CrashLoopBackOff state and have multiple restart count.

Prerequisites/Environment:

openshift-gitops-operator v 1.14.0

Following is the example of the status of the manager container:

Containers:
  manager:
    Container ID:  cri-o:// 
    Image:         registry.redhat.io/openshift-gitops-1/gitops-rhel8-operator@sha256:
    Image ID:      registry.redhat.io/openshift-gitops-1/gitops-rhel8-operator@sha256: 
    Port:          9443/TCP
    Host Port:     0/TCP
    Command:
      /usr/local/bin/manager
    Args:
      --health-probe-bind-address=:8081
      --metrics-bind-address=127.0.0.1:8080
      --leader-elect
    State:          Waiting
      Reason:       CrashLoopBackOff
    Last State:     Terminated
      Reason:       OOMKilled
      Exit Code:    137

Expected Results:

The containers under "openshift-gitops-operator-controller-manager" pod should run successfully without modifying/updating the memory limit.

Actual Results:

The manager container in "openshift-gitops-operator-controller-manager" pod goes into CrashLoopBackOff state.

Workaround (If Possible)

Currently increasing the memory from the CSV helps bring the "openshift-gitops-operator-controller-manager" pod and its containers into optimal state.