If you see that there is a crashing kommander-ui pod, and you describe the pod, you may notice that the container is exiting with code 137. Code 137 almost always means that there is a lack of memory leading to the pod's demise. But what to do about it? Editing the pod or deployment wont work for very long, you need something more resilient. It is easy to modify the resources of them via cluster.yaml, but you may not know exactly what is required and where. You can use the below as a reference to set new limits and requests for the various kommander-ui pods in konvoy:
ops-portal-kommander-ui:
kommander-kubeaddons-kommander-ui:
dispatch-kommander-ui:
ops-portal-kommander-ui:
addonsList:
- name: opsportal
enabled: true
values: |2
kommander-ui:
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: 100m
memory: 512Mi
kommander-kubeaddons-kommander-ui:
addonsList:
- name: kommander
enabled: true
values: |2
kommander-ui:
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: 100m
memory: 512Mi
dispatch-kommander-ui:
addonsList:
- name: dispatch
enabled: false
values: |2
kommander-ui:
image:
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: 100m
memory: 512Mi