Troubleshooting installation issues for IBM Spectrum Protect Plus as a set of containers

Review the available information to resolve installation issues with the IBM Spectrum® Protect Plus server in a container environment.

Table 1. Installation issues and potential resolutions for IBM Spectrum Protect Plus as a set of containers
Problem Symptom Resolution
Pods remain in the ContainerCreating state After the IBM Spectrum Protect Plus instance is created, any of the following pods remain in the ContainerCreating state:
  • virgo
  • nodejs
  • awsebs
  • awsec2
 Ensure that different zone label exists for each node by issuing the following command:
kubectl get nodes --show-labels | grep topology.kubernetes.io/zone
Set a zone label for each node if it is not already set by issuing the following command:
kubectl label nodes node_name topology.kubernetes.io/zone=label_value
where node_name is the name of a node in the OpenShift cluster, and label_value is the zone label for the node.
For example, to set the zone label for all nodes on the cluster:
n=0; for node in $(kubectl get nodes --output NAME); do kubectl label $node topology.kubernetes.io/zone="us-local-$n"; 
(n++)); done
The virgo pod is not ready The IBM Spectrum Protect Plus virgo pod is taking too long to reach a ready state, or the liveness or readiness probes are failing for more than an hour. Increase the CPU, memory, or both limits of the virgo pod. You can increase the limits by updating the spec.spp.virgo.limits field of the IBM Spectrum Protect Plus instance custom resource. The following values are the default limits for CPU and memory for the virgo pod:
limits:
  cpu: 2000m
  memory: 8Gi