Restore fails due to missing resources
Restoring a IBM® Software Hub deployment fails because resources are missing.
Symptoms
The POSTHOOK STATUS of the restore
cpd-tenant-restore-<id>-r4 is
partially_succeeded.
Environment
This problem can occur when you use the OADP backup and restore utility, NetApp Astra Control Center, or Portworx Disaster Recovery.
Diagnosing the problem
Run the following script:
Note: For information about how to run
cpd-cli from any directory, see Installing the IBM Software Hub command-line interface. restores=$(oc get restores.velero.io -n oadp-operator --no-headers | awk '{print $1}')
for restore in ${restores};do cpd-cli oadp restore logs ${restore} | grep "Skipping restore of resource because it cannot be resolved via discovery" | grep -v "klusterletconfigs.config.open-cluster-management.io" | grep -v "managedcluster.cluster.open-cluster-management.io" | grep -v "clusters.cluster.x-k8s.io" | grep -v "clusterresourcesets.addons.cluster.x-k8s.io" | grep -v "managedclusteraddon.addon.open-cluster-management.io" | grep -v "klusterletaddonconfig.agent.open-cluster-management.io" | grep -v "clusterclasses.cluster.x-k8s.io" | grep -v "clusterbootstraps.run.tanzu.vmware.com" ; done;The script returns error messages like in the following
example:
time="<timestamp>" level=info msg="Skipping restore of resource because it cannot be resolved via discovery"
logSource="/remote-source/velero/app/pkg/restore/restore.go:2188" resource=clusters.postgresql.k8s.enterprisedb.io restore=oadp-operator/cpd-tenant-restore-<id>-r2Resolving the problem
Delete the IBM Software Hub instance projects and restart the restore.