Primary pod goes in non-ready state
When jobs are running and if failover or upgrade of NetBackup primary server is triggered then primary pod may go in non-ready state. To confirm the issue,
Exec into primary pod by running the following command:
kubectl exec -it <primary_pod_name> -n <namespace> -- bash
Open the logs using the cat /mnt/nblogs/setup-server.log command and verify the last lines:
Softlink /mnt/nbdata/usr/openv/netbackup/db/rb.db does not exist or broken. Error: Issue detected while validating soft links, restarting the pod may fix this. NetBackup health check disabled.
To resolve this issue, delete the corrupted database and correct symlink as follows:
Exec into primary pod by running the following command:
kubectl exec -it <primary_pod_name> -n <namespace> - bash
In primary pod run the following commands in order:
# /opt/veritas/vxapp-manage/nb-health disable # bp.kill_all # mv -f /mnt/nbdata/usr/openv/netbackup/db/rb.db /mnt/nbdb/usr/openv/netbackup/db/rb.db # ln -sf /mnt/nbdb/usr/openv/netbackup/db/rb.db /mnt/nbdata/usr/openv/netbackup/db/rb.db # chown -h nbsvcusr:nbsvcusr /mnt/nbdata/usr/openv/netbackup/db/rb.db # bp.start_all # /opt/veritas/vxapp-manage/nb-health enable