NetBackup Snapshot Manager auto-registration fails after successful Cloud Scale deployment
NetBackup Snapshot Manager deploys successfully in a Cloud Scale environment, but does not auto-register with the NetBackup primary server.
On a fresh NetBackup Snapshot Managerdeployment, all Kubernetes resources may appear healthy. The environment and cpServer custom resources show a state, and all flexsnap pods are in state. However, the Snapshot Manager does not appear in the NetBackup primary server's list of registered Snapshot Managers.
This issue occurs when one or more flexsnap container images take longer than expected to pull during the first deployment. If the image pull exceeds the internal 10‑minute health‑check window, the operator marks the deployment as failed before reaching the Snapshot Manager registration step. As a result, registration is skipped even though the deployment completes successfully.
Run the following commands to confirm the issue. If one of the following command displays the failure indicators, the deployment is affected.
Check deployment configmap status:
kubectl get cm flexsnap-deployment-conf-<cpserver-name> -n <env-namespace> -o yaml | grep -i status
Failure indicator:
STATUS: Failed
Check cpServer registration annotations:
kubectl get cpserver <cpserver-name> -n <env-namespace> -o jsonpath='{.metadata.annotations}{"\n"}'
Failure indicator:
<empty output>
Workaround:
Remove and recreate the cpServer configuration by updating the environment custom resource. This forces the operator to re-run the full install workflow. Because the container images are already cached from the first attempt, the second run completes within the health‑check window and successfully performs auto-registration.
Back up the environment configuration:
Save a copy of the environment custom resource using the following command:
kubectl get environment <environment-name> -n <env-namespace> -o yaml > environment-backup.yaml
Verify that the flexsnap operator is running using the following command:
kubectl get pods -n netbackup-operator-system | grep flexsnap-operator
Remove the cpServer specification:
Edit the environment custom resource using the following command:
kubectl edit environment <environment-name> -n <env-namespace>
Delete the entire
spec.cpserversection. Save and exit.Wait for cpServer cleanup to complete.
The operator will run the uninstall path on the cpServer. This scales down and removes
flexsnap-*resources.Re-add the cpServer specification:
Edit the environment custom resource again using the following command:
kubectl edit environment <environment-name> -n <env-namespace>
Reinsert the
spec.cpserversection using the values fromenvironment-backup.yamlfile. Save and exit.Wait for cpServer reinstall to complete and verify that cpServer state is Success:
kubectl get cpserver -n <env-namespace>
Verification:
After redeployment completes, confirm that Snapshot Manager registration is successful.
Verify deployment configmap status:
kubectl get cm flexsnap-deployment-conf-<cpserver-name> -n <env-namespace> -o yaml | grep -i status
Expected output:
STATUS: Success
Verify cpServer registration annotations:
kubectl get cpserver <cpserver-name> -n <env-namespace> -o jsonpath='{.metadata.annotations}{"\n"}'
Expected output:
netbackup.veritas.com/registration-status: Success