NetBackup™ Deployment Guide for Amazon Elastic Kubernetes Services (EKS) Cluster

Upgrading the NetBackup application

1. To upgrade the primary server and media server, edit the environment.yaml from the new package. Copy all the fields from the preserved environment CR environment.yaml which can be obtained from the the following section:

   See Preparing for NetBackup upgrade.

   To update the primary server, update tag with new image tag in primary section in new environment.yaml file.

   To update the media server, update tag with new image tag in mediaServers section in new environment.yaml file.

   Update the storageClassName for catalog volume for primary server in Storage subsection of primary section in environment.yaml file.

   For example,

   Kind: Environment
   ...
   Spec:
    primary:
     tag: "newtag"
    mediaServers:
      tag: "newtag"
   

   Apply the changes using the following command:

   kubectl apply -f <environment.yaml>

   Primary server and media server pods would start with new container images respectively.

   Note:

   Upgrade the PrimaryServer first and then change the tag for MediaServer to upgrade. If this sequence is not followed then deployment may go into inconsistent state

   Note:

   MediaServer version should be same or lower than the PrimaryServer version after upgrade. Otherwise the deployment may go into inconsistent state.

2. At the time of upgrade, primary server and media server status would be changed to Running. Once upgrade is completed, the status would be changed to Success again.

Perform the following if upgrade fails in between for primary server or media server

1. Check the installation logs using the following command:

   kubectl logs <PrimaryServer-pod-name/MediaServer-pod-name> -n <PrimaryServer/MediaServer-CR-namespace>

2. If required, check the NetBackup logs by performing exec into the pod using the following command:

   kubectl exec -it -n <PrimaryServer/MediaServer-CR-namespace> <PrimaryServer/MediaServer-pod-name> -- bash

3. Fix the issue and restart the pod by deleting the respective pod with the following command:

   kubectl delete < PrimaryServer/MediaServer-pod-name > -n <PrimaryServer/MediaServer-CR-namespace>

4. New pod would be created and upgrade process will be restarted for the respective NetBackup server.
5. Data migration jobs create the pods that run before deployment of primary server. Data migration pod exist after migration for one hour only if data migration job failed. The logs for data migration execution can be checked using the following command:

   kubectl logs <migration-pod-name> -n <netbackup-environment-namespace>

   User can copy the logs to retain them even after job pod deletion using the following command:

   kubectl logs <migration-pod-name> -n <netbackup-environment-namespace> > jobpod.log