Restoring persistent volume claim data from Cloud Object Storage

You can copy the persistent volume claim data from Cloud Object Storage directly to the pod. Create a network policy to allow access and then create a job to provide the Cloud Object Storage credentials and restore the persistent volume claim data.

oc get deployment ${MAS_INSTANCE_ID}-service -n mas-${MAS_INSTANCE_ID}-visualinspection -o yaml | yq .spec.template.spec.securityContext.runAsUser

1. Create a network policy.
   In the Red Hat OpenShift web console, open the Import YAML page. Copy and paste the following NetworkPolicy content, replace the variables, and click Create.

   kind: NetworkPolicy
   apiVersion: networking.k8s.io/v1
   metadata:
     name: allow-vi-restore
     namespace: mas-{MAS_INSTANCE_ID}-visualinspection
   spec:
     podSelector:
       matchLabels:
         job-name: vi-restore
     egress:
       - {}
     policyTypes:
       - Egress

2. Create a Red Hat OpenShift job to restore the persistent claim volume data from Cloud Object Storage to a pod.
   In the Red Hat OpenShift web console, open the Import YAML page. Copy and paste the following Job content, replace the variables, and click Create.

   apiVersion: batch/v1
   kind: Job
   metadata:
     name: vi-restore
     namespace: mas-{MAS_INSTANCE_ID}-visualinspection
   spec:
     parallelism: 1    
     completions: 1    
     backoffLimit: 6   
     template:         
       metadata:
         name: vi-restore
       labels:
         app: vi-restore
       spec:
         serviceAccountName: ibm-mas-visualinspection-operator
         serviceAccount: ibm-mas-visualinspection-operator
         securityContext:
           runAsUser: {VI_USER}
           runAsGroup: {VI_USER}
           runAsNonRoot: true
         containers:
           - name: vi-restore
             image: rclone/rclone:1.62.2
             command:
               - sh
               - '-c'
               - >-
                 rm -rf /opt/powerai-vision/data/*; export RCLONE_CONFIG=/tmp/rclone.conf; rclone config create brcos s3 provider={S3_PROVIDER} endpoint={S3_URL} access_key_id={S3_ACCESS_KEY} secret_access_key={S3_SECRET_KEY} region={S3_REGION}; rclone --links --progress --no-check-certificate --config /tmp/rclone.conf copy brcos:{S3_BUCKET}/data /opt/powerai-vision/data; 
             volumeMounts:
               - name: data-mount
                 mountPath: /opt/powerai-vision/data
                 subPath: data
             securityContext:
               privileged: false
               readOnlyRootFilesystem: false
               allowPrivilegeEscalation: false
         restartPolicy: OnFailure    
         volumes:
           - name: data-mount
             persistentVolumeClaim:
               claimName: {MAS_INSTANCE_ID}-data-pvc

   The following values are used in the Job content:

   {MAS_INSTANCE_ID}

   The Maximo Application Suite instance ID.

   {S3_PROVIDER}

   The Amazon S3 storage provider that you are using. For more information, see Amazon S3 Storage Providers.

   {S3_URL}

   Endpoint for S3 API.

   {S3_ACCESS_KEY}

   Your Amazon Web Services access key ID.

   {S3_SECRET_KEY}

   Your Amazon Web Services secret access key or password.

   {S3_REGION}

   The region to connect to.

   {S3_BUCKET}

   The backup data is stored in {S3_BUCKET}/data.

   {VI_USER}

   The owner of the data folders.

3. In the Red Hat OpenShift web console, wait for the vi-restore job to be completed.
   You can check the log of this job for details of the progress.