zeebe-backup-job.yaml

apiVersion: batch/v1
kind: Job
metadata:
  name: zeebe-backup-job
spec:
  template:
    spec:
      containers:
      - name: zeebe-backup-script
        image: curlimages/curl
        env:
          - name: BACKUP_TIME_ID
            valueFrom:
              secretKeyRef:
                name: backup-timeid
                key: backupTimeId
        command: ["/bin/sh"]
        args:
        - -c
        - |
          # Trigger the backup
          curl -X POST "http://camunda-zeebe-gateway:9600/actuator/backups" \
            -H 'Content-Type: application/json' \
            -d "{\"backupId\": \"$BACKUP_TIME_ID\"}"
          echo "Backup trigger request sent for ID $BACKUP_TIME_ID"
          sleep 5

          # Monitor backup status
          fail_count=0
          while true; do
            response=$(curl -s "http://camunda-zeebe-gateway:9600/actuator/backups/$BACKUP_TIME_ID")
            echo "$response"
            
            # Check if the response includes the specific backupId and state COMPLETED
            if echo "$response" | grep -q "\"backupId\":$BACKUP_TIME_ID,\"state\":\"COMPLETED\""; then
              echo "Backup $BACKUP_TIME_ID completed successfully."
              break
            elif echo "$response" | grep -q "\"state\":\"IN_PROGRESS\""; then
              echo "Backup $BACKUP_TIME_ID in progress, checking again in 30s."
              sleep 30
            else
              echo "Backup $BACKUP_TIME_ID failed or in an unknown state."
              fail_count=$((fail_count+1))
              if [ $fail_count -ge 3 ]; then
                echo "Failed 3 times, exiting."
                exit 1
              else
                echo "Retrying, attempt $fail_count of 3."
                sleep 30
              fi
            fi
          done
      restartPolicy: Never