These alerts monitor the Kubernetes Job created by the Flectra backup CronJob.

Alerts:

• `FlectraBackupSucceeded` → Backup Job succeeded; PostgreSQL database, filestore, and extra-addons have been archived and uploaded to MinIO.
• `FlectraBackupFailed` → Backup Job failed; backup was not created or upload to MinIO failed.

* Success → Backup completed successfully. Data is safely stored in MinIO. * Failure → Flectra database or files may not be backed up. Could affect disaster recovery if restoration is needed.

1. Check Kubernetes Job status:

kubectl get job flectra-backup-job -n <NAMESPACE>
kubectl describe job flectra-backup-job -n <NAMESPACE>

2. Check logs of the Job pod:

kubectl logs job/flectra-backup-job -n <NAMESPACE>

3. Verify backup in MinIO:

mc ls <MINIO_ALIAS>/flectra-backups/
mc stat <MINIO_ALIAS>/flectra-backups/<backup_file>

4. Check PVC mounts if used:

kubectl get pvc -n <NAMESPACE>
kubectl describe pvc <PVC_NAME> -n <NAMESPACE>

* Pod in CrashLoopBackOff, OOMKilled, or Failed * PVC mount unavailable or insufficient space * MinIO credentials missing or misconfigured * Network issues preventing upload to MinIO * Disk space or permissions issues on the node * CronJob manifest misconfiguration

1. Inspect Job pod logs to identify errors. 2. Verify MinIO credentials and connectivity. 3. Check PVC status and node disk availability. 4. Retry backup manually if needed:

kubectl create job --from=cronjob/flectra-backup-job flectra-backup-job-manual -n <NAMESPACE>

5. Correct any misconfigurations in CronJob YAML or MinIO bucket policy. 6. Escalate to SRE or admin team if repeated failures occur.

* Escalate if backups fail for more than one consecutive run. * Notify on-call engineer if production Flectra data may not be recoverable.

* FlectraBackupSucceeded * FlectraBackupFailed * HostOutOfDiskSpace (node running backup Job) * KubernetesPodCrashLooping

* Kubernetes → Jobs & CronJobs (namespace: <NAMESPACE>) * Grafana → Backup Job status metrics * MinIO → Backup object listings