PB-14482: Helm Chart Updates for Cluster Count Limit Alerting

[t-nidhin]

This PR adds the Prometheus alerting rules and email templates required for the global cluster count limit feature in PX-Backup. When the cluster count limit is reached, alerts are sent via AlertManager to notify administrators.

Related PR

• px-backup PR: PB-14482 (implements the backend logic and metrics emission)

New Prometheus Metric (from px-backup)

The alert is triggered by the pxbackup_cluster_count_limit_status metric emitted by px-backup:

pxbackup_cluster_count_limit_status{
  user_id="<user-uuid>",
  current_count="2",
  max_count="2",
  error_reason="Cluster count limit reached: 2/2 clusters",
  backfill=""
} 100

Email Alert Preview

When the cluster count limit is reached, administrators receive an email with:

Testing

Manual Testing

1. Applied PrometheusRule directly:

kubectl apply -f test-prometheusrule.yaml

2. Applied AlertManager secret directly:

kubectl apply -f test-alertmanager-secret.yaml
kubectl rollout restart statefulset -n px-backup -l app.kubernetes.io/name=alertmanager

3. Verified alert fires when metric is present in Prometheus
4. Verified email is received with correct formatting

Prometheus Alert Verification

kubectl port-forward svc/px-backup-dashboard-prometheus 9090:9090 -n px-backup
# Navigate to http://localhost:9090/alerts
# Verify ClusterCountLimitAlert is listed and firing

Backward Compatibility

• No breaking changes: Existing deployments will continue to work
• Alert only fires when metric exists: The alert only triggers when px-backup emits the pxbackup_cluster_count_limit_status metric (i.e., when the limit is reached)

[dbinnal-px]

You need to update the index.yaml before merging to point pxb-federated branch