Skip to content

Monitoring

tuppr exposes Prometheus metrics under the tuppr_ prefix. Hit /metrics on the controller pod for the authoritative list (keeping a copy here would drift):

kubectl port-forward -n system-upgrade deployment/tuppr 8081:8081
curl -s http://localhost:8081/metrics | grep tuppr_

The chart wires up the standard observability stack on demand:

  • monitoring.serviceMonitor.enabled: true - a ServiceMonitor for Prometheus Operator scraping.
  • monitoring.prometheusRule.enabled: true - a bundled PrometheusRule with stuck-upgrade, failed-upgrade, and operator-absent alerts.
  • monitoring.dashboards.enabled: true - a Grafana dashboard ConfigMap (sidecar-discoverable). Set monitoring.dashboards.grafanaOperator.enabled: true with matchLabels to also render a GrafanaDashboard CR for grafana-operator.

All three are off by default. See the Helm chart values for the full set.