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- aServiceMonitorfor Prometheus Operator scraping.monitoring.prometheusRule.enabled: true- a bundledPrometheusRulewith stuck-upgrade, failed-upgrade, and operator-absent alerts.monitoring.dashboards.enabled: true- a Grafana dashboardConfigMap(sidecar-discoverable). Setmonitoring.dashboards.grafanaOperator.enabled: truewithmatchLabelsto also render aGrafanaDashboardCR for grafana-operator.
All three are off by default. See the Helm chart values for the full set.