Tempo Coordinator K8s

Tempo is a distributed tracing backend by Grafana.

Tempo Coordinator K8s is a Juju charm that manages the configuration, routing, and external integrations for a horizontally scalable Grafana Tempo deployment on Kubernetes. It works together with the Tempo Worker charm, following the coordinated-workers pattern. It is a component of the Canonical Observability Stack (COS): https://documentation.ubuntu.com/observability

The coordinator handles all external relations — such as trace ingestion, Grafana datasources, dashboards, and metrics — and distributes the generated Tempo configuration to the worker units. An nginx reverse proxy routes incoming traffic to the appropriate workers based on their assigned roles.

Key features:

• Centralized management of all external integrations, including trace ingestion, Grafana, Prometheus, Loki, and Alertmanager.
• Multi-protocol trace ingestion supporting OpenTelemetry (OTLP gRPC/HTTP), Jaeger, and Zipkin.
• Automatic configuration distribution to Tempo Worker units via the tempo-cluster relation.
• Built-in nginx reverse proxy for load balancing and traffic routing across workers.
• S3-compatible object storage integration for scalable trace data persistence.
• Built-in Grafana dashboards and Prometheus metrics endpoint for self-monitoring.
• Configurable trace retention period and per-protocol receiver toggles.
• TLS support via the certificates relation.
• Ingress support for cross-model trace collection through Traefik or Istio.
• Metrics generation from traces via the metrics-generator role, with Prometheus remote write support.