Skip to main content

Monitoring

Goal​

Keep enough visibility on docs, web, API, workers, and infrastructure to spot failures early.

What exists in the repo​

  • persistent Log model in the current backend
  • GraphQL log subscription support
  • beta monorepo observability direction referencing OTel, Prometheus, Grafana, Loki, and Tempo

Minimum things to watch​

  • API availability
  • docs and web ingress reachability
  • Redis and Postgres availability
  • worker liveness
  • indexing lag
  • simulation failures

Verification​

  • inspect GraphQL log output
  • inspect Kubernetes rollout and pod status
  • inspect worker logs for adaption and task execution

Troubleshooting​

  • if plans update but leaderboards do not, inspect leaderboard worker state
  • if docs and landing work but app does not, inspect API and auth routing
  • if subscriptions feel stale, inspect websocket endpoint health