Skip to content
Docs

Monitoring

The admin UI shows requests, errors, latency, the busiest routes, recent server errors, database and file sizes, realtime connections, and CPU, memory and disk. Inside a container with memory or CPU limits it shows the container’s limits, not the host’s. It warns when the data directory’s disk has less than 10% free.

The dashboard of the admin UI.The dashboard of the admin UI.

Every request except health checks and admin UI assets goes to logs.db, kept for logs.retention_days (default 7). logs.min_status keeps only errors, logs.sample_rate samples successful requests. Losing logs.db loses no data.

The request logs in the admin UI.The request logs in the admin UI.

The audit log (Audit in the admin UI) records, among others, sign-ins and sign-in failures, password and email changes, revoked sessions, changes to settings, superusers, channels and backups, and the encryption events secrets.rekey and secrets.reset.

The audit log in the admin UI.The audit log in the admin UI.

GET /metrics is off until you set metrics.prometheus_token in the settings (send it as Authorization: Bearer <token>) or metrics.prometheus_public: true.

Set RACKEND_OTLP_ENDPOINT to export traces (one span per request with database spans below it) and metrics over OTLP/HTTP.

rackend logs JSON lines to stdout; every line inside a request carries its request_id, also returned as the X-Request-Id header.