content/configuration/metrics.md
To enable performance and error measurement of connected services, Directus can provide Prometheus metrics.
| Variable | Description | Default Value |
|---|---|---|
METRICS_ENABLED | Whether or not to enable metrics. | false |
METRICS_SCHEDULE | The cron schedule at which to generate the metrics, the default is every minute | */1 * * * * |
METRICS_TOKENS | A CSV of tokens to allow access to via a Authorization: Metrics <token> header. By default it is restricted to admins | -- |
METRICS_SERVICES | A CSV of directus services to observe metrics for. Currently database, cache, redis and storage are supported | database,cache,redis,storage |
::callout{icon="material-symbols:warning-rounded" color="warning"} Metric Aggregation If Directus is running within a PM2 context, then metrics will be aggregated on a per scheduled job frequency. Ensure Prometheus' scrape frequency takes that into account. ::