How to retrieve Metrics
With the prometheus feature enabled, the Enterprise Stack serves Prometheus metrics on a separate port (default
3002, path /metrics, configured in prometheus.conf). The
endpoint has no authentication: expose it only to your monitoring.
Available Metrics
| Metric | Labels | Meaning |
|---|---|---|
enterprise_usage_total | type | The counted operations, see What is counted |
event_issuance | action, status, format, credentialConfigurationId, proofType | Issuance events |
event_verification | action, status, format, signatureAlgorithm, credentialType | Verification events |
issuer2_openid4vci_requests_total | endpoint, result | Token and credential endpoint responses |
verifier2_openid4vp_requests_total | endpoint, result | Request object and response endpoint responses |
enterprise_license_* | License expiry, restriction state, heartbeat | |
jvm_* | The JVM |
Metrics are counted per pod and for the whole installation. They have no labels per organization, tenant, service or account: each value of such a label is a series that is never dropped, so their number would grow with every service and account. Numbers per organization, tenant or service come from the usage API.
event_key and event_did are no longer exported, and event_issuance and event_verification no longer have the
target and originator labels.
Counting events
POST /v1/events/query does not aggregate: a groupBy parameter is rejected with 400. To count operations, use the
usage API, which counts them per hour, day or month and per tenant, service and type.