Metrics
The server exposes Prometheus metrics at /metrics (exact path).
Queue metrics
Section titled “Queue metrics”One series per durable queue, tagged queue="telegram", queue="redmine" or — when the
deploy review webhook is enabled —
queue="deploy-review":
| Metric | Type | Meaning |
|---|---|---|
deploy_queue_pending |
gauge | Operations spooled and not yet sent. |
deploy_queue_dead |
gauge | Operations currently in the dead/ sub-directory. |
deploy_queue_sent_total |
counter | Operations successfully sent since start. |
deploy_queue_deadlettered_total |
counter | Operations moved to dead-letter since start. |
deploy_queue_send_latency |
timer histogram | Latency of a successful send/edit call (p50/p95/p99). |
These are the same numbers shown on the dashboard’s Delivery card.
JVM & process metrics
Section titled “JVM & process metrics”Standard Micrometer bindings are also registered: JVM memory, garbage collection, threads, and process (CPU, file descriptors, uptime) metrics.