The fuller telemetry behind the
Ops & Security board — traffic, engagement, storage and trends. Counters are since the last API restart.
Usage, performance, the server, Postgres, Redis and the alerts firing, from Prometheus (15 days kept), behind Grafana’s own login.
The password is read from the server’s .env (GRAFANA_PASSWORD) only when you press Show or Copy: owners and admins only, and each view goes in the audit log. It is never in this page’s source, which anyone can download. If you change it inside Grafana, this keeps showing the old one until GRAFANA_PASSWORD is updated to match.
What's actually consuming CPU, memory and disk — not just the load-average indicator. Snapshotted by a host cron every 2 minutes.
Containers — by CPU
| Container | CPU | Memory | Net I/O | Block I/O | PIDs |
|---|
Top host processes — by CPU
Disk — by directory
Request latency distribution (cumulative buckets, seconds)
Logins per day — success / failed
Largest tables by total size (data + indexes + toast).
ChessMoves AI