Monitoring

System Check

License & System → System Check reports the health of everything OpenCelium depends on.

Service	Notes
OpenCelium	The application itself; the Info column shows the running version.
MariaDB	Users, groups, connectors, schedules, license tracking.
MongoDB	The workflow documents.
Email	Performs a real SMTP connection test. Down until spring.mail is configured and reachable.
Polyglot	The polyglot-engine service for Python and Ruby enhancements. Down unless you deploy it — harmless if you do not use those languages.
Operating System	Host information.

Each row reports Operational, Down or Unknown, plus an info and an error column, so a failing component names its own reason.

The command palette shortcut is check system. The underlying endpoint is /actuator/health, which requires the Admin authority.

Dashboard

The dashboard is the operational overview. Metrics stream over the WebSocket, so values update without reloading; each tile carries a small dot reporting the state of that connection (Live, Connecting, Connection lost — data may be stale).

• Tiles — Executions, Failure rate, Avg runtime, Run time, Logs. A tile with no data shows — rather than a misleading zero.

• Executions & failures — per-day counts, seven days by default.

• Resource usage — host CPU and memory.

• Top workflows — highest all-time execution count with failure rate.

Note

Attention required, Recent activity and System health are laid out but marked Coming soon. They are placeholders in 5.0 and do not show live data.

If the dot says the connection was lost, the usual cause is a reverse proxy that does not forward /ws — see Configuration.

Logs on the host

journalctl -xe -u opencelium -f

opencelium.debug_mode raises verbosity. Execution logs themselves are in the UI, not on disk — see Debug a workflow.

Netdata

A Netdata template for host-level dashboards ships in files/oc-mode.html. Deploy it next to your Netdata installation to mirror the example layout.