Diagnose the unhealthy service
Work from the exact health-monitor message toward the first concrete error. Record the evidence before restarting the service so the original cause remains visible.
Open the service diagnosis
In Health monitor, select the affected service and open Diagnosis settings. Note the service name, status message, and the time of the last successful check. This identifies the component you need to inspect.
Retry the health check once
Wait a few seconds and run the check again. A service may be starting after a device reboot or update. If the result stays unhealthy, continue with the status and logs instead of repeatedly restarting it.
Check the service status
Open the terminal on the pyCANdo device and replace the placeholder with the exact service name shown by the diagnosis. Read the Active line and the first error displayed below it.
systemctl status <service-name> --no-pagerRead the most recent log entries
The journal usually explains whether the service could not read its configuration, open a port, access hardware, or start a dependency. The pyCANdo dashboard also exposes system and application logs where available.
journalctl -u <service-name> -n 100 --no-pager- For dashboard API problems, also inspect `/var/log/pcd/pcddash/`.
- For project runtime problems, inspect `/var/log/pcd/pyrbs/` and `/var/log/pcd/pyrbs_core/`.
- For update problems, inspect `/var/log/pcd/pcdup/pcdup.log`.
Check the named dependency
Follow the first concrete error. Check the relevant configuration or dependency, then run the health check again. Restart the service only after you have recorded the error and corrected its cause.