Under Settings → Maintenance the edge offers a set of repair actions. They are executed by the Maintenance Sidecar — a service with elevated privileges that exists exactly for this kind of intervention. The same actions are also available to Brinkhaus support remotely via the FleetManager; the on-site button and the remote command do exactly the same thing.
The list is state-dependent: actions that make no sense in the current state (e.g. “Emergency-stop data services” when they are already stopped) are hidden.
The actions
| Action | What happens | Typical reason |
|---|---|---|
| Reboot edge | Restarts the entire device (with a confirmation prompt). | General malfunction that a service restart does not fix. |
| Reset Grafana dashboards | Restores the Grafana dashboards and data sources shipped with the firmware. Previous states are backed up; dashboards marked as “locked” are left untouched. | Misconfigured or accidentally deleted dashboards. |
| Reset DataStore settings | Restores the DataStore configuration from the shipped template (the service is briefly stopped and restarted for this). | DataStore misconfiguration. |
| Reset IDEA4S-Bridge settings | Same as above, for the IDEA4S-Bridge. | Bridge misconfiguration. |
| Reset DataStore historic tables | Discards the historic measurement-data tables; they are recreated with the current sensor signal types on the next measurement. Data acquisition pauses in a controlled way during this — no messages are lost. | After a sensor hardware swap, or when a measurement was created with the wrong data type (e.g. rotational speed appears as text in Grafana). |
| Emergency-stop / start data services | Stops or starts PostgreSQL, DataStore, and IDEA4S-Bridge in the correct order. | Triage when database load gets out of control. |
Maintenance migrations in the background
On every firmware update the Maintenance Sidecar additionally runs one-time, version-gated maintenance migrations — e.g. deploying corrected Grafana templates onto devices already in the field. This happens automatically and without any user action; at most, you may notice a brief restart of the affected service.