Device down

A device is “down” when it stops sending telemetry (status flips to offline; no samples within a few intervals). Heating is not necessarily affected: the controller runs the control loop and safety interlocks locally and holds heat on last known good config even with no cloud link. Triage the connectivity, not the heat, first.

1. Confirm scope

• One device or many? Many at once → likely a backend/ingest issue (incident.md), not the field.
• Operator app → device page: last_seen_at, status, last telemetry.

2. Likely causes (single device)

• Cellular/SIM: check the SIM in Integrations — suspended or over data cap? Re-activate / failover (POST /integrations/sims/{id}/activate). Check signal at site.
• Power: site power loss. The agent comes back up in safe state on restart (systemd Restart=always); heat stayed available via native controls meanwhile.
• Agent crash: systemd restarts it safe. Pull diagnostics/logs from the device.

3. Verify heat (if reported cold)

• This is a field/HVAC issue, not a platform one. The interposing relay de-energizes to the safe state (boiler reverts to its own aquastat). Dispatch to confirm the boiler’s native controls are calling for heat; check the relay wiring and the aquastat.

4. Recover

• On reconnect the agent drains its local buffer; telemetry backfills. Confirm samples resume and status returns to online. Acknowledge any alerts raised during the outage.