⚡ Reduxi Configurator · User manual
Health check
See the problems the controller has detected with devices, control and itself, and what to do about each one.
📋 In this article
→ Overview
The controller checks its devices and itself every 10 seconds. When it finds a problem — a device that does not communicate, ignores its setpoint, reports implausible energy values, or is not under automatic control — it opens a health event. The event stays open while the problem lasts and closes by itself when the controller sees that the problem is gone.
The Health check page lists these events. Open events are the ones to act on; closed events show the history. While the cloud connection is up, events are also reported to the Reduxi Cloud as they open and close.
1 Review the health events
1
Open the page
In the left menu, click Settings, System, Diagnostics, then Health check.
2
Set the period
Click the date range at the top right. By default the page shows events that started in the last 7 days. Your choice is remembered in this browser.
3
Find open events
Events with a dash
- in Time end are still open. Sort by Time start or filter by Severity to see the important ones first.4
Open the details
Click the eye icon in Details to see the device, its type and the measured values behind the event.
5
Fix the cause
Use the table in Event codes to find what to check. After the fix, the event gets a Time end on its own.
Settings → System → Diagnostics → Health check with one open DEVICE_STATE_UNKNOWN warning for the Sonnen FlexStack battery
2 Parameters
Table
| Parameter | What it does | Unit / default |
|---|---|---|
| Date range | Period shown, by start time of the event. At most 500 events are loaded; if there are more, a red bar Showing … of … rows. Select a shorter time range to see all results. appears. | default last 7 days |
| Time start | When the problem was first detected. Newest first by default. | date and time |
| Time end | When the problem cleared. - = still open. |
date and time |
| Code | Type of problem, see Event codes. Filter field below the header. | — |
| Category | Area of the problem: COMMUNICATION, ADHERENCE, MEASUREMENT, CONTROL, STATE or SYSTEM. Filter field below the header. |
— |
| Severity |
WARNING (orange) or ERROR (red). While an event is open its severity can rise from WARNING to ERROR, but never falls back. |
— |
| Subject | What the event is about: DEVICE: <name> or CONTROLLER: <unit ID>. Filter field below the header. |
— |
| Details | Eye icon: opens the Details dialog. | — |
| Rows per page, page, Prev, Next | Paging of the list. | 20 rows |
Details dialog
Health check Details dialog for a closed DEVICE_CONTROL_NOT_IN_AUTO warning on the Etrel Inch charger with details MANUAL
| Parameter | What it does | Unit / default |
|---|---|---|
| Start, End | Time the event opened and closed (- = open). |
— |
| Code, Category, Severity | As in the table. | — |
| Subject | Kind, internal ID, name and device type, for example DEVICE #4 (Etrel Inch, CHARGER_ETREL). |
— |
| Details | Check-specific data, for example the control type (MANUAL), the signals in an unknown state, or the commanded and measured power with the deviation in %. |
text or JSON |
3 Event codes
| Code | Category | Severity | What it means | What to do |
|---|---|---|---|---|
DEVICE_COMM_ERROR |
COMMUNICATION | ERROR | The controller cannot communicate with the device. Closes as soon as communication works again. | Check power and cabling of the device, its IP address and port (or serial settings) and the Modbus unit ID. Test reachability with the TCP port test or Ping test. |
SETPOINT_NOT_FOLLOWED |
ADHERENCE | WARNING or ERROR | A controlled device does not do what the controller asks (checked for the device's main, battery, PV and EV charging power). After each setpoint change the device gets 20 s to settle. A deviation of 10–25 % is a WARNING; 25 % or more, the wrong direction (for example charging instead of discharging) or almost no output is an ERROR. With a setpoint below 500 W, an output of 500 W or more is a WARNING and 1000 W or more an ERROR. The event opens after 6 bad checks in a row (about 1 minute) and closes after 6 good checks in a row. | Check that the device accepts external control (remote/EMS mode enabled in the device), that its own limits (SOC limits, power limits, temperature derating) are not holding it back, and that the sign and scaling of its power readings are right. |
POWER_ENERGY_MISMATCH |
MEASUREMENT | WARNING or ERROR | Over 15 minutes, the energy calculated from the device's power readings does not match the change of its energy counter (checked separately for import and export). A difference of 10–50 % is a WARNING, 50 % or more an ERROR. Periods with less than 50 Wh are not judged. | Check the measurement settings of the device: current transformer ratio and direction, scaling, and that power and energy come from the same measuring point. |
DEVICE_CONTROL_NOT_IN_AUTO |
CONTROL | WARNING | A device that can be controlled has its control type set to something other than Auto (for example Manual or Off), so strategies do not control it. Details show the current control type. | If the device should be controlled by a strategy, set its control type back to Auto on the device page. If Manual is intended, the warning stays as a reminder. |
DEVICE_READ_ONLY |
CONTROL | WARNING | A device that could be controlled is set to read-only in its settings, so the controller never writes to it. | Switch read-only off in the device settings if the device should be controlled. |
DEVICE_STATE_UNKNOWN |
STATE | WARNING | The device reports a status the controller does not recognise. Details list the affected state signals. | Check the device's own display or app for its state. Report the device model and firmware version to Reduxi support if the warning persists. |
DEVICE_NOT_RUNNING |
STATE | WARNING | A configured device has not been running for about a minute — for example because it could not be started or the license device limit is reached. | Check the device settings and Settings → System → License. Save the device again or restart the controller; contact support if it stays. |
CONTROLLER_REBOOT_EXCESSIVE |
SYSTEM | WARNING | The controller software started more than 3 times in the current day. The event stays open until midnight. | Look for CONTROLLER_STARTUP, CONTROLLER_REBOOT and CONTROLLER_UPDATE_INITIATED entries in the Audit log. Frequent starts without a reason point to a power supply or hardware problem. |
When events close
An event closes only when the controller has positively seen that the problem is gone. While a device cannot be looked at — for example during a restart — its open events stay open. An event also closes when its device is deleted.
Warnings you expect
A device you keep in Manual on purpose, or a read-only meter, keeps its WARNING open. That is fine; focus on ERROR events and on warnings you cannot explain.
4 Checklist
- No open event (Time end
-) with severity ERROR. - Every open WARNING is understood, for example a device deliberately in Manual.
- After a fix, the event shows a Time end (for
POWER_ENERGY_MISMATCHthis can take 15 minutes, forSETPOINT_NOT_FOLLOWEDabout a minute).