Maintenance Mode

This page explains what Maintenance Mode is and when to use it. Press the Maintenance Mode button before you change a battery, re-pair a sensor, or touch broken hardware. It tells the integration that you are actively fixing things, preventing your manual repairs from polluting the system's learned statistics.

Why It Exists

Because the integration cannot see your hands, a manual fix (like a battery swap, a power cycle, or a device re-pair) looks identical to a device recovering on its own from a long silence.

If the system learned that long silence as part of the device's normal rhythm, it would teach the integration that being broken for hours is just what that device does. That permanently widens the freeze window for that specific hardware, making every future freeze verdict much slower.

The integration recognizes several kinds of intervention by itself. It catches a recovery during a Zigbee2MQTT pairing window, one that spans a coordinator or WiFi outage, one that follows a restart, and, on ZHA, one on a device the gateway announced as re-paired, reconfigured, or removed.

Maintenance Mode is for everything those systems cannot announce: a battery pull, a power cycle, a button press, or any physical work on a stack that says nothing. It gives you a way to supply a signal to the integration that something frozen or unavailable is about to be fixed.

Using It

Press the Maintenance Mode button on the Device Sentinel integration page before touching the hardware.

The window opens for ten minutes by default, which you can adjust from 5 to 60 minutes under Advanced. Press the button again to end the window early; otherwise, it closes itself automatically when the time expires.

What It Changes

While the window is open, any watched device that recovers, meaning it goes from frozen, unavailable, or unknown back to reporting, has that silence gap ignored and restarted rather than learned.

Detection is untouched. A device that goes down during Maintenance Mode is still detected and added to the problem list normally. It only filters the recoveries.

The Fine Print

Opening and closing are both written to the system events log, so your daily brief can explain any discarded recoveries between them. If a genuine self-recovery happens to occur inside your ten-minute maintenance window, it is discarded. The silence episodes report marks those gaps no (maintenance), so nothing disappears without a record.