Devices & Setup

Zigbee bridge / coordinator lost — devices unreachable

Recover when the Zigbee coordinator (Hue Bridge, SmartThings, ConBee, Sonoff Dongle, Home Assistant ZHA/Zigbee2MQTT) goes offline: backup-restore, network healing, channel changes, and the cases where re-pairing every device is unavoidable.

Zigbee mesh topology + recovery model

Reference images and diagrams. Click any image to view full resolution.

Original concept diagram (not vendor copyright). A Zigbee network has exactly one coordinator. Routers (always-powered devices) extend the mesh; end devices (battery-powered) route through nearby routers. Coordinator backup is the difference between minutes and days of recovery work.

Problem summary

Every Zigbee network has exactly one coordinator — the device that owns the network credentials, network key, and the routing table. When the coordinator goes offline or is replaced, every Zigbee device disconnects until the coordinator returns or a backup is restored to a new coordinator. The recovery options depend on whether you have a coordinator backup and which platform manages it.

Operator snapshotEvidence first

First proof

Coordinator hardware is powered and reachable.

Screen to open

Frontend > Settings > Coordinator. Verify connected, firmware version, channel, network address.

Expected signal

Bridge status LED on; USB stick host (Home Assistant) running; coordinator UI reachable.

Stop boundary

Stop before factory-resetting devices that are battery-powered with hard-to-access reset buttons (in-wall sensors).

Layer path

1Every Zigbee network has exactly one coordinator that owns the network credentials and routing table. Coordinator failure or replacement disconnects every Zigbee device until the coordinator returns or a backup is restored to a new coordinator.

2Recovery depends on whether a coordinator backup exists: with a backup, the network state can be migrated to a new coordinator and devices auto-rejoin. Without one, every device must be factory-reset and re-paired.

3Channel changes (firmware-driven or manual) break the mesh until devices re-discover the new channel.

4Coordinator backups should be taken regularly via Home Assistant / Zigbee2MQTT; closed-ecosystem bridges (Hue, SmartThings cloud) have limited backup options.

Runbook

Step-by-step runbook

Start here. Do each check in order, compare it to the expected result, and stop when the evidence explains the failure or the safe stop point applies.

Verify coordinator is online and reachable

Check: Bridge LED on; USB stick host (HA) running; coordinator UI loads.

Expected result: Coordinator hardware is fine.

If not: If not, power-cycle; if still dead, vendor support OR replacement coordinator + restore.

Check coordinator UI for device list state

Check: Hue app / SmartThings app / Home Assistant ZHA or Zigbee2MQTT frontend.

Expected result: Devices listed (even if 'unavailable') OR empty.

If not: Devices listed but unavailable → network heal (step 4). Empty → restore backup OR re-pair (step 3).

Restore from backup if coordinator state is lost AND backup exists

Check: Copy the backup files to the new coordinator install; restart the integration.

Expected result: Devices reappear in the coordinator UI.

If not: If no backup exists, factory-reset every device and re-pair — no shortcut.

Trigger network heal if devices are listed but unavailable

Check: Reconfigure Device on powered router devices; wait 24h for battery devices to check in.

Expected result: Mesh re-establishes; devices return to available state.

If not: Some battery devices check in only every few hours — patience.

If channel change broke the mesh, wait or change Wi-Fi

Check: Identify Zigbee channel; verify Wi-Fi 2.4 GHz channels are 1/6/11; let mesh heal.

Expected result: Wi-Fi and Zigbee on non-conflicting channels.

If not: Don't change Zigbee channel back arbitrarily — the change forces all devices to re-discover.

Take a coordinator backup now if you didn't have one

Check: Zigbee2MQTT > Settings > Backup. HA: enable automatic backups including the Zigbee integration.

Expected result: Backup exists outside the coordinator host.

If not: Future coordinator loss recoverable in minutes instead of days of re-pairing.

Decision tree

If: Coordinator UI shows the device list intact but devices 'unavailable'.

Then: Mesh issue, not coordinator data loss.

Action: Network heal: Hue app > Search for new lights; Home Assistant ZHA: Reconfigure Device on router devices. Wait 24h for battery-powered devices to check in.

If: Coordinator UI is empty / 'no devices found' AND a backup exists.

Then: Restore from backup.

Action: Home Assistant ZHA: copy `/config/zigbee.db` to the new install. Zigbee2MQTT: copy `coordinator_backup.json` and `database.db`. Restart the integration.

If: Coordinator UI is empty AND no backup exists.

Then: Re-pair every device — no shortcut.

Action: Factory-reset each Zigbee device (manufacturer-specific procedure). Add to coordinator. Group by room to make the work less painful.

Safe stop: Stop before factory-resetting devices that are battery-powered with hard-to-access reset buttons (in-wall sensors).

If: Moving Home Assistant to a new server and want to keep the Zigbee network.

Then: Copy the full Zigbee state + USB stick BEFORE starting the new install.

Action: ZHA: copy /config/zigbee.db. Zigbee2MQTT: copy /config/zigbee2mqtt/ directory. Move USB stick to new host. Start integration. Network is preserved.

If: Hue Bridge dead, new bridge needed.

Then: Hue migration tool if old bridge partially works; otherwise factory-reset every light.

Action: Hue app > settings > Hue Bridges > Migrate from old Bridge. If dead: factory-reset lights (turn off/on 5x in 8 seconds for most Hue lights) and re-pair to new bridge.

If: Suspected channel collision with Wi-Fi.

Then: Wi-Fi on Zigbee channel range causing instability.

Action: Wi-Fi: channels 1, 6, or 11 only (these don't overlap with each other). Zigbee: default 11 or 15. Avoid configurations where Wi-Fi channel sits between Zigbee channels.

Evidence

Evidence table

Symptom	Evidence to collect	Likely layer	Next action
Hue app shows all lights 'unreachable' simultaneously.	Hue app status indicator on every light.	Coordinator unreachable	Power-cycle the Hue Bridge; if still unreachable, check Hue Bridge status LED — vendor support if dead.
Home Assistant ZHA shows devices 'unavailable' (yellow/red status) but coordinator is healthy.	Settings > Devices & Services > ZHA > devices list.	Mesh routing broken	Reconfigure Device on router devices (powered Zigbee devices); wait 24h for battery devices to check in.
Zigbee2MQTT frontend shows coordinator firmware version is much older than the latest stable.	Frontend > Settings > About > coordinator firmware.	Outdated coordinator firmware (potential instability)	Flash the latest stable coordinator firmware via the documented procedure for that USB stick model (zigbee2mqtt.io guide).
After Home Assistant reinstall, Zigbee2MQTT shows 'no devices'.	Frontend > devices tab is empty.	Coordinator state lost (backup not restored)	Stop. Restore `coordinator_backup.json` + `database.db` BEFORE re-pairing any device; otherwise you create a new network and lose ability to recover the old one.

Reference

Commands and settings paths

Check coordinator status (Home Assistant Zigbee2MQTT)

Frontend > Settings > Coordinator. Verify connected, firmware version, channel, network address.

Where: In the Zigbee2MQTT web frontend (typically port 8080).

Expected: Coordinator shows Connected; firmware version current; channel matches expected (default 11).

Failure means: If Disconnected, USB stick or host issue.

Safe next step: Check USB stick is plugged in firmly; check host (Home Assistant) is running.

Take coordinator backup (Zigbee2MQTT)

Frontend > Settings > Backup > Download backup. Stores `coordinator_backup.json` + database.

Where: In the Zigbee2MQTT web frontend.

Expected: Backup file downloaded; size > 0 bytes.

Failure means: Without periodic backups, coordinator loss = re-pair everything.

Safe next step: Schedule automatic backups via Home Assistant automation.

Restore coordinator backup (Zigbee2MQTT)

Stop Zigbee2MQTT add-on. Copy `coordinator_backup.json` and `database.db` to `/config/zigbee2mqtt/` on the new host. Move USB stick to new host. Start Zigbee2MQTT.

Where: On the new Home Assistant host.

Expected: Devices appear in the Zigbee2MQTT frontend; states reflect the backup snapshot.

Failure means: If devices don't appear, network key mismatch — verify coordinator is reading the correct backup files.

Safe next step: Wait 24h after restore for battery-powered devices to check in.

Trigger network heal (Home Assistant ZHA)

Settings > Devices & Services > ZHA > Configure > Reconfigure Device on powered router devices.

Where: In Home Assistant.

Expected: Devices re-establish mesh routing; previously-unavailable battery devices reconnect.

Failure means: Battery devices check in on their own schedules (hours-to-days).

Safe next step: Patience helps — don't force-reset devices that just haven't woken up yet.

Hardware boundary

Hardware and platform boundary

Change only when

Migrating to Zigbee2MQTT from a closed bridge (Hue, SmartThings) is the right next step only when you understand the migration cost (factory-reset every device) — the open coordinator has better backup/restore but the transition is painful.

Evidence that matters

Coordinator backup cadence, channel selection (avoid Wi-Fi overlap), and powered-router-device placement for mesh health matter most.

Evidence that does not matter

Newer coordinator hardware (Sonoff Zigbee Dongle Plus, SkyConnect) doesn't change recovery patterns; what matters is whether you have backups.

Avoid

Avoid moving the coordinator USB stick between hosts without backing up state, factory-resetting devices speculatively, or running multiple coordinators on the same channel.

Related tool/checklist

Use the linked tool when you need a guided plan from your exact symptoms instead of a static checklist.

Device setup troubleshooter

Last reviewed

2026-05-18 · Reviewed by HomeTechOps. Reviewed against the Connectivity Standards Alliance's Zigbee documentation and Zigbee2MQTT's adapter/coordinator documentation; specifically the coordinator-owned-network-state model, the backup-and-restore migration flow, and the channel-collision interaction with Wi-Fi.

Sources/assumptions

Assumes a Zigbee 3.0 network (the current standard); older Zigbee Pro / Zigbee Light Link networks have additional migration constraints.
Assumes the coordinator platform supports backups; some legacy bridges (very old Hue Bridge v1) don't have a documented backup/restore flow.

Source-backed checks

HomeTechOps turns official docs and conservative safety rules into a shorter runbook. These links are the source trail for the page direction.

Connectivity Standards Alliance: ZigbeeUsed for Zigbee mesh architecture, coordinator role, and network key behavior in Zigbee coordinator-loss diagnostics.Zigbee2MQTT: Adapter (coordinator) docsUsed for Zigbee coordinator/adapter recovery, backup, and replacement flows on Home Assistant / Zigbee2MQTT setups.

Diagnostic rule

1. Check the simplest signal first.
2. Change one thing at a time.
3. Avoid unsafe electrical or data-loss steps.
4. Stop when the evidence says the risk moved.

FAQ

I'm moving my Home Assistant install to a new server — how do I keep my Zigbee network?

For ZHA: copy the entire `/config/zigbee.db` plus `/config/.storage/core.config_entries` (or use Home Assistant's built-in Backup feature). For Zigbee2MQTT: copy the entire `/config/zigbee2mqtt/` directory (includes `coordinator_backup.json`, `database.db`, `state.json`). On the new host, move the USB stick over BEFORE starting the Zigbee integration, restore the files, then start. The new system picks up the existing mesh.

Hue Bridge died and the new one won't see my old lights — what now?

Hue Bridge has a migration tool (Hue app > settings > Hue Bridges > Migrate from old Bridge) that works only if the old bridge is still partially functional. If it's fully dead, you need to factory-reset each light (most Hue lights: turn off/on 5x in 8 seconds) and re-pair to the new bridge. The lights themselves are fine; you're just rebuilding the network.