[ZHA] SONOFF ZBMINIL2 firmware 0x0000100e repeatedly becomes unavailable with Connect ZBT-2 while local relay still works
The problem
Several SONOFF ZBMINIL2 no-neutral modules repeatedly become unavailable in ZHA.
Only the ZBMINIL2 devices are affected. My other Zigbee devices, including covers and sensors, remain connected and work normally.
When a ZBMINIL2 becomes unavailable:
- Remote control from Home Assistant no longer works.
- The physical wall switch and local relay still work.
- The relay can still be heard clicking and the light can be controlled locally.
- The module remains powered, but its status LED stays permanently off.
- A full mains power cut for several minutes does not reliably restore the Zigbee connection.
- Opening the ZHA device-joining window and holding the ZBMINIL2 button for approximately 5 seconds makes it rejoin without deleting it from Home Assistant.
- Remote control and normal LED behaviour return, but the device can become unavailable again after a few minutes or a few hours.
All affected modules are SONOFF ZBMINIL2 devices running firmware 0x0000100e.
The issue was first observed on Home Assistant Core 2026.8.3 and is still present after updating to Core 2026.9.1.
The attached diagnostics were downloaded while the device was unavailable. They show available: false, no LQI/RSSI value, and a last communication time of 2026-08-30T22:57:57 UTC.
What version of Home Assistant Core has the issue?
core-2026.9.1
What was the last working version of Home Assistant Core?
No response
What type of installation are you running?
Home Assistant OS
Integration causing the issue
Zigbee Home Automation (ZHA)
Link to integration documentation on our website
https://www.home-assistant.io/integrations/zha/
Diagnostics information
zha-01KF5GHVAX80VE35X4P59MN515-SONOFF ZBMINIL2-486360ca8c2256ef1f687752702939ae.json
Example YAML snippet
None. ZHA is configured through the Home Assistant user interface and I have no custom ZHA configuration related to this issue.
Anything in the logs that might be useful for us?
Relevant messages observed in the ZHA logs include:
ZIGBEE_SOURCE_ROUTE_FAILURE
ZIGBEE_NO_APS_ACK
Failed to send request: device did not respond
ping attempts have been exhausted, marking the device unavailable
STANDARD_SECURITY_UNSECURED_REJOIN
STANDARD_SECURITY_SECURED_REJOIN
After a 5-second pairing operation, the device rejoins with its existing IEEE address but can receive a new NWK address. It later becomes unavailable again.
Additional information
System and device information:
- Home Assistant Core: 2026.9.1
- Installation: Home Assistant OS
- Home Assistant OS: 18.2
- Supervisor: 2026.08.0
- Hardware: Home Assistant Green
- Architecture: aarch64
- Python: 3.14.6
- Zigbee integration: ZHA 2.2.1
- ZHA quirks: 2.2.1
- Coordinator: Home Assistant Connect ZBT-2
- Device manufacturer: SONOFF
- Device model: ZBMINIL2
- Device type: EndDevice
- Installed firmware: 0x0000100e
- Latest firmware offered by ZHA: 0x0000100e
- Quirk applied: false
- Device status in the attached diagnostics: unavailable
- LQI and RSSI while unavailable: null
I have not enabled the TRUST_CENTER_POLICY / ALLOW_UNSECURED_REJOINS workaround.
Multiple users have reported very similar behaviour with SONOFF ZBMINIL2 firmware 0x0000100e, ZHA and Silicon Labs coordinators:
Zigbee device signature:
{
"node_descriptor": {
"logical_type": 2,
"complex_descriptor_available": 0,
"user_descriptor_available": 0,
"reserved": 0,
"aps_flags": 0,
"frequency_band": 8,
"mac_capability_flags": 128,
"manufacturer_code": 4742,
"maximum_buffer_size": 82,
"maximum_incoming_transfer_size": 1024,
"server_mask": 11264,
"maximum_outgoing_transfer_size": 1024,
"descriptor_capability_field": 0
},
"endpoints": {
"1": {
"profile_id": "0x0104",
"device_type": "0x0002",
"input_clusters": [
"0x0000",
"0x0001",
"0x0003",
"0x0006",
"0x0007",
"0x0b05",
"0xfc57"
],
"output_clusters": [
"0x0019"
]
}
},
"manufacturer": "SONOFF",
"model": "ZBMINIL2",
"class": "zigpy.device.Device"
}
Source: home-assistant/core