Nuki Pro Gen4 (Matter over Thread) becomes unreachable after a few hours – only battery restart helps, HA Matter/Thread restart does not

[CoLeTheGamR123]

Hi,

I am using a Nuki Smart Lock over Matter/Thread with Home Assistant and a Sonoff Thread/Matter dongle as my Thread border router. I was using this solution for 14 months and since 2 week its stopped working correctly.

I am having a recurring issue where the lock works normally at first, but after a few hours it becomes unreachable in Home Assistant. At that point the device is effectively offline from the Matter side.

What is important here:

• Removing the battery / restarting the Nuki itself fixes the issue, but only for a few hours
• Restarting the Thread service and Matter service in Home Assistant does not fix it
• So the workaround currently is only to power-cycle the lock itself

From the logs, it looks like Home Assistant can no longer maintain or re-establish the secure Matter session with the lock. I see repeated:

• Subscription liveness timeouts
• CASE session timeouts
• Failed re-subscription attempts
• The node being marked unavailable / offline

This makes it look like the lock is still being discovered at times, but communication over Matter/Thread eventually stops working until the lock is rebooted.

What I am seeing technically

My interpretation of the logs is:

1. The lock initially works normally
2. The Matter subscription times out
3. Home Assistant tries to re-subscribe
4. Repeated CASE session establishment attempts fail
5. The node is marked unavailable/offline
6. Even if the node is rediscovered later, the CASE session still cannot be established
7. Only removing and reinserting the Nuki battery makes it work again temporarily

Because restarting Home Assistant’s Matter/Thread services does not recover it, this does not look like just a Home Assistant service issue. It looks more like the Nuki Matter/Thread stack becomes unresponsive or stuck after some uptime.

Example log snippets

Initial subscription timeout / node becomes unavailable

2026-03-15 22:40:17.933 CHIP_ERROR [chip.native.DMG] Time out! failed to receive report data from Exchange: 20575i with Node: <0000000000000003, 1>
2026-03-15 22:40:17.934 INFO [matter_server.server.device_controller] Node:3 Subscription failed with CHIP Error 0x00000032: Timeout, resubscription attempt 2
2026-03-15 22:40:17.935 INFO [matter_server.server.device_controller] Node:3 Marked node as unavailable

CASE session timeout during re-subscription

2026-03-15 22:41:18.023 CHIP_ERROR [chip.native.SC] CASESession timed out while waiting for a response from peer <0000000000000003, 1>. Current state was 4
2026-03-15 22:41:18.023 CHIP_ERROR [chip.native.DMG] Failed to establish CASE for re-subscription with error ‘src/protocols/secure_channel/CASESession.cpp:594: CHIP Error 0x00000032: Timeout’
2026-03-15 22:41:18.024 INFO [matter_server.server.device_controller] Node:3 Subscription failed with CHIP Error 0x00000032: Timeout, resubscription attempt 3

Device reports “resource busy” while HA tries to re-establish the session

2026-03-15 22:41:33.326 CHIP_ERROR [chip.native.SC] Received error (protocol code 4) during pairing process: src/protocols/secure_channel/CASESession.cpp:2383: CHIP Error 0x000000DB: The Resource is busy and cannot process the request
2026-03-15 22:41:33.326 CHIP_ERROR [chip.native.DMG] Failed to establish CASE for re-subscription with error ‘src/protocols/secure_channel/CASESession.cpp:2383: CHIP Error 0x000000DB: The Resource is busy and cannot process the request’
2026-03-15 22:41:33.327 INFO [matter_server.server.device_controller] Node:3 Subscription failed with CHIP Error 0x000000DB: The Resource is busy and cannot process the request, resubscription attempt 4

Later the same pattern happens again

2026-03-15 23:48:55.011 CHIP_ERROR [chip.native.DMG] Subscription Liveness timeout with SubscriptionID = 0xead2fd9e, Peer = 01:0000000000000003
2026-03-15 23:48:55.012 INFO [matter_server.server.device_controller] Node:3 Subscription failed with CHIP Error 0x00000032: Timeout, resubscription attempt 0

Repeated retransmissions and CASE failures until the node is considered offline

2026-03-16 00:24:01.528 CHIP_ERROR [chip.native.EM] <<5 [E:20599i with Node: <0000000000000000, 0> S:0 M:194976404] (U) Msg Retransmission to 0:0000000000000000 failure (max retries:4)
2026-03-16 00:24:11.646 CHIP_ERROR [chip.native.SC] CASESession timed out while waiting for a response from peer <0000000000000003, 1>. Current state was 4
2026-03-16 00:24:11.647 INFO [matter_server.server.device_controller] Node:3 Subscription failed with CHIP Error 0x00000032: Timeout, resubscription attempt 12
2026-03-16 00:24:11.647 INFO [matter_server.server.device_controller] Node:3 Node considered offline, shutdown subscription

Rediscovery happens, but setup still fails

2026-03-16 00:29:10.774 INFO [matter_server.server.device_controller.mdns] Node:3 Discovered on mDNS
2026-03-16 00:29:10.774 INFO [matter_server.server.device_controller] Node:3 Setting-up node…
2026-03-16 00:30:06.544 CHIP_ERROR [chip.native.SC] CASESession timed out while waiting for a response from peer <0000000000000003, 1>. Current state was 4
2026-03-16 00:31:05.320 WARNING [matter_server.server.device_controller] Node:3 Setup for

Please try to help me this is a really important things for me and for my wife. We are using daily multiple times the matter connection and also its automated door opening happening through smart home when we arrive with the car and many other solutions.

[Juergen (Jürgen Pansy)]

Looks a little bit like the Smart Lock looses the thread connectivity. If you want to dig deeper, you can try to find the SLs IPv6 address via MDNS and then try to ping it. If connection is lost it will not be reachable.

If you have other devices in your Thread network that act as router, you might try to disable them temporarily and see if that solves the problem.

[CoLeTheGamR123]

Hello.

Sadly I dont really know how to do that first check what you mentioned, but I can answer your second question. I have one SINGLE thread router and one single matter device which is your Nuki door lock, so one router → one client. It was working until 14 months without any disconnect and now its started and the issue is on Nuki’s side thats clearly visible. Do you have any idea what could I do? I dont know if that means anything but if I go to Matter settings in Nuki app Remote Access is disabled and cannot be turned on the button is buggy.

[CoLeTheGamR123]

@Juergen

Looks a little bit like the Smart Lock looses the thread connectivity. If you want to dig deeper, you can try to find the SLs IPv6 address via MDNS and then try to ping it. If connection is lost it will not be reachable.

If you have other devices in your Thread network that act as router, you might try to disable them temporarily and see if that solves the problem.

OK AI helped me a little bit with Ping. I found the IPV6 address and I was able to ping from Home Assistant node to Nuki.

First try when its showing unavailabel: Ping does not work dropping error, cannot ping.

After I put out the battery, put it back than trying again:

Thank I got this answer, so looks like the ping also does NOT work when its showing unavailable so yes looks like the thread connectivity is dropped after a few hours in Nuki.

[CoLeTheGamR123]

I think Nuki team also realized the issue, they released a stability improvement firmware update for Matter. I just started to install right now. I’ll test it for 3 days and let you know about the news.