Can't send command to Nuki Lock over Thread

Configuration Matter/Thread
1

Hello

I read recommendations for Matter/Thread in this section.

I have HA OS working on Proxmox and SMLight SLZB-MR1 as OpenThread Border Router. I can see Nuki in the Matter devices with fdb3:a9d2:e7c3:1:b5ef:e6cb:7e05:bd76 ip adress. I can interview with the device. I can ping the lock from HA inside device settings ping command. HA OS has the fixed ipv4 adress set in the system network settings and ipv6 adresses are automatic as fd7e:f16f:c9e2:7c2d:592f:57f5:b26a:2cc5 and fe80::6776:d44d:a6b8:e826. Proxmox Host has a fixed ipv4 adress and no ipv6 adress set. When I send a open command after 30 sec I get the message Smart Lock Ultra became unknown. When I look at the Matter Server logs I see below messages.

2025-10-30 00:41:30.506 (MainThread) ERROR [matter_server.server] Error doing task: Task exception was never retrieved
Traceback (most recent call last):
File “/usr/local/lib/python3.12/site-packages/matter_server/server/client_handler.py”, line 213, in _run_handler
raise err
File “/usr/local/lib/python3.12/site-packages/matter_server/server/client_handler.py”, line 196, in _run_handler
result = await result
^^^^^^^^^^^^
File “/usr/local/lib/python3.12/site-packages/matter_server/server/device_controller.py”, line 717, in send_device_command
return await self._chip_device_controller.send_command(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/usr/local/lib/python3.12/site-packages/matter_server/server/sdk.py”, line 308, in send_command
return await self._chip_controller.SendCommand(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/usr/local/lib/python3.12/site-packages/chip/ChipDeviceCtrl.py”, line 1552, in SendCommand
return await future
^^^^^^^^^^^^
chip.interaction_model.InteractionModelError: InteractionModelError: Failure (0x1)
Please note that all devices are in the same ipv4 subnet. I can’t enable ipv6 over service provider. It is not supported. What can be the reason for the problem?

I also tried using sysctl -w net.ipv6.conf.vmbr0.accept_ra=2 && sysctl -w net.ipv6.conf.vmbr0.accept_ra_rt_info_max_plen=64 commands to Proxmox host, but then Nuki doesn’t receive ip adress anymore after restart of open thread borser and Matter server. Proxmox host receives an ip adress fd7e:f16f:c9e2:7c2d:eaff:1eff:fedd:bda1/64.

Regards

Gokalp

2

What do you see under the Thread integration about number of TBR available ?

3

I have one TBR, that is SLZB-MR1.

4

Does the Nuki Lock every come back from being unavailable?
Has the “Open” command worked before?

Since you could successfully pair the device and later ping it from the HA UI, I don’t see this as an IPv6 problem. If the device comes back from being unavailable, most likely it is a Thread networking related. If the “Open” command has not worked before, it could be that the “InteractiveModelError” means Matter doesn’t like something about the Cluster/Attributes used by the device with the “open” command.

5

If you have only one TBR via USB, that discard already a lot of issue.

Like my experience those days when my house switch from copper line to fiber line. I removed my own synology router to the router provided by ISP because it’s having 10Gbps port but this box have special routing for the IPTV and then all my Matter devices are now not working anymore with HA.
I’m also not able to do the pairing with google home and TBR nest gen2 anymore which is completely weird.

IMO, I would check everything related to router if accessible instead of Proxmox.

6

The problem was Nuki lock wasn’t calibrated. I calibrated and it worked without problems. Thks a lot for your answers and support.