Matter devices offline

[nocturne tapir]

All my Matter over Thread devices stopped responding on the 10th of May, after updating to 2025.5.1. The Matter log was flooded with CHIP errors, and resetting the Matter server, or Thread integration did nothing. Nor did restarting Home Assistant, or rebooting the VM. The Thread network had been stable for over a year and a half.

I tried air-gapping switches, resetting them, and resetting and attempting to re-add them. Nothing worked. I went so far as to shut down HA and spin up a clean instance of it. I've a case open in Git (144683), but I've seen nothing on the case. I'm down to resetting my Apple Thread routers (2x AppleTVs and 1x HomePod Mini in the hope that resetting these forces a new Thread network to be created.

EDIT: new HA instance shows no errors in the Matter log, the homeassistant.log file, or the fault file. Debugging is turned on for Thread.

Help?

[odd urchin]

I would suggest checking your IPv6 config on network. It's always DNS 😉

[nocturne tapir]

Thanks - there’s been no updates to the network or DNS for months. I did just enable ipv6 on the VLAN a couple of days back to see if that helped, but no go so far.

[subtle mist]

Power off all your thread border routers. Leave them off for 15-20 minutes to completely destroy any vestiges of the current thread network. Then power on one of the thread border routers, choose whichever one you would like to be primary. Wait another 15-20 minutes then power on the other thread border router options and see after another 30-60min if the thread network stabilizes.

[nocturne tapir]

I'll give that a try. At one point, I powered them all off for about an hour, but that was before I realized a couple of Inovelli switches were still on the network (just not responding) and, according to Eve, had set up as routers/leaders. I'll give it a try again.

[nocturne tapir]

Nope. That didn’t work. The network name remained the same. I tried resending the credentials to Home Assistant and to the phone.

Plugged my skyconnect in and set up openthread. Set it as the preferred network - same issue.

[nocturne tapir]

Okay, I reset all my Apple devices to factory settings, deleted the preferred network, and started completely from scratch. All testing has been done on three separate MoT devices, and the factory settings have also been reset. The "setting up" screen runs for about 30 seconds before erroring out—still no errors in the Matter logs.

[nocturne tapir]

Now I'm getting "Failed to save thread network credential, error: Thread network credentials does not match with any of the active thread networks around
" Failed to save thread network credential." The HomePod disappeared from the preferred Thread network. I'm removing OBTR and resetting the Thread network for now.

[nocturne tapir]

And back to the Matter log being spammed with:

2025-05-20 17:20:58.144 (MainThread) INFO [matter_server.server.device_controller] Starting Matter commissioning with code using Node ID 7.
2025-05-20 17:21:11.955 (Dummy-2) CHIP_ERROR [chip.native.EM] <<5 [E:48134i with Node: <0000000000000000, 0> S:0 M:100891927] (U) Msg Retransmission to 0:0000000000000000 failure (max retries:4)
2025-05-20 17:21:19.595 (Dummy-2) CHIP_ERROR [chip.native.SC] PASESession timed out while waiting for a response from the peer. Expected message type was 33
2025-05-20 17:21:28.149 (Dummy-2) CHIP_ERROR [chip.native.CTL] Discovery timed out
2025-05-20 17:21:31.928 (Dummy-2) CHIP_ERROR [chip.native.EM] <<5 [E:48135i with Node: <0000000000000000, 0> S:0 M:100891928] (U) Msg Retransmission to 0:0000000000000000 failure (max retries:4)
2025-05-20 17:21:40.391 (Dummy-2) CHIP_ERROR [chip.native.SC] PASESession timed out while waiting for a response from the peer. Expected message type was 33
2025-05-20 17:21:53.989 (Dummy-2) CHIP_ERROR [chip.native.EM] <<5 [E:48136i with Node: <0000000000000000, 0> S:0 M:100891929] (U) Msg Retransmission to 0:0000000000000000 failure (max retries:4)
2025-05-20 17:22:01.189 (Dummy-2) CHIP_ERROR [chip.native.SC] PASESession timed out while waiting for a response from the peer. Expected message type was 33

[nocturne tapir]

Interesting the inovelli switch blinked green indicating it was paired.

[nocturne tapir]

Attempted again, and I'm getting errors in the Matter logs again. Enabled debugging again and I get the following debug errors:

2025-05-21 09:00:24.946 (MainThread) INFO [matter_server.server.device_controller] Starting Matter commissioning with code using Node ID 10.
2025-05-21 09:00:25.005 (ThreadPoolExecutor-0_1) DEBUG [matter_server.server.storage] Saved data to persistent storage
2025-05-21 09:00:25.477 (MainThread) DEBUG [matter_server.server.device_controller.mdns] Discovered commissionable Matter node: AsyncServiceInfo(type='_matterc._udp.local.', name='D9609621D76BC07D._matterc._udp.local.', addresses=[], port=5540, weight=0, priority=0, server='AA7AF20D245B0B32.local.', properties={b'VP': b'4961+1', b'SII': b'800', b'SAI': b'2000', b'T': b'0', b'D': b'1074', b'CM': b'2', b'PH': b'36', b'PI': None}, interface_index=None)
2025-05-21 09:00:25.541 (MainThread) DEBUG [matter_server.server.device_controller.mdns] Commissionable Matter node disappeared: AsyncServiceInfo(type='_matterc._udp.local.', name='00D5D2D1182AD583._matterc._udp.local.', addresses=[], port=None, weight=0, priority=0, server=None, properties={}, interface_index=None)

I'm finally able to add these devices back to HomeKit, but I'm not able to add them to HA.

[nocturne tapir]

I’ve now managed to add five Nanoleaf essentials bulbs back into HomeKit and six Inovelli switches. I’ve got four more to add, and two Eve Light switches but have stalled on two Inovelli switches.

Still can not add anything to Home Assistant.

[nocturne tapir]

I'm still unable to add devices to Home Assistant. Devices have successfully been re-added to HomeKit, but I can't add to HA either by sharing a device already added to HK, or a "new" device. Log attached. How do I get eyeballs on this/support?