1.35.2

[Koenkk]

Feel free to discuss the v1.35.2 release here!

[vjonas]

Thank you! ♥‿♥

[Trebor87]

Many thanks

But still getting an error with TS011F when trying to OTA it.

Best regards

[Nerivec]

There's no firmware image available in Z2M for that combination of specs. Many devices use that combination of image type & manufacturer code (as often with Tuya), so it is further filtered by manufacturer name to ensure it's getting a proper firmware image for your specific device. _TZ3000_cayepv1a isn't in the list.
If you know of one, you can follow the instructions in the previous link to get support for it in Z2M.

[Nardol]

#20482 Notify systemd for start, stop, watchdog (@chrthi)

It looks the installation documentation is not yet updated, Koenkk/zigbee2mqtt.io#2475 is not merged.
Is it expected? Should we avoid to change SystemD unit?

[Koenkk]

Merged the pr!

[pietrouk]

Following the update, all my devices have disappeared from the list. They're still working though. How do I get them back?
database.db file has them listed.

I'm on rpi4-64
Core 2024.1.6
Supervisor 2023.12.1
Operating System 11.4
Frontend 20240104.0
Version of add on 1.35.2-1

[Nerivec]

I assume you tried restarting Z2M. Did you try a "Reboot system" of HAOS?

[pietrouk]

That's right and no luck

[ivancea]

After the update, I could still use everything as expected, but after some minutes, I lost connection with all my devices. Both Hass and Z2M worked the first seconds.
I rollbacked, checked everything worked again, re-updated, and checked that it failed again, so looks reproducible at least in my env.

Everything failing with a no response received (63). Example log:

Publish 'set' 'state' to 'Luz sala' failed: 'Error: Command 0x001788010b05e90f/11 genLevelCtrl.moveToLevelWithOnOff({"level":2,"transtime":0}, {"timeout":10000,"disableResponse":false,"disableRecovery":false,"disableDefaultResponse":false,"direction":0,"srcEndpoint":null,"reservedBits":0,"manufacturerCode":null,"transactionSequenceNumber":null,"writeUndiv":false}) failed (no response received (63))'

Versions:

Versión de Zigbee2MQTT
    1.35.2 commit: unknown
Tipo de coordinador
    ConBee2/RaspBee2
Versión de coordinador
    0x26720700
Coordinator IEEE Address
    0x00212effff07ed0d
Versión de la interfaz
    0.6.152
Zigbee-herdsman-converters version
    18.25.1
Zigbee-herdsman version
    0.33.5

[TheRealWaldo]

Having the same issue. Rolling back to the previous version returns it to normal operation.

[dc911x]

Yep same here, also using a conbee II stick.

[Albruaje]

Exactly the same for me. Using Phoscon RaspBee II

[Nerivec]

Make sure you check if you have devices stuck in "interviewing" (little icon on top of the device's picture in Z2M UI). After large updates that bring new features to your devices, if some indeed get stuck, it can help to push the button on them (should be "sleepy" end devices, on battery), just a push-and-release (not long-push reset). They otherwise fail interviewing (because of that sleepy mode), and that can lead to your network trying requests again and again, and disrupting other devices. (PS: This applies to EZSP, but I'd guess the same applies to deconz.)

Otherwise, if someone has a herdsman debug log, open an issue and post it there (if someone hasn't already).

[Koenkk]

This is Conbee specific, let's continue in #21162

[bergetun]

This relase breaks the support for Namrom dimmers :(

[Seger85]

@Koenkk just wanted to say thank you, what a great piece of software, donation on its way! Sensational and overwhelming!

I have seen a reported error:

IKEA repeaters dead after update to 1.35.2-1
#21147

which is still preventing me from updating. Can anyone else confirm the error and help solve the problem?

Best regards
Seger

[merlinpimpim]

Thanks for the update.

1. But I now get this type of message with my adapter: "Adapter disconnected, stopping"
   (Nabu casa Skyconnect (up to date) / Multiprotocol / EZSP)
   Every 6/7 minutes it seems.
   

2. I also have this problem with Lixee: https://github.com/Koenkk/zigbee2mqtt/issues/21151

3. I have a NodOn SIN-4-1-20 device showing an OTA firmware update but it doesn't really seem to exist, the update never applies:
   
   

4. And, Is this update now compatible with Silicon Labs Multiprotocol 2.4.4?

5. Edit: The OTAUpdate/deviceMessage' is still there:
   

Thanks

[merlinpimpim]

I migrated SLM to version 2.4.4 and it works with Z2M 1.35.2 😊
However, topics 1, 2 and 3 from my message above persist.
I'm going to mention point 1 in the thread you mentioned.

[Nerivec]

Could you open an issue for 1 and post a herdsman debug log with it? I'm looking into it, but I want to make sure what is triggering the reset, I need the herdsman debug for that.
Seems it is recovering properly now though, so that's a good start... 😄

PS: post a link here once you have, to avoid duplicates.

[merlinpimpim]

@Nerivec Thanks for the feedback.
Just a detail: I've just discovered that Z2M crashes and restarts every time the message (Adapter disconnected, stopping) appears in the logs. Also every time I extract the container logs (with docker logs CONTAINER_ID > log.txt 2>&1), I have the Z2M startup sequence and a bit of information but probably not what was happening at the time of the problem.
Are you still interested in these logs?

[Nerivec]

Yes, Adapter disconnected is the hard fail of Z2M, it relies on Home Assistant (or whatever system you are using to monitor the Z2M instance) to restart Z2M from scratch. For EZSP that can happen for 3 reasons, too many watchdog failures, failure to send data to the serial port (after a couple of tries), or failure of the serial port. Since the whole stack relies on the adapter and Z2M to keep in sync, any of these requires a full restart of the stack. It previously restarted internally (without reaching the Z2M/HA level, but that lead to issues, so now it's a proper restart -like the other adapter types-).

I'd need the logs right before the crash indeed. See if there are any error (using search) in the logs you have, in which case, they might help.
Is the timespan between restarts pretty close each time? That could be your system suspending your USB port I'd guess...

cat /sys/module/usbcore/parameters/autosuspend should give you the value of the autosuspend (-1 is disabled), depending on your system, as always.

[merlinpimpim]

As expected, here is the link to the new issue: #21198

[nickb24]

The update broke the Stelpro HT402 Outdoor temperature command for some reason. I opened an issue here #21152

Rolling back to 1.35.1 immediately makes the command work again.

[rrakso]

After performing the update, the add-on does not start 😕

[MrFaberinho]

u can click the refresh button

[arteck]

same problem with #21078

[weljajoh]

After upgrading to 1.35.2 the binding of my IKEA remote (E1524/E1810) to a Tradfri LED driver (ICPSHC24-30EU-IL-1) group was gone. It was now bound to something called 'default_bind_group'. Did a successfull rebind to the correct group through the GUI.

[MrFaberinho]

Anyone familiar with this RangeError?
Trying to connect EnOcean switch with Conbee III stick.

When pressing one of the four buttons I see this error in the debugger.

[chris-1243]

It seems to be related to this issue #12031