Zigbee devices no longer working in 2023.6.2 #94733
Comments
|
Additional info: this person described my issue better than I did originally. |
|
Hey there @dmulcahey, @Adminiuga, @puddly, mind taking a look at this issue as it has been labeled with an integration ( Code owner commandsCode owners of
(message by CodeOwnersMention) zha documentation |
What the error specifically is saying is that the serial port isn't working. ZHA will try to reconnect to the coordinator indefinitely but if it never does and only a reboot fixes it, this is a problem with either the OS or your hypervisor. I'm leaning towards the latter. |
|
would a way of ruling this in or out be to take my 'fresh' install of 2023.06.01 and upgrading to 2023.06.02 and see what happens? |
|
If you can consistently replicate the problem with one specific setup and not with another, that would be a place to start. I suspect it would have to do with the version of Home Assistant OS itself, not of Core. |
|
I'm encountering the same problem. Since upgrading earlier today, ZHA will not start and just says After some diagnosing, I found that it stemmed back to the Silicon Labs Multiprotocol addon, which is allowing both Zigbee and Thread to run on my SkyConnect. The problem seems to be that there is some underlying issue with Zigbee and ZHA since this release, which then has a knock-on effect on Thread when the addon is running. If I restart the addon and don't restart ZHA, then my Thread network comes back up and works fine. But as soon as I try reloading the ZHA integration (Which always fails), my Thread network goes down and I lose control of all devices, requiring another restart of the Silicon Labs Multiprotocol addon. I can also confirm that neither restoring the backup created just before the update nor restoring a nightly backup fixes this problem. Something in the update has had a permanent impact on ZHA, SkyConnect, and/or the Silicon Labs Multiprotocol addon to the point where restoring a backup cannot fix it. |
|
A few hours ago on my restore'd NEW vm, i upgraded the OS back to 2023.06.02 - ZHA has been working much longer than it has been. so my timeline here is:
|
|
Hi same issue for me, have rolled back to 2023.6.1 for now will have a look at doing the above fix later. |
|
Hi all, I have the same problem. I am using a Conbee II stick and running HA OS as VM on Unraid. ZHA stopped working after an update, about 2 weeks ago (I don't remember exactly which OS version). Here are some evidence of the problem: ... and listed when trying to reconfigure ZHA but I still get the unable to connect error. Here is the log: The above exception was the direct cause of the following exception: Traceback (most recent call last): Looks like Some additional infos:
It's about w weeks that I am unable to use any Zigbee device... and I am really struggling to find a solution. Anyone managed to fix it? Someone has an idea of what to do? Thanks for the support! |
|
same problem here! |
|
For anyone reading this, try unplugging your dongle and plug it back in and report back if this works please. That fixed it for me, but I'm already back on 2023.6.1. I'm curious if 2023.6.2 will work for others after you reset the dongle or not. |
|
I'm also seeing this since upgrading OS from 2023.03 to 2023.06 this week, on Home Assistant Blue with a ConBee II.
There is no /dev/serial/* hierarchy at all, which is why ZigBee has collapsed, as the kernel is constantly trying and failing to initialize the ConBee USB device: Power cycling the Blue doesn't help. |
I tried multiple times and doesn't work. |
|
Same issue here, updated to Home Assistant 2023.6.2 a few days ago. Other Versions: SkyConnect Stick worked without problems before. After the Update, some ZigBee Devices (Aqara Temperature sensors) disconnected one by one. After a reboot the Stick is not available in the system at all. Checking via Terminal: Device Path should be like: /dev/serial/by-id/usb-Nabu_Casa_SkyConnect_v1.0_* |
Didn't work for me yesterday only fix so far to me is to stop at 6.1. will look into loading a fresh version 6.2 and then applying a system backup as some one suggested further up. |
This worked to get my system back to somewhat stable, correct (I am OP). |
|
Backup restore worked. Devicepath is back under /dev. |
|
I too have SkyConnect and am experiencing this issue since the upgrade. Restarting the HA host recovered some but not all Zigbee devices. |
Could you maybe also look via SSH if your Skyconnect Dongle is missing in /dev/serial/by-id/? |
I am using a Conbee II stick and I am able to connect using Deconz addon (which is disabled when I confgure ZHA). |
|
Hi still not fixed mine. Running generic x86-64 on an old laptop Have just attempted to re flash with a fresh new install and then run my backup but still no joy. mine is running a ziggbe wireless sonoff controller. |
|
Mostly same setup as @cpdowsett , and same issue as of all you Same zigbee controller as well. Something that I noticed is that, even if I repair the zigbee devices, they became unavailable after a period of time (cannot ensure how long it takes, sorry) |
Issues continues after Home Assistant OS 10.3 upgrade yesterday, now at 10.3 / 2023.6.2; /dev/serial/* still absent, USB errors still scrolling continuously in dmesg. |
|
@floatingatoll the issue you're seeing is a problem with the Conbee. If it fails to start up due to 2.4GHz interference, it will continuously reset and since it manages its own USB-serial interface, that will reset too. Make sure your Conbee is on a USB extension cable and away from 2.4GHz interference sources such as USB 3.0 ports and SSDs. I would also update its firmware with Phoscon as well. |
|
I have also been facing issues the last couple of days: Logger: homeassistant.components.zha.core.gateway
Source: components/zha/core/gateway.py:205
Integration: Zigbee Home Automation (documentation, issues)
First occurred: June 20, 2023 at 20:21:37 (4 occurrences)
Last logged: 12:23:01
Couldn't start EZSP = Silicon Labs EmberZNet protocol: Elelabs, HUSBZB-1, Telegesis coordinator (attempt 1 of 3)
Couldn't start EZSP = Silicon Labs EmberZNet protocol: Elelabs, HUSBZB-1, Telegesis coordinator (attempt 2 of 3)
Couldn't start EZSP = Silicon Labs EmberZNet protocol: Elelabs, HUSBZB-1, Telegesis coordinator (attempt 3 of 3)
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/components/zha/core/gateway.py", line 205, in async_initialize
self.application_controller = await app_controller_cls.new(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/zigpy/application.py", line 219, in new
await app.startup(auto_form=auto_form)
File "/usr/local/lib/python3.11/site-packages/zigpy/application.py", line 194, in startup
await self.initialize(auto_form=auto_form)
File "/usr/local/lib/python3.11/site-packages/zigpy/application.py", line 164, in initialize
results = await self.energy_scan(
^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/bellows/zigbee/application.py", line 699, in energy_scan
channels_to_scan.remove(channel)
KeyError: 11 |
|
Sorry nothing helpful to add. I did try everything I could think of last weekend but I am not at home in the week. When I get back tomorrow I will attempt to upload the requested info. My system all runs fine untill I update to 6.2. the only fix I have to keep my home running and the not wife complaining about things not coming on and dead switches it to take the system back to the morning before the update... Good luck all |
|
Here is my log from ZHA
|
|
@languy99 I meant the entire |
|
Here you go
|
|
Note that my issue started with the upgrade to OS 10.2 and 2023.6.1, not 10.3, in case that’s of any interest.I’ll need to secure a USB extension cable since one hasn’t been necessary in this setup until now, so I’ll report back when I have more data.On Jun 21, 2023, at 15:42, dereitz ***@***.***> wrote:
I’ll try that, I suppose. Not sure why it’s suddenly upset after an OS/HA upgrade and was fine for the six months prior. Such is the nature of radio interference, I guess!On Jun 20, 2023, at 13:09, puddly @.> wrote: @floatingatoll the issue you're seeing is a problem with the Conbee. If it fails to start up due to 2.4GHz interference, it will continuously reset and since it manages its own USB-serial interface, that will reset too. Make sure your Conbee is on a USB extension cable and away from 2.4GHz interference sources such as USB 3.0 ports and SSDs. I would also update its firmware with Phoscon as well. —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: @.>
Please let us know if you have any success -- those of us who are standing by (until I have the time to troubleshoot/debug) would really appreciate the feedback!
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
That fixed mine, all working after the upgrade. I don’t know why it worked until the update, but I appreciate the fix.On Jun 23, 2023, at 00:36, Rowan Green ***@***.***> wrote:Note that my issue started with the upgrade to OS 10.2 and 2023.6.1, not 10.3, in case that’s of any interest.I’ll need to secure a USB extension cable since one hasn’t been necessary in this setup until now, so I’ll report back when I have more data.On Jun 21, 2023, at 15:42, dereitz ***@***.***> wrote:
I’ll try that, I suppose. Not sure why it’s suddenly upset after an OS/HA upgrade and was fine for the six months prior. Such is the nature of radio interference, I guess!On Jun 20, 2023, at 13:09, puddly @.> wrote: @floatingatoll the issue you're seeing is a problem with the Conbee. If it fails to start up due to 2.4GHz interference, it will continuously reset and since it manages its own USB-serial interface, that will reset too. Make sure your Conbee is on a USB extension cable and away from 2.4GHz interference sources such as USB 3.0 ports and SSDs. I would also update its firmware with Phoscon as well. —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: @.>
Please let us know if you have any success -- those of us who are standing by (until I have the time to troubleshoot/debug) would really appreciate the feedback!
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
Hi All, I don't mean to be rude, but is someone from HA team looking at this issue? Is there a way to escalate this issue? If people bought a SkyConnect, paying HA Cloud and the setup is not working I think that a fix should be prioritized... |
This issue has @puddly's attention. If you're able please supply the information that he requested in an earlier comment: I'll do the same when I'm able... (not anytime soon unfortunately) |
You have to submit the debug log, and detailed description of your setup. Half of people in this thread, are getting "no such file or directory" which means the problem is not with ZHA, but with the underlying OS. whether the device file is no longer there, or link to the serial port is not there, etc. |
I am getting that also. Almost like it got uninstalled and nothing I do can get the usb stick to show up anymore. I even tried to roll back and nothing. |
Thanks for your reply, I tried to describe the issue at my best in my first comment: Some additional stuffs that I tried after I posted that comment:
Please let me know if the entire log would help (I should have pasted the most critical part). I am also available for a call if could be useful. Thanks again! |
|
My devices drop off after an hour or so, using the SkyConnect, HA stick. |
|
I posted last week saying I was having the same issues, and as no progress was being made to a solution I decided to try fixing the issue myself. I had 2 plans in mind:
I'd purchased a second SkyConnect with the plan of running 2 Zigbee networks, one ZHA and one Z2M, so that I could use the service with the best compatibility depending on the device. However, I'm now running Thread on my original SkyConnect (Running multi-pan firmware), as that still worked after the update, and I have transferred my Zigbee network to my second SkyConnect (Just running on Zigbee only firmware). After the transfer, a few of the powered devices came back online, but the majority of other devices required repairing. However, if you still have all the devices in HA, then a repair will still use the device that already existed, so no further configuration is needed, just a quick run around the house with my phone checking each device re-connected, and then my network was back up and running. It has been over a week since I did that, and I've not had a single device drop off the network, even after multiple HA restarts and reboots. I am now, however, much more dubious of updating HA, and will be monitoring issues for a few days after an update before I commit to it myself. Once I see the root of the issue has been identified and hopefully fixed, I will try migrating my ZHA network back to my original SkyConnect so that I can use my second SkyConnect for Z2M, but for now, everything is back working. Some further information about my solution:
|
|
Not sure if this is important, but I am seeing this log entry regularly in my host event logs: |
|
As mentioned above, it's next to impossible to debug these things without logs and a detailed description of your setup. See my comment above (#94733 (comment)) for what exact info would be helpful. |
|
If it’s of interest, we used a two meter USB extension cable, on the theory that if noise was newly an issue, that we’d best make it as much not an issue as possible.This resolved the file not found / missing device. I didn’t downgrade to see if the March kernel was somehow less sensitive to noise. But our setup probably needed that extension cable regardless.Changed over from ConBee II to SkyConnect today, everything is still working fine, so I’m confirmed stable and all good now. On Jun 27, 2023, at 06:57, puddly ***@***.***> wrote:
As mentioned above, it's next to impossible to debug these things without logs and a detailed description of your setup. See my comment above (#94733 (comment)) for what exact info would be helpful.
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
@JMoro-GH I'm responding to your email here. From the logs, it appears your Conbee is unresponsive. ZHA tries to connect to it, the connection is made, but the Conbee never sends back a single byte of data. There are a few reasons this can happen:
|
|
I found my issue. I run my HA in Qnap virtual machine. For some reason the usb pass through to the VM was turned off. I turned it on and reboot the VM and now it shows up. I don’t know how it got turned off but it did. |
Thanks for the reply. This helped me find the solution finally after 3 weeks. You were right, the issue was point 2. VM USB passthrough. For people facing a similar issue using unRaid, install usb manager plugin and re-configure ZHA with the new device path! More info here: https://community.home-assistant.io/t/solved-conbee-2-fails-to-connect-to-zha-ha-in-unraid-vm/431276/2 Please note that after the reconfiguration, you need to put all devices in pairing mode once. The good news is that they get automatically detected by HA. |
|
I just had this problem and couldn't figure out what was wrong. Change USB port, tried with extension and no extension. Nothing. Rebooted HA. Nothing. Then I rebooted Unraid which my HA VM is running on, worked immediately... So if you're running a VM, try rebooting the hypervisor with the Sonoff device plugged in to help "refresh" the USB connection or something. |
|
Hi have still not worked out how to fix mine. Just says Error setting up entry . the log mentions a missing file but i don't understand the change if someone would like to have a look then please find my log file bellow. Any ideas My instance is running on and old laptop using the generic 86x64 and my Ziggbe hub is running on a SONOFF Zigbee Bridge that connects to my WiFi on the other side of my home. 2023-07-04 19:56:36.878 WARNING (SyncWorker_2) [homeassistant.loader] We found a custom integration hive which has not been tested by Home Assistant. This component might cause stability problems, be sure to disable it if you experience issues with Home Assistant |
|
This seems relevant: homeassistant.config_entries] Error setting up entry socket://[192.168.86.250]:8888 for zha |
|
Thanks for the nudge |
|
I would restore from a backup (with the broken path) and then perform a radio migration from within ZHA: 
You can then select "Re-configure current radio" and then correct the URI. This will ensure none of your entity customizations are lost. This looks like a CPython bug: python/cpython#103848 |
|
Hi So anyone that setup ZHA with socket://[192.168.86.250]:8888 then all you need to do is remove the [ ] and then do the update. Thanks for the help puddly |
|
Glad you got it working! This will be fixed automatically in 2023.7.1 once #95917 is merged. |
and others
The problem
This (Zigbee devices stop working/reporting) started to happen as soon as I upgraded to the latest 2023.6.2 and for whatever reason I also can't revert the setup to the previous to try downgrading it.
Essentially if I reboot the entire VM cold, the zigbee integration works for... under an hour, then stops and won't until rebooting the VM again. Z-wave is working on the same stick seemingly without issues for now.
I think I also see... is any of this helpful? never contributed a bug report before.
I built a new VM and installed 2023.6.1 - after a little finagling zigbee is back to stable.
Stability for me looks like this setup:
What version of Home Assistant Core has the issue?
core-2023.6.2
What was the last working version of Home Assistant Core?
core-2023.6.1
What type of installation are you running?
Home Assistant OS
Integration causing the issue
ZHA
Link to integration documentation on our website
No response
Diagnostics information
home-assistant_zha_2023-06-16T09-17-07.745Z.log
Example YAML snippet
No response
Anything in the logs that might be useful for us?
No response
Additional information
No response
The text was updated successfully, but these errors were encountered: