Zigbee devices drop off regularly

[jclendineng]

The problem

I have bulbs that drop off a lot. It’s never the same bulb, but I get the same error each time. A power cycle gets them back on the network. This has been consistent across all versions since I got my HA Yellow.

What version of Home Assistant Core has the issue?

core-2023.5.0b1

What was the last working version of Home Assistant Core?

No response

What type of installation are you running?

Home Assistant OS

Integration causing the issue

Zigbee Home Automation

Link to integration documentation on our website

https://rc.home-assistant.io/integrations/zha

Diagnostics information

config_entry-zha-43a53bca9914c8f44545268db208e1f6.json.txt home-assistant_2023-04-28T19-04-44.289Z.log

Example YAML snippet

No response

Anything in the logs that might be useful for us?

Logger: bellows.zigbee.application
Source: runner.py:179 
First occurred: 11:45:26 AM (2 occurrences) 
Last logged: 11:45:26 AM

NWK conflict is reported for 0x79ed
Found cc:cc:cc:ff:fe:c6:d4:d3 device for 0x79ed NWK conflict: The Home Depot Ecosmart-ZBT-A19-CCT-Bulb

Additional information

No response

[home-assistant[bot]]

Hey there @dmulcahey, @adminiuga, @puddly, mind taking a look at this issue as it has been labeled with an integration (zha) you are listed as a code owner for? Thanks!

Code owner commands

Code owners of `zha` can trigger bot actions by commenting: - `@home-assistant close` Closes the issue. - `@home-assistant rename Awesome new title` Renames the issue. - `@home-assistant reopen` Reopen the issue. - `@home-assistant unassign zha` Removes the current integration label and assignees on the issue, add the integration domain after the command.

_{^{(message by CodeOwnersMention)}}

---

zha documentation zha source _{^{(message by IssueLinks)}}

[puddly]

Routing is not controlled by ZHA, nor can it reach devices that are unreachable, but I can offer three suggestions:

1. Click the "reconfigure" button on the affected devices, in case attribute reporting isn't set up (this will not work if the device is unreachable):

2. According to your diagnostic info, your RF environment is pretty noisy:

       "energy_scan": {
         "11": 39.90320178295578,
         "12": 91.05606689948522,
         "13": 84.164247274957,
         "14": 49.512515447068886,
         "15": 92.0598007161209,
         "16": 33.860880820104335,
         "17": 52.75969252664325,
         "18": 49.512515447068886,
         "19": 55.9836862725909,
         "20": 25.74050169409602,
         "21": 33.860880820104335,
         "22": 78.25348754651363,
         "23": 55.9836862725909,
         "24": 31.01324838787301,
         "25": 78.25348754651363,  // your channel
         "26": 87.33047519856483
       }

   You should be getting a warning on startup about channel utilization being high, which explains common interference sources and what you can do to alleviate problems.

3. If all else fails, you can try to move your network to another channel. Since you're running the beta, there is a way to migrate most devices, but it isn't exposed in the frontend. You can paste the following into your browser's dev console to do this:

   var hass = await window.hassConnection;
   
   await hass.conn.sendMessagePromise({type: "zha/network/change_channel", new_channel: "auto"});  // or a specific channel 11-26

[jclendineng]

Yes, tried 1 but as you said, devices are unreachable. 2 is a solid choice but I switched to 20 precisely for this reason, I do get congestion alerts but it's 85-90% which isn't terrible?

[jclendineng]

Also why does it say I'm on channel 25? I'm on 20 and I can confirm this in the zha config...weird.

[puddly]

The config is not used unless you form a new network. What's in the diagnostics is correct.

[jclendineng]

So my zha states that I'm on 20 and the logs that say I have congestion also say 20, those are both wrong? I did form a new network with 20 when I switched...

Edit. Scratch that I am on 20 not sure why the diag above says 25. I checked the diag to make sure and it clearly says 20 and has completely different energy scans so not 100% sure where that data above came from 😂

[jclendineng]

That said:

"energy_scan": {
  "11": 88.70042934643088,
  "12": 93.76433891498253,
  "13": 92.95959997754716,
  "14": 80.38447947821754,
  "15": 85.82097888710312,
  "16": 80.38447947821754,
  "17": 84.164247274957,
  "18": 82.35373987514762,
  "19": 36.830390267097734,
  "20": 93.76433891498253, // My channel
  "21": 89.93931580241996,
  "22": 78.25348754651363,
  "23": 82.35373987514762,
  "24": 43.057636198227904,
  "25": 49.512515447068886,
  "26": 82.35373987514762
},

I really don't have any good options. I'm in a house and neighbors aren't THAT close, this looks absolutely terrible. 19/24/25 look best but iirc 25 wasn't recommended as certain devices wouldn't support it.

[puddly]

Oh, my mistake: you are on 20. I would try migrating to 25 with the method I showed above, most of your ZigBee 3.0 devices should migrate on their own.

The number of devices with channel issues is absolutely miniscule so unless you're buying Legrand, Konke (which likes 25), or a few other obscure brands, you're fine migrating to 25.

[jclendineng]

I have no zigbee 3 😂 that's ok. I have re-paired them all before and it's fine. I only have 75 or so. I know zigbee mesh's usually only support up to 70 reliably so maybe it's just a matter of adding a mesh node.

[puddly]

I'd give the migration a try regardless, the worst that will happen is that it won't work.

Regarding mesh size: people routinely have 200+ devices without issues so I think something else is going on with your network. How far away are these bulbs from other devices on the network?

[issue-triage-workflows[bot]]

There hasn't been any activity on this issue recently. Due to the high number of incoming GitHub notifications, we have to clean some of the old issues, as many of them have already been resolved with the latest updates. Please make sure to update to the latest Home Assistant version and check if that solves the issue. Let us know if that works for you by adding a comment 👍 This issue has now been marked as stale and will be closed if no further activity occurs. Thank you for your contributions.