Zigbee failing, needs restart

Hello All

Quite new to HA.If I am posting in a wrong category, then sorry :slight_smile:

My setup:
HA running on PI4 2GIG. Zigbee via tasmota flashed Sonoff bridge. Zwave via USB dongle on PI.

Zigbee devices:
8 outdoor lights Phillips Hue
9 Sonoff temperature and humidity sensors
3 Sonoff door switches
3 Sonoff switches
3 Sonoff PIR

Using node-red to control my outdoor lights via PIR and sunrise/sunset.

My problem:
My zigbee networks partly crashes once a day or so. Lights freezes in any given state. Sometimes dimmed, sometimes fully lit. I cannot control my sonoff switches either.

I am still getting temperature, humidity and PIR readings.

In my logs i get these:

Watchdog heartbeat timeout:

Logger: homeassistant
Source: runner.py:115
First occurred: 10. december 2020 20.43.25 (8 occurrences)
Last logged: 8.20.34

Error doing job: Task was destroyed but it is pending!

And a lot of these:

2020-12-11 07:04:23 WARNING (MainThread) [bellows.zigbee.application] Watchdog heartbeat timeout:

2020-12-11 07:06:13 WARNING (MainThread) [bellows.zigbee.application] Watchdog heartbeat timeout:

2020-12-11 07:08:03 WARNING (MainThread) [bellows.zigbee.application] Watchdog heartbeat timeout:

2020-12-11 07:08:04 WARNING (bellows.thread_0) [bellows.uart] Reset future is None

A restart of HA solves it. But this is not a permanent solution :frowning:

Any ideas of how to troubleshoot this problem?

1 Like

Did you ever get this figured out? I am having the same problem

No :frowning:

My fix is a nightly reboot, a patch but it works

Is your Home Assistant db file over 300 mb? I was just dealing with this and think that is the issue.

I moved my recorder over to Mysql using the MariaDB addon, then deleted the db file, and rebooted.

I am no longer getting watchdog errors, or the reset future is none.

Now as to why that was the issue, I am not sure. It wasn’t a space issue, and this drive is brand new. But it appears to have solved the issue.

Nope…I lied.

It worked for a bit, but once I got my 17 devices ,loaded its back to completely unusable. Need to see if its related to the Home Assistant Blue that I just migrated too.

Did you ever got this fixed? My Zigbee via tasmota flashed Sonoff bridge just started to act up and I noticed the logs are full of:

2021-03-09 10:09:11 WARNING (MainThread) [bellows.zigbee.application] Watchdog heartbeat timeout:
2021-03-09 10:10:52 WARNING (bellows.thread_0) [bellows.uart] Reset future is None
2021-03-09 10:10:56 WARNING (bellows.thread_0) [bellows.uart] Reset future is None

I get the same sort of WARNINGs and ERROR here and there regarding zigbee and UART…

Same problem with ZHA network based on Sonoff zigbee bridge and sonoff zbmini’s…
Any solution?

Read the warning on the ZHA integration page, similar warning on the zigbee2mqtt website, and this issue:

1 Like

Shit! Gotta find something else that works reliably…