deCONZ integration HASSIo 0.88 , all disappeared. Help!

Hi @Robban.

Are you referring to the deconz logs?

My log config is as below, as per /components/deconz/#debugging-component

logger:
  default: info
  logs:
    pydeconz: debug
    homeassistant.components.deconz: debug

My Hassio logs are as follows:

INFO (MainThread) [homeassistant.setup] Setting up deconz
INFO (MainThread) [homeassistant.setup] Setup of domain deconz took 0.0 seconds.
------ Later in the log
DEBUG (MainThread) [pydeconz.utils] Sending {} to http://172.30.33.2:80/api/API_KEY
ERROR (MainThread) [pydeconz.utils] Timeout getting deCONZ data from http://172.30.33.2:80/api/API_KEY.
------ Later in the log
ERROR (MainThread) [pydeconz] Couldn’t load data from deCONZ
WARNING (MainThread) [homeassistant.config_entries] Config entry for deconz not ready yet. Retrying in 5 seconds.
------ Later in the log
ERROR (MainThread) [homeassistant.components.deconz] Config entry failed since one deCONZ instance already exists

Yesterday i tried to clear all deconz entries in the .storage directory, and then re-adding the intergration. Results where the same.
Should i try to uninstall the add-on and clearing out deconz entries in .storage and the start from scratch?

So what you’re saying is that even if you completely remove the integration and redo it the setup fails?

I’m sorry if I’m not clear, the sequence of failure is this.

Reboot Home Assistant with working deCONZ integration results in the above with all entities being unavailable.
If I then remove the integration and reboot a second time of course it doesn’t work.
After another reboot I can then add deCONZ integration again and all works flawlessly.
Until I reboot again and I need to start the process over…

In summary I need to reboot Home Assistant in between removing the integration and adding it again.

I’m sorry, English is not my primary language.

yes for me is same: my little trick is before rebooting, removing the integration, after reboot redo the integration

verry strange that not everyone is affected with this bug?

I’m doing that trick as well, it’s just a pain…

Yep, and now I have frequent restarts of HASS (still investigating the issue) and each time breaks deconz, for which entire family relies now for lights and most power sockets and switches

A big pain

https://github.com/dresden-elektronik/deconz-rest-plugin/issues/1288

Work with us to get to the bottom with this

Here is an interesting thread, maybe a solution for you guys… Keep us posted

I didn’t see this thread…that linked thread is where I discovered removing custom-updater solves this for me and another user.

yes, i just posted here for information only, it will probably help others too

I am also struggling with this. The integration adds an internal temporary docker IP, which won’t work after reboot. Editing .storage/core.config_entries to use external IP/port gives me this error:

2019-05-17 00:17:22 ERROR (MainThread) [pydeconz.websocket] Unexpected error 400, message='Invalid response status'

Is it possible to add it properly using YAML? What should I do about API key in that case?

Have you tried enabling discovery, that should help with updating IP on new address

Thanks, I might try that. I don’t like the new “wizard-clicky” way of doing things, it’s hard to control such things now.

It was already enabled. I am a bit surprised more people don’t have this problem, maybe deconz and HASS in separate dockers is an uncommon setup?

No its a common problem. But it only manifests if you restart deconz.

Maybe deconz discovery is not enabled?

OK, restarting is an important feature :smiley:
I am not familiar with deconz discovery, is that a separate thing? I just verified that I had discovery: in configuration.yaml.

I think it is disabled sometimes

I’m seeing this same issue since a while back. Running hassio 0.93.1 right now but problem appeared a few releases back. Using marthoc:s hassio add-on (https://github.com/marthoc/hassio-addons)

Rebooting / restarting hassio will result in a non working integration with deconz. Re-adding the integration fixes everything until the next restart. References to 172.x.x.x IP in the log files.

The 172… addresses are docker internal routing range. So that will change on each restart.

Enabe discovery in Hass and investigate how to enable deconz discovery, when both are running discovery should automatically update the address of deconz