Nuki entities become unavailable after some time

kvj / hass_nuki_ng

Better support for Nuki devices in the Home Assistant

MIT License

176 stars 33 forks source link

Nuki entities become unavailable after some time #79

Open erd82 opened 2 years ago

erd82 commented 2 years ago

I could successfully control my bridge and Nuki Smart Lock 3.0 using this integration. Thanks for your great work!

After around an hour, the entities become unavailable.

At the same time the mobile app from Nuki shows that all components are online. In my router I can also see that the bridge is connected. So it seems to be a problem within the integration.

Reloading the integration helped to get the entities "online" again. As a workaround I guess I could create an automation which regularly reloads the integration (> workaround).

Do you have this issue "Nuki entities become unavailable after some time" on the agenda? Or is there any solution (no workaround) available?

Diddlik commented 2 years ago

I have the same issue. I have teo nuki (2.0 and 3.0) and both go unavaible.. Screenshot_2022-08-13-14-32-37-77_c3a231c25ed346e59462e84656a70e50

sgi320 commented 2 years ago

seems everybody faces the same issue:

copo03 commented 2 years ago

seems everybody faces the same issue:

I solve it by putting the external url in configuration.

erd82 commented 2 years ago

I solve it by putting the external url in configuration.

Just to understand your solution correctly: Where exactly did you put the external URL? And did you configure Nuki by adding it to configuration.yaml or using the integration UI?

kvj commented 2 years ago

is it a persistent issue? e.g. only integration reload helps?

I have Nuki 2.0 with Bridge and I've never experienced something similar. Sometimes entities become unavailable because integration fails to get the latest state, but it recovers after next successful polling

Diddlik commented 2 years ago

I have two HA instances (main server and backup). First I installed nuki_ng on the backup, then on the main server. On the backup server it runs 90% of the time, very rarely integration needs to be reloaded. On the main server the integration always needs to be reloaded so that the elements are updated.

copo03 commented 2 years ago

I solve it by putting the external url in configuration.

Just to understand your solution correctly: Where exactly did you put the external URL? And did you configure Nuki by adding it to configuration.yaml or using the integration UI?

Settings > System > Network > External URL

kvj commented 2 years ago

Please try version 0.3.10

I've added connection timeouts to each bridge API call, this should unblock polling thread

Diddlik commented 2 years ago

@kvj I will watch it the days and hope it is now fine ;)

Diddlik commented 1 year ago

I think the problem still exist: grafik