Homeassistant.components.light.hue Unable to reach bridge (flapping)

[sj3fk3]

Home Assistant release with the issue:
0.67.0
but also reports from version 0.66

Last working Home Assistant release (if known):
0.65.6

Operating environment (Hass.io/Docker/Windows/etc.):
Linux virtualenv

Component/platform:
hue

Description of problem:

Flapping connection.. (disconnect and reconnect)

Problem-relevant configuration.yaml entries and (fill out even if it seems unimportant):

hue:
  bridges:
    - host: a.b.c.d
      filename: phue.conf
      allow_unreachable: true
      allow_hue_groups: true

Traceback (if applicable):

2018-04-09 19:31:15 ERROR (MainThread) [homeassistant.components.light.hue] Unable to reach bridge 10.0.1.17
2018-04-09 19:32:09 INFO (MainThread) [homeassistant.components.light.hue] Reconnected to bridge 10.0.1.17
2018-04-09 19:57:30 ERROR (MainThread) [homeassistant.components.light.hue] Unable to reach bridge 10.0.1.17
2018-04-09 19:58:15 INFO (MainThread) [homeassistant.components.light.hue] Reconnected to bridge 10.0.1.17
2018-04-09 20:06:12 ERROR (MainThread) [homeassistant.components.light.hue] Unable to reach bridge 10.0.1.17
2018-04-09 20:07:01 INFO (MainThread) [homeassistant.components.light.hue] Reconnected to bridge 10.0.1.17

Additional information:

[nelbs]

I think this is the same issue as #13685

[sj3fk3]

It's not the same issue, but it might be related to the same underlying problem.

[Mariusthvdb]

since i updated to 067.1
component light.hue never connects, and refuses to load Hassio consequently:

8-04-28 16:41:22 ERROR (MainThread) [homeassistant.components.light.hue] Unable to reach bridge 192.168.1.212
2018-04-28 16:41:23 ERROR (MainThread) [homeassistant.components.hassio.handler] Timeout on /supervisor/ping request
2018-04-28 16:41:23 ERROR (MainThread) [homeassistant.components.hassio] Not connected with Hass.io
2018-04-28 16:41:23 INFO (MainThread) [homeassistant.setup] Setup of domain hassio took 15.0 seconds.
2018-04-28 16:41:23 ERROR (MainThread) [homeassistant.setup] Setup failed for hassio: Component failed to initialize.

lights show fine though and are switchable. When they're not shown as unavailable while in fact they are (App, api, etc)

[balloob]

This happens if the Hue bridge is not able to respond to our query within the set time. Which is very rare. Is your Home Assistant instance with bad wifi connection and thus cannot always reach the hub ?

[Mariusthvdb]

No, everything is wired. Even have the hue Hub and Pi on their own direct cable to the Asus router.

[sj3fk3]

I don't see the ERRORS anymore in 0.70 \o/

[Mariusthvdb]

well, i only wish it were true.
below (and sorry for the rather large snippet) is a very small bit of the log after having updated to .70.
what ever the reason is, it isnt solved at all, on the contrary...please let me know what to do to be able to help debug/solve this:
Ive purposely copied the error on the 'listener', maybe anyone can confirm this is related, or what this is about in the first place. I can only see these errors appear together, as if they are related to the Hue hub being connected and disconnected repeatedly:

2018-05-29 21:48:03 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:48:20 ERROR (MainThread) [homeassistant.components.light.hue] Unable to reach bridge 192.168.1.212
2018-05-29 21:48:23 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:48:26 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:48:28 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:48:35 WARNING (MainThread) [homeassistant.core] Unable to remove unknown listener <function async_track_point_in_utc_time.<locals>.point_in_time_listener at 0x6de3d390>
2018-05-29 21:48:35 ERROR (MainThread) [homeassistant.core] Error doing job: Exception in callback <function async_track_point_in_utc_time.<locals>.point_in_time_listener at 0x6de3d390>
Traceback (most recent call last):
  File "uvloop/cbhandles.pyx", line 49, in uvloop.loop.Handle._run
  File "/usr/lib/python3.6/site-packages/homeassistant/helpers/event.py", line 211, in point_in_time_listener
    hass.async_run_job(action, now)
  File "/usr/lib/python3.6/site-packages/homeassistant/core.py", line 253, in async_run_job
    target(*args)
  File "/usr/lib/python3.6/site-packages/homeassistant/helpers/script.py", line 95, in async_script_delay
    self._async_listener.remove(unsub)
ValueError: list.remove(x): x not in list
2018-05-29 21:48:35 WARNING (MainThread) [homeassistant.core] Unable to remove unknown listener <function async_track_point_in_utc_time.<locals>.point_in_time_listener at 0x6e1356f0>
2018-05-29 21:48:35 ERROR (MainThread) [homeassistant.core] Error doing job: Exception in callback <function async_track_point_in_utc_time.<locals>.point_in_time_listener at 0x6e1356f0>
Traceback (most recent call last):
  File "uvloop/cbhandles.pyx", line 49, in uvloop.loop.Handle._run
  File "/usr/lib/python3.6/site-packages/homeassistant/helpers/event.py", line 211, in point_in_time_listener
    hass.async_run_job(action, now)
  File "/usr/lib/python3.6/site-packages/homeassistant/core.py", line 253, in async_run_job
    target(*args)
  File "/usr/lib/python3.6/site-packages/homeassistant/helpers/script.py", line 95, in async_script_delay
    self._async_listener.remove(unsub)
ValueError: list.remove(x): x not in list
2018-05-29 21:48:35 WARNING (MainThread) [homeassistant.core] Unable to remove unknown listener <function async_track_point_in_utc_time.<locals>.point_in_time_listener at 0x6d7b7810>
2018-05-29 21:48:35 ERROR (MainThread) [homeassistant.core] Error doing job: Exception in callback <function async_track_point_in_utc_time.<locals>.point_in_time_listener at 0x6d7b7810>
Traceback (most recent call last):
  File "uvloop/cbhandles.pyx", line 49, in uvloop.loop.Handle._run
  File "/usr/lib/python3.6/site-packages/homeassistant/helpers/event.py", line 211, in point_in_time_listener
    hass.async_run_job(action, now)
  File "/usr/lib/python3.6/site-packages/homeassistant/core.py", line 253, in async_run_job
    target(*args)
  File "/usr/lib/python3.6/site-packages/homeassistant/helpers/script.py", line 95, in async_script_delay
    self._async_listener.remove(unsub)
ValueError: list.remove(x): x not in list
2018-05-29 21:48:37 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:48:40 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:48:48 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:49:15 INFO (MainThread) [homeassistant.components.light.hue] Reconnected to bridge 192.168.1.212
2018-05-29 21:49:20 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:49:23 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:49:30 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:49:47 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:49:51 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting
2018-05-29 21:49:54 ERROR (MainThread) [homeassistant.components.light.hue] Unable to reach bridge 192.168.1.212
2018-05-29 21:50:00 ERROR (MainThread) [homeassistant.core] Timer got out of sync. Resetting

[balloob]

Your stack trace is unrelated to Hue. It has to do with a delay in a script/automation.

[jawilson]

Could this be related to #13389?

[Mariusthvdb]

Your stack trace is unrelated to Hue. It has to do with a delay in a script/automation.

ok thanks, i will try to weed the scripts. Is it regular (yaml) or python scripts Im looking for, or cant you say based on this trace.

Still, please do follow up on the flapping of the OP here, it's really there

[ViperRNMC]

Have the same error here, my hue system doesn't respons as fast as it did a few versions before (i think before the multiple bridges).

2018-06-11 22:47:19 ERROR (MainThread) [homeassistant.components.light.hue] Unable to reach bridge 10.0.0.212

[roflcoopter]

This started happening for me as well today. All wired connections. I have sub 20 devices so its not a whole lot.

[balloob]

We give the bridge 4 seconds to reply before marking it as unreachable. I wouldn't be surprised if there is another integration that is doing I/O inside the event loop which is slowing down the Hue.

People that are having this issue:

• Please list the platforms that you use besides Hue
• Try running your Home Assistant instance with just the Hue platform and nothing else configured. Does it still occur?

[ViperRNMC]

Try it coming week only with hue hub

list of platforms

• tado
• hue
• zwave
• mqtt
• influxdb
• buienradar
• nmap
• ring
• dsmr
• homekit
• braviatv
• samsungtv
• appletv
• pushover
• powned
• sabnzbd
• speedtest
• systemmonitor

[roflcoopter]

I reverted back to 0.70.0 and i still see this problem which is really strange since i did not get this error before. Whats interesting tho is that the error occurs EXACTLY 7 minutes apart for me.

If i post to /api//config and check the last_used timestamp its only Home Assistant who is calling the API so that rules out any other apps that might cause this

I can test disabling other components during the weekend hopefully, for now heres my platforms:

• hue
• zwave
• ios
• owntracks
• nmap
• mqtt
• rfxtrx
• systemmonitor
• xiaomi_aqara
• proximity
• radarr
• sonarr
• harmony
• broadlink
• google_assistant
• timer
• generic thermostat
• chromecast
• braviatv
• songpal
• arlo
• ffmpeg
• twilio
• facebox

I probably missed a few. Is there a good way to extract all of them?

[roflcoopter]

I turned off all other components and i still received the error. I have a few osram lights that are constantly marked as Unreachable in the Hue app, could it be this that is screwing with the integration?

[roflcoopter]

@sj3fk3 @hvbhome @balloob @ViperRNMC

Did you guys find a solution? This practically makes Home Assistant unusable for me.

[ViperRNMC]

No solution here, also try with only hue without succes. But i also have trouble again with history graphs.

[Mariusthvdb]

No, unfortunately this keeps bugging the great support and integration for Hue we had before the change to Asyncio. As a matter of fact I've just had to change some automations to forgo on the state change of hue lights and reachability, since it keeps 'flapping', to stay on topic...

@balloob suggested before in this thread my issues had nothing to do with Hue. Unfortunately I have been able to establish it in fact has.
Restarting Hassio most of the time kills connection to Hassio itself, and takes it out of the side panel. If I continue to keep Hue alive during restart, through the Hue App, Hassio comes back in fine...

Im surprised no more action is taken to resolve the issue, many people are reporting on in a variety of details and issues, and asking for a solution. Would be glad to be of assistance in tracking this down , since, as you say, this is a major letdown for such an industry leading tool as Hue is.

Made a few rest sensor showing reachability, and this directly show the 'flapping' in the frontend...
Hopefully the restored support for scan_interval can solve the issue soon? Or isnt it restored for Hue after all? Not really sure.

[balloob]

Im surprised no more action is taken to resolve the issue, many people are reporting on in a variety of details and issues, and asking for a solution.

I ask for help in reporting the platforms people use so we can find a common platform. I don't see your platforms? Just a complaint.

I can't reproduce it. I also have limited time and can't just be chasing random reports. Feel free to spend your time on it instead of expecting other people to do so.

[sj3fk3]

The flapping as I reported (and I reported this bug) has been fixed a while ago. I think the problem with this bug report is, that it's fragmented. Maybe it's a good idea to report a new bug with specifics to reproduce. Because the bug I initially reported has been fixed for me.

[Mariusthvdb]

Feel free to spend your time on it instead of expecting other people to do so.

this is a bit unfair, to be honest, as you can see in this thread alone, spending time trying to find the solution to the breaking change of Hue integration is exactly what we've been doing...

great to see some functionality get back on track, hue.hue_activate_scene. Which is much appreciated!

tried to compile several issues for my own and others reference https://community.home-assistant.io/t/any-hope-to-get-hue-hue-activate-scene-back/52741/9?u=mariusthvdb

nevertheless i forgot to mention the platforms you asked for, sorry for that. This should be getting close:
my (sensor)platforms, which are all working as they should, (or at least so it seems..) and show up in the frontend without issues, except when annotated, which is in fact only on Asuswrt

api_streams
asuswrt (could be a culprit, having previous issues of false reporting home/away)
cast
file
filesize
fitbit
google
group
hue
iCloud
influxdb
ios
microsoft (unused)
mpd
mqtt
nmap_tracker
notify
openweathermap
owntracks
ping
plex
proximity
pushbullet
rest
speedtest
spotify
systemmonitor
time_date
tradfri
trend
tts (unused)
uptime
version
workday
yweather

hope this helps, glad to assist in any other way i could
cheers!

[roflcoopter]

Actually (and i am a bit embarrassed about hits) a reboot somehow fixed this for me!(?)

I do not really understand how, I am running Home Assistant in Docker on an Intel NUC Ubuntu 18.04.
I guess an update to some package on the host OS or something messed this up. (I didn't reboot for 10 days)

Hope this helps someone

[balloob]

@hvbhome sorry if I came off a bit strong.

So the reason I want the platforms, is that it could be that one of them, although it seems like it works well, is stalling our event loop and then impacting Hue.

I am going to close this issue as the original poster had the issue resolved and also rebooting seems to help (which might hint at a misbehaving platform actually). 0.72 will fix the flapping sooner, as it was accidentally scanning every 30 seconds instead of the intended 5.