airq integration stops updating

The problem

I’m trying the AirQ integration on the dev branch. I was able to set it up fine, and it started gathering and reporting data.

However, a few hours later (at 8:45 local time), it suddenly stopped updating and providing new values.

During the night (these times are in UTC), I found these in the logs - there’s nothing else after the start up:

2022-11-14 01:30:47.030 ERROR (MainThread) [homeassistant.components.airq] Timeout fetching airq data
2022-11-14 01:30:57.142 INFO (MainThread) [homeassistant.components.airq] Fetching airq data recovered

However, the airq then continued reporting data until 7:45 UTC. The only error message in the logs which matches this is a timeout from the telegram integration, which seems unrelated. The AirQ itself is still reachable on its web interface and via the app.

Manually reloading the integration resolved this temporarily and the AirQ is now reporting values again, but obviously it would be preferable if this wasn’t necessary.

What version of Home Assistant Core has the issue?

Home Assistant 2022.12.0.dev20221113

What was the last working version of Home Assistant Core?

No response

What type of installation are you running?

Home Assistant Container

Integration causing the issue

AirQ

Link to integration documentation on our website

https://next.home-assistant.io/integrations/airq

Diagnostics information

No response

Example YAML snippet

No response

Anything in the logs that might be useful for us?

Unfortunately not.

Additional information

AirQ Pro by CorantGmbH Firmware: 1.20.2.r6_D_1.79 Hardware: D

[l-mb]

It hasn’t occurred again. But I’ve also had to upgrade the Air-Q firmware to 1.80.0 (the new ones shipped with -rc1 which wouldn’t dim the LEDs during the night 😃 ), plus moved on to a more recent dev release from HA.

Let’s close unless I can reproduce with more logs, and hopefully on a stable HA branch too.

[Sibgatulin]

Thanks for the logs! I am assuming those records which state success: False are proceeded by another log entry, such as Timeout fetching airq data? Just wanted to comment on the ~15.5 seconds reported by all failed updates. It is most likely due to the default timeout of 15s set in aioairq.AirQ.__init__ and passed through aioairq.AirQ.get to aiohttp.ClientSession.get. Hitting this timeout raises aiohttp.ServerTimeoutError (derived from asyncio.TimeoutError) which then must be caught by DataUpdateCoordinator. As a design decision this timeout was added to curb unreasonably long (5min by default) and predestined to fail queries when device and the HA host are not connected to the same network. This timeout is, of course, arbitrary and should not be expected to apply universally. At this point, however, the integration neither changes this timeout nor even exposes it for configuration, which may be a sensible point of further improvement. I cannot, however, point a finger at what may be causing the longer delays in those requests which finally succeed.