Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Update to Supervisor 2022.08.03 breaks RaspiMatic addon

See original GitHub issue

Describe the issue you are experiencing

22-08-10 21:02:04 WARNING (MainThread) [supervisor.addons.addon] Watchdog found addon RaspberryMatic CCU is unhealthy, restarting…

RaspiMatic addon gets restarted every some minutes with this log entry.

What operating system image do you use?

ova (for Virtual Machines)

What version of Home Assistant Operating System is installed?

8.4

Did you upgrade the Operating System.

Steps to reproduce the issue

Update Supervisor from 2022.07.xx to 2022.08.03, no other updates done.

Anything in the Supervisor logs that might be useful for us?

22-08-10 21:02:04 WARNING (MainThread) [supervisor.addons.addon] Watchdog found addon RaspberryMatic CCU is unhealthy, restarting...

Anything in the Host logs that might be useful for us?

22-08-10 21:02:04 WARNING (MainThread) [supervisor.addons.addon] Watchdog found addon RaspberryMatic CCU is unhealthy, restarting...

System Health information

No response

Additional information

No response

Issue Analytics

State:
Created a year ago
Comments:5 (3 by maintainers)

Top GitHub Comments

1reaction

agnerscommented, Aug 12, 2022

So it seems this is a add-on issue which just cropped up with the new Supervisor issue. I’ll close this ticket then as the problem is being tracked in https://github.com/jens-maus/RaspberryMatic/issues/1907. Thanks for your help to get the bottom of this!

1reaction

mdegat01commented, Aug 11, 2022

@jens-maus In 2022.08.3 Supervisor now actually uses HEALTHCHECK statements in watchdog. If Docker tells us that a container for an addon is unhealthy then supervisor restarts it. Previously any HEALTHCHECK statements in addons were essentially ignored. Watchdog didn’t look at the health status of containers and we never presented that information to users.

So I guess my question would be why is the healthcheck marking the containers as unhealthy? If the result is intended to be ignored then it should be removed I would think.

EDIT: Reading the thread on the other bug looks like you already found that. Just to be clear though, Supervisor never looked at the output of HEALTHCHECK statements before. So its possible it has been returning false for a long time and no one noticed since the health status of containers was not presented to HA users or used by Supervisor.