Netatmo becomes unavailable and won’t reload

Third party integrations
1

Mainly the title.

I have an automation that attempts to reload the Netatmo integration if it becomes unavailable. It routinely gets stuck with a “failed to reload” message. Only restarting HASS fixes this.

Of course the larger problem is that the Netatmo integration becomes unavailable with great regularity.

2

There is some ongoing maintenance Cloud

That being said, I have no issues with the integration myself.

3

Is this relevant?

5

“Ongoing” sure is right. This issue has been happening for WEEKS.

Also almost every time I check the Netatmo site when the integration has failed, the site is up and reports no issues.

6

Haven’t tried this, and I don’t experience the majority of the symptoms the poster complains about (because I’m not trying to add anything new) but if some other response doesn’t offer a solution, I’ll try this even though it would show that the integration as it currently exists and is documented is just straight up broken.

7

Is the problem documented in GitHub where the developer might fix it?

8

This has happened to me twice in the past 6 months, one occasion their servers were down.

A restart of HA fixes it.

9

Fixes, or bypasses it?

Should it be robust enough to recover from unanticipated events?

An automation to 'fix' another automation failure? Am I the only one to see something wrong here?

Where are the logs and problem isolation activity? Return codes? Detailed description? Platform and network description?

Are you having credential issues?

You need to supply more information if you expect any progress towards finding a solution.

10

Well when the cloud servers were down nothing fixed it except for time and waiting for them to fix it.

The other occasion it took a HA restart as the reloading the integration did not seem to work.

Over the multiple years of using my Netatmo Weather Station I have not had very much trouble with it at all.

11

Twice is different from great regularity. Maybe the integration needs some love to make it respond to network unavailability in a more graceful fashion?

More critically, without further detailed information to debug the problem, this thread so far is anecdotal hearsay.

12

I have found Netatmo quite stable except the odd hiccup.

13

I experienced exactly the same issue that several other Home Assistant users have reported recently.

My Netatmo integration was working for a long time, although one of my cameras had stopped working correctly. As a troubleshooting step, I removed the Netatmo integration and tried to add it again.

The authorization process consistently failed:

  • Home Assistant redirected me to the Netatmo authorization page.
  • I logged in successfully.
  • Netatmo displayed the list of permissions and devices.
  • I clicked “I Accept”.
  • Nothing happened afterwards. The page simply stayed on the authorization screen and never redirected back to Home Assistant.

I tested multiple browsers, multiple devices, a different Netatmo account, and even enabled a Home Assistant Cloud (Nabu Casa) trial subscription. The result was always the same.

What finally solved the issue for me was creating an Application Credential in Home Assistant and using my own Netatmo developer application from dev.netatmo.com.

Steps that worked:

  1. Create a Netatmo application on dev.netatmo.com (in my case called “Homeassistant_Rien”).
  2. Copy the Client ID and Client Secret from the Netatmo developer portal.
  3. In Home Assistant, create a new Application Credential for Netatmo and enter the Client ID and Client Secret.
  4. Restart Home Assistant.
  5. Add the Netatmo integration again.
  6. When Home Assistant asks which credential to use, select the newly created application credential instead of the default Home Assistant Cloud authorization flow.

After doing this, the authorization completed successfully and all my Netatmo devices, including cameras and weather station, were added again.

This suggests that the problem may be related to the default Home Assistant OAuth client or authorization flow, while using a custom Netatmo developer application still works correctly.

14

Works, thank you so much! :star_struck:

15 
logger:%20homeassistant.components.netatmo%0ASource%3A%20components/netatmo/__init__.py:167%0AIntegration:%20Netatmo%20(documentation,%20issues)%0AFirst%20occurred:%204:38:08%20PM%20(7%20occurrences)%0ALast%20logged:%207:01:58%20PM%0A%0AError%20during%20webhook%20registration%20-%20400%20-%20Bad%20request%20-%20Invalid%20url%20parameter%20(21)%20when%20accessing%20'https://api.netatmo.com/api/addwebhook'%0AError%20during%20webhook%20registration%20-%20429%20-%20-%20API%20limit%20exceeded.%20This%20could%20be%20your%20Application%20limit%20or%20User%20limit.%20(26)%20when%20accessing%20'https://api.netatmo.com/api/addwebhook'%0A

logger:%20homeassistant.components.automation.reload_netatmo_when_unavailable%0ASource%3A%20helpers/script.py:2185%0AIntegration:%20Automation%20(documentation,%20issues)%0AFirst%20occurred:%206:53:52%20PM%20(10%20occurrences)%0ALast%20logged:%207:01:58%20PM%0A%0AReload%20Netatmo%20when%20Unavailable:%20Already%20running
16

I have posted one example of what’s going on in the logs, as far as I can tell.

This is happening to me much more often than “twice”. I’m restarting HASS about every other day to fix this problem.

As for my automation to attempt to reload the integration, I call that an attempt to fix something broken that I need to be able to rely on. So yes, something IS wrong here, and it’s the integration being unreliable.

Also the NEED for that automation should tell you that this integration is not reliable.

17

Works, thank you

18

Just had to restart HASS again today, only a few hours after restarting because of a HACS installation update. Because Netatmo would not reload.

As is, out of the box, this integration is not reliable.

19

Once again, my reload animation fails as Netatmo "failed to unload."

20