Integration keeps losing connection

[Axsgranted]

Integration was working for a couple of weeks flawlessly. Was asked to reconfigure, which I did assuming maybe there was a 30 day token and now it asks me to reconfigure every 5 minutes.

[Axsgranted]

Error Log Details after authenticating successfully and then it "loses" connection:

Logger: custom_components.petsafe
Source: helpers/update_coordinator.py:343
Integration: PetSafe (documentation, issues)
First occurred: 6:53:07 PM (6 occurrences)
Last logged: 7:25:48 PM

Authentication failed while fetching PetSafe data: Client error '403 Forbidden' for url 'https://platform.cloud.petsafe.net/scoopfree/product/product' For more information check: https://httpstatuses.com/403

[dcmeglio]

I think they changed something or broke something. It might fix itself but if not it might be best to disable the integration so they don’t lock you out for too many failed attempts.

[cadwizzard]

I'm seeing this too. But I was asked to reconfigure again just once (15h ago).
I input email again and got the access code.
Now i have my email address showing twice in the integration, but only one with device and entity behind it.... but all entities are unavailable

UPDATE: i've since reloaded the integration from menu next to my email address - entities are now live again

[Axsgranted]

I'm seeing this too. But I was asked to reconfigure again just once (15h ago). I input email again and got the access code. Now i have my email address showing twice in the integration, but only one with device and entity behind it.... but all entities are unavailable

UPDATE: i've since reloaded the integration from menu next to my email address - entities are now live again

I've seen this exact behavior. I also did a reload and it worked for a little bit then stopped again.

[Axsgranted]

Update: I removed the integration and re-added it and it appears to be working normally again. But when it wasn't, how it handled the issues created the duplicate configuration entries @cadwizzard mentioned.

Update2: Spoke too soon. The entities went offline shortly after.

[Zeunas]

Are you guys still having this issue? It just started happening to me ( that I noticed)

[MivraMe]

Are you guys still having this issue? It just started happening to me ( that I noticed)

Yep same here

[Axsgranted]

Happening to me again after the latest updates

[dcmeglio]

I'm working on a fix for this it's just kind of hard to prove it fixes it since I haven't found a pattern to when the issue occurs.

[dcmeglio]

If you want to test out my fix and you know what to do, update the manifest.json for the integration to use petsafe 2.0.7

  "requirements": [
    "petsafe==2.0.7"
  ],

Since it's kind of a "wait and see" bug having more people let me know if they can see an improvement will help. After changing the file just restart HA

[ablack89]

I have been seeing the issue every few days recently - have applied your fix and will report back if I see it happen again.

[Axsgranted]

If you want to test out my fix and you know what to do, update the manifest.json for the integration to use petsafe 2.0.7

  "requirements": [
    "petsafe==2.0.7"
  ],

Since it's kind of a "wait and see" bug having more people let me know if they can see an improvement will help. After changing the file just restart HA

Trying this as well. The connection usually stays for 5 days or so, so it might be a few days before I can report back.

[Axsgranted]

Just lost connection and it wants me to re-authenticate even after setting the requirements version.

[ablack89]

I've also had it happen again today, with the updated version. The error in the log at the time is below.

`This error originated from a custom integration.

Logger: custom_components.petsafe
Source: helpers/update_coordinator.py:407
integration: PetSafe (documentation, issues)
First occurred: 17:48:33 (1 occurrences)
Last logged: 17:48:33

Authentication failed while fetching PetSafe data: Client error '403 Forbidden' for url 'https://platform.cloud.petsafe.net/scoopfree/product/product' For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/403`

[MivraMe]

Same for me, just lost connection, do we ping the server at a high interval so we are kicked out ?

[siobhanellis]

I just lost connection after reconfiguring today

[dcmeglio]

Darn happened to me today too. I'll keep investigating.

[dcmeglio]

I want to check if you're experiencing the same thing I am while yes I am getting a 403 that causes a reauth, it does seem the data is continuing to update? I'm wondering if there API is just a bit flaky and throws 403s sometimes and I just need to add retries. Seems like an odd thing, but it does appear my litter box is still updating even though I got that error.

[ablack89]

Yes, if I ignore the re-auth prompts the data keeps coming in for me.

[MivraMe]

No data for me... and it's seem to create a new device every time..

[dcmeglio]

@MivraMe the duplicate items is a different bug, I'll look into that, but this sounds like you did the reauth. It seems if you ignore the reauth, everything continues to work.

[MivraMe]

All my sensors are unavailable for my petsafe food..

[dcmeglio]

Right, as I said, you did the reauth workflow, correct? That is causing the bug I just filed on your behalf #18. If you delete the devices and reinstall it should begin working again.

[dcmeglio]

I've been testing a fix for this that so far seems good. If it's still working this weekend I'll release a new version

[dcmeglio]

I ran this for a week and didn't have any auth issues or data issues. I think their API just occasionally throws 403s so I added some retries to detect multiple errors before triggering a reauth can you guys try out 1.3.5 and let me know if it works? https://github.com/dcmeglio/homeassistant-petsafe/releases/tag/1.3.5

[ablack89]

Thanks, have updated and will report back if I see any issue.

[dcmeglio]

I haven't heard any further reports so I'm going to mark this as closed.