Something very weird happened to my account... anyone else?

[shill]

A few days ago when I opened the Tesla app on my phone I got a message that the vehicle XXXXXXX (I didn't screenshot this, wish I had - pretty sure it was the internal ID they use and therefore I use) was no longer on my account, and promptly signed me out. I tried signing back in a few times and it failed... Eventually I got back in and things seemed fine, but when I got home and checked on HS, I found the updates were all failing. A quick check in Postman and I saw that this primary key had changed!

I then added a quick option to see/change the ID on the root device page (along with a warning that you better not mess with that if you don't know what you're doing). That got my updates going again, but during my testing, I discovered it fouled all my events, since I chose to use the vehicle ID as the key indicating which vehicle you wanted to control And since HS has never provided any methods to iterate through events and their actions, there's no way for my plugin to propagate that change to all events.

Has this happened to anyone else? Trying to decide how much energy to invest in accounting for this possibility.

[will40]

All is good here. The only thing I have noticed in my HS3 log lately is quite a few of the following. It is slow to update HS with stats but doesn't seem like a major issue right now.

Tesla Vehicles

Web request failed: The remote server returned an error: (422) Unprocessable Entity. (422)

[shill]

Turns out they made a slight change behind the scenes, thankfully other developers also ran into this so the fix was a known, simple thing. I just published 1.2.1; expect to see it in the updater on Monday or so.

[integlikewhoa]

I'm on 1.2.1 but my logs are getting flooded with

Tesla Vehicles

Web request failed: The remote server returned an error: (404) Not Found. (404)

Any ideas?

[will40]

I'm on 1.2.1 but my logs are getting flooded with

Tesla Vehicles

Web request failed: The remote server returned an error: (404) Not Found. (404)

Any ideas?

same here...something must have changed again for sure.

[shill]

I'm pretty sure that's what happened to me, too. They changed the ID the plugin relies on as a key and that's part of the URL...

I added a way to update your ID, but I realize that not everyone has a Postman project set up that lets you check the raw data! I need to add some logic to help automate this, but I was hesitant to invest in that effort if it was a one-time fluke on my account.

If you want to get things up and running again quickly, you should be able to fix most things by:

1. turn file logging to TRACE then go to the options page, which will call Tesla and get your list of vehicles.
2. Turn logging back down,
3. then look in the plugin log file for the JSON and find the "id" entry (don't be confused by "vehicle_id", which isn't what I use as the "Vehicle ID", since I use the "id" for that...).
4. Go to your root device for your car and there's a field where you can update the ID and hit save.
5. Restart the plugin.
6. You'll need to manually reselect your vehicle from any event actions (send navigation, schedule update, etc.) since HS doesn't provide any mechanism for me to update events

In the meantime, I have a plan for solving this and a way to prevent it from being an issue with events in the future, but that's going to take some time to write and test on my end.

[will40]

Thanks for looking into it. I believe I did as you said but the id # in the json is the same the plug in and the device root has already if I understand you correctly?

Sep-03 8:11:19 PM

Tesla Vehicles

Making request: (Authorization=Bearer **********************************, Content-Type=application/json, Accept=application/json, User-Agent=HomeSeer Tesla Plug-In/1.2.1) https://owner-api.teslamotors.com/ap...98686477xxxxxx

No rush on this. I see from your location you have more to worry about this week than this. Hope all is well in you're area.

[shill]

My ID has changed yet again as well, so I'm now seeing the 404. Clearly I picked the wrong field to make the key, and I've identified another value that hasn't changed throughout all this. I'm working on an update this weekend that will fix old devices and use the new key going forwards, but as before, if you use any of the plugin actions (Send Navigation, Schedule Update) in events, they'll have to be fixed again (and since they'll use this new key, that shouldn't be an issue after this).

[will40]

great, thanks. I did end up seeing a change in my ID the following day. I had to add it as another vehicle under my account and it works fine. Except it created another vehicle on device page and I cannot delete my old vehicle # in the plug in and errors in the log file so I left as is. I will try your update when you are done.

[shill]

1.3.0 has been submitted. (I posted about it here.) If you can get your plugin back to the state it was in before you added the 2nd vehicle, once you update to 1.3.0 it should take care of migrating everything but your any events that use "Navigate" or "Schedule Update". Those need to be fixed manually, unfortunately.

[will40]

Ok I updated the plugin to new version. I believe the PI vehicle changed ID again but all appears to work anyway. Only one same vehicle id # in the root device, the PI vehicle config, and the INI file. I can see the status and control the vehicle from HS3 with no issues that I can see.

Every 30 to 60 mins I get this error in log file if it means anything?

Tesla Vehicles			Referenced Vehicle Device ("matching vehicle id #") not found

Thanks for the support.

[shill]

I'm not seeing that in my logs, so I'm going to need some more info to help track that down. Can you set the plugin log file to TRACE, then stop/start the plugin and wait until you see that message in the HS log, then change the plugin back to whatever it was set to before (default is INFO I think) and download/email me the log file (steve@skware.net)?

[will40]

Sent...

[shill]

Thanks, pretty sure I found the issue - working on a quick bug fix.

[integlikewhoa]

Thanks, pretty sure I found the issue - working on a quick bug fix.

Thanks for forward to it. I also have that same error.