API Call Limit #211
Comments
|
Just to follow on from this, adjusting the API calls from 30 seconds to 2 minutes fixed the issue and the errors have disappeared. Did VW do something with the API call limit? |
|
Where did you change the limits? I have similar problems since a few days. |
@bagges you need to open custom integrations in Home Assistant: /config/custom_components/volkswagen_we_connect_id/init.py Here you will find: update_interval=timedelta(seconds=XX) XX relates to the number of seconds. Default is 30. I changed mine to 90 and it seems to work fine. 60 seconds was hit and miss. |
|
Thanks, I'll give it a try. |
|
Still no luck, I am getting 403 errors. I have two different accounts integrated with this custom component. |
|
Periodically unavailable: Every five miniutes, for one to two minutes. With update_interval unchanged. ID4 Plug Connection State changed to connected |
|
@PeterAndreS I had the same issue, and fixed this by setting the poll rate to 90 seconds. It appears that VW have capped the rate or something. I also had issues during boot up, in that if I rebooted HA and it had polled shortly before, the integration failed to launch. The 90 second change appears to have fixed that. |
@bagges did you reboot HA after making the change? |
|
I think this messed up my app and tibber integration too. VW is really the worst. Their support is totally useless and they are changing these things constantly. Now trying with 120 seconds lets see how that works. |
|
@mnorrsken the mobile app is kit and miss, and this integration sometimes misses updates. @mitch-dc is this integration still being maintained? |
for me it is working at the moment. I restarted HA after the change but I took a while to start working again. |
|
I suspect vws limit has a way too small timeframe, causing my app and other integrations to fail in a random way, very frustrating, I sent vw a mail about this but i usually only get corp support back.13 okt. 2023 kl. 22:50 skrev Markus Backes ***@***.***>:
Still no luck, I am getting 403 errors. I have two different accounts integrated with this custom component.
@bagges did you reboot HA after making the change?
for me it is working at the moment. I restarted HA after the change but I took a while to start working again.
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
@mnorrsken i downloaded weconnect MQTT and integrated this today via node-red. It took me a good 8 hours to do 2 cars. For reference the minimum allowed API call limit is 300 seconds in this integration so would assume it should be the same for any other integration? |
|
For me the integration has fully stopped working for over a week now. I changed the polling to 120s few days agi, that did not help. Maybe I need to test with 300s. That is already 5 minutes so a bit too long to wait e.g. confirmation of climatisation starting, charging starting... But better than nothing 😅 |
|
For me, changing the polling to 90 sec interval does do the job. Used to fail retrieving data quite often, now I have a continuous success again. |
|
i just changed to 90 and it loaded on restart. it have been very unsable past two weeks i think, unsuable last 3 days. |
|
@t112013 fhange to 300 seconds and see how you get on. I changed to weconnect-MQTT and it seems to work / stable, but the minimum call rate is 300s. |
|
After updating requirements in manifest.json to latest weconnect.py (weconnect==0.59.1) the integration started working again, at least for the time being... |
|
Connection stabel again without any changes, sinde October 19 12:00 UTC+1. |
|
Changing update_interval to 90s has fixed it for me. |
Solved the issue for me, too. |
|
Looks like the pre release version has had the setting altered to 45 seconds |
|
After much experimenting, I finally have the unavailable status gone and connection seems stable - for 24 hours now. My most important use case for this integration is to give the car a command to start charging when electricity spot price is low, so I don't mind the looong 4 min interval. For those still wondering how to fix their setup, here is what I did. Others have explained it much better than I ever can, but here it is:
For stability review I used History view and particularly sensor for "state_of_charge". If the line is not continuous, you have an issue, i.e. the sensor is unavailable. If it is continuous, i.e. there is a numeric value at any given time, you're good. |
|
I am not using this integration but I've been following the discussion. I use another integration https://github.com/skagmo/ha_vwid with 30 second update interval and have no issues. The other integration has not all the signals and features which this has so perhaps it is not for all. I have modified it for my own purposes and added some functionality together with some other custom integrations so that I'm able to charge my ID.4 when electricity is cheap. My version is too specifically for my own purposes, so I cannot share it, and I do not have enough time to keep maintaining public version of it. It has been functioning well for almost two years already. I do not believe that the issue is at VW cloud side as it works fine for me without any issues with 30 s interval. You should take a look how often and what API calls WeConnect-python library is making. Perhaps there is happening something what is not needed. I used to use we_connect to login and my own functions to fetch data from API but changed to ha_vwid because it is much more simpler and straight forward. WeConnect-python library has originally been developed for another purpose and there is nothing wrong with that implementation, and Till Steinbach has done great job with it. The thing is that it is far too complex for HomeAssistant use. For Homeassistant it would be better and lighter just to call VW API and return back JSON response as is, and read needed data from JSON response, and ensure that only those API calls are made which are really needed for HomeAssistant use. That is how ha_vwid custom integration is doing it. |
|
My conclusion from this discussion is that the VolksWagen-We_Connect-ID Integration is quite processor demanding. And, dependent on the number of other processor running on the same processor, the update_interval needs to be adjusted to make it run smoothly. This explains why different users have different results. |
|
I switched to this integration from robinostlund/homeassistant-volkswagencarnet and expierence the same error. It seems that the standard Call has been changed to 120 seconds but it's causing the erros anyway. Maybe it is because I have two e-UPs and running this on 2 different HA installations? So maybe I have to change it to 240 seconds? |
|
@gmp91 the standard call in the we connect MQTT integration is 300 seconds. This works flawlessly. |
|
The issue is definitely caused by VW throttling back on the frequency with which requests can be made. |
Related to issues mitch-dc#211, mitch-dc#215 and mitch-dc#222 insofar as reducing the polling frequency reportedly avoids WeConnect API errors (e.g. 403, 429) when the servers are busy (API call rate limiting / throttling).
|
On the ID Buzz i use 90s and it works pretty well so far. However as the data in the backend seems to be very recent, I actually want to poll more often, while driving (45-60s), and less while stationary (120s or less?). Another workaround might be to define 1min during the day and 5mins during the night. Is there any way, to dynamically change the frequency? Because I would rather make this based on an automation instead of modifying the config and restarting HA. |
Hi,
Running the latest release, and getting an "unknown error" when logging into the API.
Anyone with the same issue? Only happening since adding a VW Up to my account.
Logger: homeassistant.components.device_tracker
Source: helpers/entity_platform.py:359
Integration: Device Tracker (documentation, issues)
First occurred: 19:39:06 (1 occurrences)
Last logged: 19:39:06
Error while setting up volkswagen_we_connect_id platform for device_tracker
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/update_coordinator.py", line 290, in _async_refresh
self.data = await self._async_update_data()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/src/homeassistant/homeassistant/helpers/update_coordinator.py", line 246, in _async_update_data
return await self.update_method()
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/config/custom_components/volkswagen_we_connect_id/init.py", line 53, in async_update_data
await hass.async_add_executor_job(_we_connect.update)
File "/usr/local/lib/python3.11/concurrent/futures/thread.py", line 58, in run
result = self.fn(*self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/weconnect/weconnect.py", line 193, in update
self.updateVehicles(updateCapabilities=updateCapabilities, updatePictures=updatePictures, force=force, selective=selective)
File "/usr/local/lib/python3.11/site-packages/weconnect/weconnect.py", line 202, in updateVehicles
data = self.fetchData(url, force)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/weconnect/weconnect.py", line 388, in fetchData
raise RetrievalError(f'Could not fetch data. Status Code was: {statusResponse.status_code}')
weconnect.errors.RetrievalError: Could not fetch data. Status Code was: 403
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/entity_platform.py", line 359, in _async_setup_platform
await asyncio.shield(task)
File "/config/custom_components/volkswagen_we_connect_id/device_tracker.py", line 29, in async_setup_entry
await coordinator.async_config_entry_first_refresh()
File "/usr/src/homeassistant/homeassistant/helpers/update_coordinator.py", line 262, in async_config_entry_first_refresh
raise ex
homeassistant.exceptions.ConfigEntryNotReady: Could not fetch data. Status Code was: 403
The text was updated successfully, but these errors were encountered: