FoxESS has Cloud timeout fetching Device Detail #268
Comments
|
Have you changed your apikey recently ? If you can restart home assistant, wait 60 seconds when it has started up and then got to Settings, System, Logs and click the ‘Load Full Logs’ button - if you could copy and paste every log entry that relates to foxess here. |
|
I did change the api key this morning as an attempt to fix the issue. I generated a new key a replaced it on the configuration.yaml. These are the latest Logs: |
|
If you log into the Fox cloud, and go to the user profile, api management screen (as if you were going to request a new apikey) what are the values for your ‘Number of Calls per day’ and ‘Remaining calls for today’ |
|
I did an HA restart this morning, I let it run for 1 hour and still no readings. These are the logs since restart:
API calls: Before the restart I had over 600 calls left. FYI the Inverter Datalogger is connected and working fine I can see the power reading on Foxcloud.
|
|
I’m going to need some debugging enabled to see whats happening, your api-call count is low if it is only this integration making calls. Can you add this to your configuration.yaml near the top Then restart your home assistant again, leave it a couple of minutes and copy the logs again - there will be a lot more detailed information and you might want to obscure your apikey and inverter serial number as they will be shown. |
|
Here's the debug logs on the attached file. Looks like there's a lot api calls being sent, today's remaining api calls are about to be exhausted only have 6 left. |
|
Thanks, that's very strange - I can see 2 things, the first is that your inverter is reporting it is in alarm state, I can't see what the alarm state is because the get real time variables is failing with a response of 'query data failed' - i'm not sure why it would do that, it could be related to the inverter alarm. Because it fails to startup properly, it's skipping round this loop every minute and burning through 3 api calls (that will need fixing). But for now we need to understand why the inverter is in alarm - can you check by logging into the website, what alarm is your inverter reporting ? |
|
I have released a pre-release version 0.42-beta-3 which fixes the integration exhausting your api count when an inverter is in alarm. If you could upgrade to that as soon as possible, and even though your inverter is reporting a fault it won’t waste your api_calls - just need to understand what the fault is now. |
|
The fault is related to not having a battery connected to the inverter.
I've also installed version 0.42-beta-3. |
|
If you are on the old HACS, go to the integration in HACS, click the 3 dots in the top right corner and click 'Redownload' - it will show a dialogue box with the current version and above it is a switch for 'Show beta versions' - turn that on and it will show it. You are on a very old version of firmware, if you upgrade there is a switch within your inverter Settings, Feature 'BMS Set' that allows you to disable the no battery alarm - it's also possible the age of your firmware versions is causing a problem. Fox ESS did an update on the OpenAPI on 13th September and it's broken several Energy Stats app users (also uses OpenAPI) in the same way with an error 41200 'query data failed' - but this is the first one I have seen using this cloud integration. As it is an internal error to the API, you will need to ask Fox to investigate and fix the problem - I would suggest you email your local Fox service and tell them that for your inverter serial number the OpenAPI is failing with this error It could be your lack of battery, or old firmware versions - but obviously it used to work and it failed after Fox updated the API recently. If you are willing to email me your apikey and inverter serial number to b5djf@yahoo.com (you can change your apikey later), I will do some testing and try and find out which variable within the call is failing - it may be I can get something working for you whilst you wait for Fox to investigate the problem. |
|
Hi @NunoOliveira, Thanks for sending over your apikey and inverter SN. I've been running debug on your system and found that the corruption is in a certain field returned by the get device variables call - the variable in question is You might find that updating to the latest version of firmware on your inverter clears this problem - if you do that you can go into your inverter settings and tell it you don't have a BMS installed which will make the error clear anyway. But for now, i've created a pre-release v0.42-beta-4 which will avoid addressing that particular variable and your system will work. If you update to this release, and then add the If you update your inverter, you can test if it works by removing the 'restrict' switch and it will revert to its normal 'default' operation i.e. get all available variables. |
|
Hi @FozzieUK, Thanks for the help solving the issue. I've installed the new v0.42-beta-4 and it seems to be working fine now I'll try to get my inverter updated to a newer firmware version and I'll let you know what the outcome is. Cheers |
|
Ok glad you have it working again Please let me know how you get on after the inverter is updated - I suspect that when Fox did the update they have changed the data type of the current fault variable and as you had an active error it had the old string text in it which is causing them an internal server side error. Hopefully once you've been able to clear the 'old' text error it will work normally again 👍 |
|
now resolved with release v0.43 |
Describe the bug
The integration just stopped working 7 days ago and not able to pull any data.
Is was working fine for several months before and no changes were made with the exception of HA version updates.
Debug logs:
2024-09-19 10:39:28.019 WARNING (MainThread) [custom_components.foxess.sensor] FoxESS has Cloud timeout fetching Device Detail, will retry in 1 minute.
HA version:
core_2024.9.2
foxess-ha version:
0.42
The text was updated successfully, but these errors were encountered: