Panda88

I see the same issue - not sure what causes it (yet) My new release in testing does not see the issue, but I have far fewer devices on my test platform Polls are not truly needed - MQTT updates when actions happen - short poll is mostly to send a heartbeat there are cases where an update sometimes takes longer so the state has not changed when the display updates on the Admin Console - in this special case the poll causes the display to update - Not if using programs this does not matter

Polls are not truly needed - MQTT updates when actions happen - short poll is mostly to send a heart beat there are cases where an update takes longer so the state has not changed when the display updates on the Admin Console - in this special case the poll causes the display to update - Not if using programs this does not matter

top is installed - it shows the issue Trying to figure out what causes this

I did not study the CPU usage in great detail - I am in the progress of changing the message handling approach - It used to be every device was waiting for input from a queue - I am changing that to start a thread when a message arrives - I think that will lower the CPU usage I cannot think of anything else causing a lot of CPU usage - unless you are using the delay timers

Each device is really a node server by itself - I'll see if there is something obvious wrong but it may take some time to get to the bottom of it

I'll take a look How many devices have you installed? Do you know if Memory increases over time?

I think I fixed it - I made a wrong commit about 12 days ago and that code caused this to happen I am working on redoing the message handling and hope to have a new and improved version soon

Can you send log with Debug logging enabled?

Yes - if you can stop and start and get the log file I can check to see what is happening - it looks like it thinks there is connection to both cloud and local, but it looks like cloud is not working correctly there is no real documentation - even on github, but you are welcome to check out the code there. I am just doing this for fun, and the charging is mostly to give some extra income to universal-devices (they get half of what you pay).

What is your setup - are you running both cloud and local - or local only? -1 means no value retrieved? If running local only you need to run for 24hours before yesterday can be deployed - (it only reports current status). I'll take a closer look once I understand your setup (there is no initialization in the log you provided) Generator is an optional mode where a generator can also be enabled when battery runs out

I did notice a network error in the log as well 2022-11-19 07:13:48,998 Thread-1259 udi_interface ERROR TeslaInfo:pollSystemData: Exception PollSystemData: Powerwall is unreachable: HTTPSConnectionPool(host='192.168.1.71', port=443): Max retries exceeded with url: /api/meters/aggregates (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x8031acee0>: Failed to establish a new connection: [Errno 50] Network is down')) i'll try to see if I can find a way to reconnect if this happens

Looks like you are not looged into the local power wall - Can you log in from Browser? 2022-11-19 08:03:05,414 Thread-6 udi_interface ERROR TeslaPWController:tesla_initialize: local authenticated failed.

Can you send me a log with debug enabled It usually happens if system does not start correctly - it may be you need to power cycle your power wall during start up - to log into it locally - something they recently changed

I actually mean to remove the yolink node (delete it) and then reinstall it - If you replace 1 to 1 I think the programs will remain unaffected - I have tried with my own nodes but not purchased nodes Maybe you can contact Michel and see if he can reimburce you if you purchase again - I am sure it should be ok. Naturally, we should try to get to the bottom of what happened, but it will allow you to move forward

The latest YoLink node server does support the speaker - not as elegant as I would like, but there are some limitations with regarding to text in the AC (you basically need to predefine you text (announcements) and then you can pick them in the AC) -

did you reboot PG3 after the upgrade? I am not 100% this will work, but if you do not start AC - could you erase the node in PG3 - that should free up the node slot) and you can reinstall in the slot - I have done it before with success - you may need to refresh the program and save afterwards - but no editing should be needed

You do not see this option

In my case, there is an option 2/3 way up on the install page to install in existing slot - do you not see this?

It should be 0.6.15 - it shows in store as 0.6.1 but it is a bug in PG3 - when you enter the node you should see 0.6.15

Just released the YoLink server to the production store I finally found the issue causing the server to sometimes disconnect - YoLink has a limit if maximum 5 connections and I somehow did not release the connections so after 5 connections the node server can still talk to YoLink but it was ignoring the requests - I think I found the correct way to handle the disconnects - It has been running for 2 weeks now, and I have been trying what I can to break it but it keep coming back I also added support for the smart lock - It actually works well and is very reasonably priced I am charging 10$ for the node - mostly to provide some cash to UD - (they take about 1/2) - rest will be used to buy new YoLink devices and rest will be donated to a good cause - I do this mostly for fun My next devices will be the IR remote controller and the new dimmer switch

Did you try to run an update modules from the AC - if not try to contact Michel - he helped a similar issue I had

Sorry - it is an issue with having to update the version number multiple places (and remember to do it everywhere) when releasing the node. There are at least 3 places that needs to be in sync. BTW - you are correct - I did not update the run-time display of the node number to latest version I think PG3 compare that to the number in the node store to determine if an update is needed I did update the node number now - but I'll be looking to add a sleep parameter to the status node to update the car state so a new version will be required

Thanks - What about the node version besides purchase options in the developer page? I try to update both - My impression is it actually uses the node version for checking - you can show different version under purchase options as I understand - I may be wrong

Thanks It is a slightly different behavior than I would expect - I had expected the car to set online to False (and my code is working that way) - it almost looks like an issue connecting to getting data from the Tesla Cloud. The sleep feature is there indirectly with time since last update fields (assuming they work correctly) - but I can add a dedicated sleep state in the status node - with offline, sleep, online, and unknown states - does than make sense? I feel it is too much to say the node is not online - as it is after you wake the car - I see the connection as the indicator - not the car Do you have a little longer snippet of the log so I can see what happens next

Sorry - it is an issue with having to update the version number multiple places (and remember to do it everywhere) when releasing the node. There are at least 3 places that needs to be in sync. BTW - you are correct - I did not update the run-time display of the node number to latest version I think PG3 compare that to the number in the node store to determine if an update is needed I did update the node number now - but I'll be looking to add a sleep parameter to the status node to update the car state so a new version will be required