Charger Reconnects #307
Replies: 2 comments 53 replies
-
|
As well as network transport layer issues it could be loading on your HA server, router etc. Basically a localised issue to your environment causing the websocket to drop. You can expect a few reconnects over the course of a week or two on a solid setup. |
Beta Was this translation helpful? Give feedback.
-
|
Wow!!! A massive improvement. Latency values are sitting close to 0 now and the log looks like this... |
Beta Was this translation helpful? Give feedback.
-
|
I've only had one reconnect so far, and this is what lead up to it... The underlying network issue is obviously still there as I've frequently seen delays of up to 9 seconds between PING and PONG when originated by HA. When originated by the charger, the delay is always less than a second. On the occasion above, the delay was 10 seconds which looks to be the threshold where the charger determines it's too long and closes the connection. Sensor charger.Latency.Ping always sits at zero now, even though there are frequent delays of several seconds between PING (originated by HA) and PONG (returned by charger). So it seems that the latest changes broke that sensor. Now that the OCPP integration is working stably again, I'll try switching various things off around the house and track down the network issue. There's nobody else to complain that the internet is down for a while 😉 |
Beta Was this translation helpful? Give feedback.
-
|
Then again, another reconnect just happened and this is what lead up to it... There's no delay at all with those responses so it seems completely random when/why the charger decides to reconnect. At least we can now deal with reconnects without the comms becoming choked up. |
Beta Was this translation helpful? Give feedback.
-
|
the latency sensors are fixed in v0.4.14 |
Beta Was this translation helpful? Give feedback.
-
Awesome, thanks. Yes indeed, they are working perfectly now. |
Beta Was this translation helpful? Give feedback.
-
|
Earlier today I saw this...
Two hours have now passed since installing the update, and it looks very promising that this issue has been resolved, once and for all...
The 'PONG Latency' has also reduced dramatically since the upgrade...
|
Beta Was this translation helpful? Give feedback.
-
Just wondering if anybody else is experiencing frequent Charger Reconnects?
They can be as little as 10 minutes apart and occur for no apparent reason. I'm not sure if it indicates a problem, but has been happening with at least the last 3 releases. Here's a screenshot showing the relevant Lovelace card...
And here's an excerpt from the log, showing all entries relating to the Reconnect event...
2022-01-18 08:27:00 DEBUG (MainThread) [websockets.server] server < GET / HTTP/1.1
2022-01-18 08:27:00 DEBUG (MainThread) [websockets.server] server < Headers([('Connection', 'Upgrade'), ('Host', '192.168.1.6:9000'), ('Sec-WebSocket-Key', 'IUAswvKNnFJfZ127obKmNw=='), ('Sec-WebSocket-Protocol', 'ocpp1.6'), ('Sec-WebSocket-Version', '13'), ('Upgrade', 'websocket'), ('User-Agent', 'WebSocket++/0.8.2')])
2022-01-18 08:27:00 DEBUG (MainThread) [websockets.server] server > HTTP/1.1 101 Switching Protocols
2022-01-18 08:27:00 DEBUG (MainThread) [websockets.server] server > Headers([('Upgrade', 'websocket'), ('Connection', 'Upgrade'), ('Sec-WebSocket-Accept', '0FBLkyLz9+vDkI1NTfD4ipEjFo8='), ('Sec-WebSocket-Protocol', 'ocpp1.6'), ('Date', 'Mon, 17 Jan 2022 19:27:00 GMT'), ('Server', 'Python/3.9 websockets/9.1')])
Before anybody asks, the WiFi signal to my Pulsar Plus is shown as 'High' on the Wallbox Portal, or 3 bars on the app, so I'm certain that's not the cause.
Beta Was this translation helpful? Give feedback.
All reactions