Closed brucedavidson closed 10 months ago
Hi, probably it writes shortly a 0 until the data is fetched and then it provides the correct data. Have you an one phase or three phase system?
Could you please also check your logs, if there are any relevant entries like that a request took longer than 5 seconds?
Hi, probably it writes shortly a 0 until the data is fetched and then it provides the correct data. Have you an one phase or three phase system?
I have a one phase system. :)
Could you please also check your logs, if there are any relevant entries like that a request took longer than 5 seconds?
Certainly - will check the logs now and confirm.
Could you please also check your logs, if there are any relevant entries like that a request took longer than 5 seconds?
I can see in the logs multiple instances through the day of the following:
'Max retries exceeded with url: /production.json?details=1 (Caused by NewConnectionError' 'ERROR:root:--> fetch_meter_stream(): 5 errors accured. Stopping driver...' ' WARNING:root:--> fetch_production_historic(): HTTP request took longer than 5 seconds: 21.285921 seconds'
Try to change the 0
in this line
to None
like this
'/Ac/Energy/Forward': {'initial': None, 'textformat': _kwh},
Then restart the driver by launching bash /data/etc/dbus-enphase-envoy/restart.sh
. Tell me if this solved the problem.
I found no other evidence what else could cause this. Have you maybe an InfluxDB and Grafana where you could monitor the value?
Thank you so much for this package. Works super well and saves me buying an ET112 🙏🏻.
One small issue - if / when the script is disconnected from the Envoy and then gets reconnected (assume bad WiFi on Envoy), it seems to load the EnPhase system’s full (lifetime) production values (27kwh in my case). It then reflects in VRM as if that 27kwh was produced in that moment in time. It does this every time it gets disconnected and reconnects (as a result of poor WiFi this has happened 5 times today). As a result, my solar production and values for today are wildly inaccurate (over 150kwh of production being reported where only around 30kwh has in reality been produced).
Would be awesome if it might be possible to fix this bug.