Just noticed that since around the time I updated from 2023.4.2 to 2023.4.4 on April 13th I now have funky jumps in my utility meter result graphs. Its random looking, nothing to do with restarting or anything I can tell.
Oh. Any idea why that would suddenly have become a problem at almost exactly the same time as the recent update?
I’ve had these sensors working flawlessly for months without this issue. And I think I had to add some default value because it spams errors/warnings otherwise.
I guess without getting too off-topic, is there a way to tell HA to stop spamming my log for battery powered ESPHome device being unreachable in deep_sleep?
I assume if there might be a line about whatever intermittently is happening with the input iotawatt sensor, its probably stuck among the mountains of expected can’t-connect-something-sleeping from the ESPHome sensor I have on battery power
That’s actually a constant…so that I don’t have to re-enter the same hard coded value a bunch of places when the power rates change. But yeah here’s a graph for good measure.
I have meters/graphs for the use over hour, day, week, month and wanted to simplify where to change it. All the graphs exhibit this same behavior fwiw…and one of my earlier posts had the “sourcecode” to them. They’re all identical except for the time-interval on the meters.
I think I may have found something! Happened to be looking at the IoTaWatt integration for a bit while watching TV and noticed everything go unavailable.
Then in the log I found Err: [Errno -2] Name does not resolve.
Now the question is…why? My local pfSense DNS server reports no errors, the hostname resolves just fine. And it apparently resolves “most of the time” on HassOS.