Hassos + InfluxDB = high cpu usage

I have read up a little bit on this, but still can’t find a solution to make influxdb useable again on my pi4 with hassos. Background… the pi4 has been running fine for over a year, influxdb included. I went about a month without logging in to HA for maintenance, and when I did I noticed influxdb contunuously using a lot of CPU (~50%). I have read about large db’s and 32/64bit, but this is hassos. I also tried the INFLUXDB_MONITOR_STORE_ENABLED = false config which failed to work for me. My instance is not crazy busy as far as data rates… normal small home with power sensors etc setup so nothing is flooding HA. As I said, it has worked fine for years… and now this seems to have snuck in somehow.

I still have the issue wit I saw this post, but figured I may need to make a new one to get any help.

Anyone else figure out how to fix this with hassos?

1 Like

Hi,
I have also an Rpi4, and I´m “batteling” with this issue for a longer time, but I was ignoring it as it was not causing that much an issue. But I wanted to get rid of it.
By any change, are you running glances addon from within HASS, or somewhere else and you just send the data to influxdb?

What I did, and I also read about it somewhere here on the forum, I have disabled the Glances Integration.
Let me get more clear on this:
Im running glances, and sending this data to HomeAssistant to display the current status (cpu load, memory, and free disk space) but I have disabled glances to be recorded in the recorder, history or logbook. In exchange, as I wanted to have this data to look back, I configured the glances addon to send data directly to influxdb, into a different database.

What helped for me was, to disable the “export” of the Glances data to influxdb, restart glances addon, re-enable the export of glances data to influxdb restart addon once more, and afterwards my load dropper from averaging 7 to 2.
I think you can see what InfluxDB is in this graph… :smiley: influx_usage

Running HASS in Docker, Supervised, RPI4 4GB RAM, 64bit OS (as I tried 32 → 64bit which did not help )