Recorder does not start, Recorder, History, Logbook, etc affected

The recorder does not start, therefore my history cant be setup.

I am using Maria DB. I will try to upload the error log below.

Setup failed for recorder: Integration failed to initialize.

09:36:21 – (ERROR) setup.py

Error during connection setup: (MySQLdb.OperationalError) (2002, “Can’t connect to MySQL server on ‘core-mariadb’ (115)”) (Background on this error at: Error Messages — SQLAlchemy 1.4 Documentation) (retrying in 3 seconds)

09:36:18 – (ERROR) recorder - message first occurred at 09:35:50 and shows up 10 times

Setup of recorder is taking over 10 seconds.

09:36:00 – (WARNING) runner.py

I have exactly the same problem, which arose withour apparent reason (i.e. not after an update, new integration, or any other apparent action form my side).
Hassio on an Intel NUC.
Happened on 31-1

Error log:

[plus a few lack of history related errors]

15:05:08 – (WAARSCHUWING) helpers/frame.py
The filesize platform for the sensor integration does not support platform setup. Please remove it from your config.

15:05:05 – (FOUT) Sensor
Setup failed for recorder: Integration failed to initialize.

15:05:00 – (FOUT) setup.py
Error during connection setup: (MySQLdb.OperationalError) (2002, “Can’t connect to MySQL server on ‘core-mariadb’ (115)”) (Background on this error at: Error Messages — SQLAlchemy 1.4 Documentation) (retrying in 3 seconds)
15:04:57 – (FOUT) recorder - bericht kwam voor het eerst om 15:04:30 en verschijnt 10 keer
Setup of recorder is taking over 10 seconds.15:04:40 – (WAARSCHUWING) runner.py


Did you come up with any solutions?

no :frowning: I started the whole HA from scratch and set up a scheduled back up

No I haven’t. I actually lost motivation and interest for a few months, after which I grabbed myslef together and build a new setup (HAOS on proxmox) with easier backup/restore options…
I have a suspicion that my database got corrupted by a very big dataflow from my ‘slimme lezer’ power meter hub. It generated a ton of data per second, which, I’m afraid all got written to the databse. Not sure of this, but my current implementation simply makes use of a simple cable between my power meter port P1 and the NUC, and that generates a lot less data.