Updates after updates things tend to break for me

Configuration
1

Hi!
Been happily using HA on a RPI 4 for maybe a year now and little by little added stuff. Maybe about 4-5 months ago things broke, had to downgrade (had a backup too) and then it went back to (almost) normal.
Unfortunately little by little I’ve lost recorder and other things. The main stuff I need still work and I am happy for that (HA is just not rebooting after the last update - I’ll have to take the Pi and look at the SD card data).
I’ve recently read the release notes and it seems sometimes things do break (“normally” - in the sense that we are warned) but as I didn’t read any release note I didn’t update anything.
So considering I have backups and my setup is rather simple should I just install a new hassio on my PI and reconfigure or will I face the same issues?
Path 2 is: should I stay with my current OS/config and try to address what is not working?
I have some spare time but I don’t really feel like spending 3 days on it :wink:
Advice welcome.
(FWIW I have BT thermometers, MI lamps, Smart plugs, UPS, Sonoff switch(es) and that’s about it)

Thank you.

2

Yes, you have to address the breaking changes with each release.

You really do need to do this.

1 Like
3

thanks… I thought there was some kind of magic that would … right!
Thank you for the light.

Let’s say it’s probably a good time to remove what I don’t need and just redo properly the things I use everyday!

Wish you the best… loved the answer!
Fred

4

So a little update. I did the “start from scratch” which was not so troublesome. I notice great improvements on my RPI 4:
CPU Use down to 1% (from 10%)
CPU Temp down to 42-43C (from 59C)
Disk Use down to 10% (from 27%)

And everything works again (I also got rid of things I don’t use). So double thank you!

Fred

5

It’s unlikely that’s because you started from scratch. There have been major improvements in how HA manages these things in recent releases.

6

oh yes… Definitely better HA. And from upgrade after upgrades I had a lot of things timing out, not being “discovered”, errors etc which probably created some extra CPU load (and services which I have now removed).