You did restore from a point in time before you had the corruption issues, right?
Yes. Tried to restore from 2 full backups - one a few days ago prior to upgrading to 7.2023, and over a month ago. Both are having slow boot up times to get the integrations and add-ons started. Been trying to reboot but it’s been taken an hour for everything to load, then it’s been crashing.
Try disabling all add-ons and start them one at a time. I see at least the VS code one is complaining about incompatibility.
I don’t see any errors in the logs with sqlite DB issues, is this completely unrelated?
Hard to say for certain. Try what I mentioned in the previous post.
When you delete the DB file, be sure to delete the WAL and SHM files too.
Thanks, trtying both recommendations. HA isn’t staying online long enough at boot. It just keeps loosing its connection, and I’m having to do a hard reset for it to come back online.
Really frustrating that I’m restoring it from a backup long before the corruption, but yet it will not fully boot.
I’ve tried doing a partial backup from just core, and not any of the addons but still continues.
I feel you. Do you have access to some other hardware to run it on it could be worth a try.
So I basically wiped and installed HAOS on a new SSD, same result. I have another spare Pi I can try on. Also I run Proxmox, but dont’ have any experience running Supervised.
But it does run as it should before you start restoring?
Yes, it runs snappy when I go through the initial login prompts to get setup. It’s when I restore, and reboot that everything craps out.
I dread having to run from scratch again after having run HAOS for over 2 years. Is there a simple way to restore - maybe with just the essentials and automations and start anew?
Never had to restore so I’m not sure, sorry.
No worries, appreciate you chiming in. I’ve dug into the forums, but haven’t found an identical scenario. I’m hoping anyone in the community has seen/experienced this.
You can take any backup-file and extract them(if they are not corrupt, it’s simple tar/targz files, keep on extracting until you come to the /data inside you find /config
Thank you, I did extract that and can access all files in /data.
Would the configurations be fine to get started to just copy and paste configuration.yaml & automations.yaml ?
I’m wondering what would happen if I copied all of /data folder except the home-assistant_v2.db files if that will help me boot into a stable system with the old integrations setup.
well you never said from which version you tried to upgrade from ? > to 2023.7 … so i assume your backups are from your “old” version, if it’s 2 years old, doubtful 
If it’s 2023.6 or slightly older, i suggest you install 2023.6 (or same again) again, just in case
Sorry for clarity, I upgraded from 2023.5. The installation process I’m following is flashing my SSD to OS 10.3, then restoring from there. Should I install from an older version of OS?
no, i think 2023.5 is fine ( if your backups are 2023.5 i suggest you install this, as your backups also are from this version … but pick a backup from before your problems started( a 2023.5 or .4 backup), and restore this, if it also fails you can install the integrations again, and copy/paste(configurations-files/dashboards etc one by one) ( i.e in general most, if not all yaml-files is straight forward to copy/paste
1 Like
Ok, I’ve gone as far back as restoring a full backup from 2022.12 but still the problem continued - slow boot after restore, integrations not fully starting (stuck in loading state), then eventually HA stalls and crashes. Meanwhile I was monitoring the Supervisor logs, but nothing indicative of the problem.
I’m waiting for the RPI to boot from another attempt to restore from full backup (think it’s been about 5 attempts with different backups).
If the same outcome persists, I’ll try your suggestion of - copy / pasting YAML one by one. Really appreciate your guidance. I’m guessing it’s some specific integration that’s causing everything to break but cannot pinpoint.