I am running HA on proxmox with a virtual disk of 80GB. I am using multiple add-ons and integrations and often the complete 80GB is used. I deleted the complete database from influx and home assistant and all the backups but the system is still using around 60GB of data.
Tried to find big files and folders but in total I do not get more than 5GB of data. Where is all other data stored en how can I find it?
Hope someone can help?
Do you have any Backups/Snapshots created? Look under Supervisor > Backups. They take up a LOT of space. They’re stored on disk under /backup.
Additionally, I’ve found that corrupted backups/snapshots can be orphaned. They won’t appear in the UI, but they’re on disk. These are MUCH harder to get to, as you’ll need console access to the OS that your HA container resides in (SSH via the Add-on, or via Terminal in the UI won’t cut it).
These are found in /mnt/data/supervisor/tmp.
I think it might be your last option. within the /mnt/data/supervisor/tmp folder i have this list of files/ folders, is it possible to see what data is corrupted or can I just delete all tmp folders?
Deleted all tmp folders and I reduced my instance with approx 20GB. Still works on large amount of storage but this looks reasonable!
thnx for the help!
Yep, my pleasure. I’m not sure when/if the “tmp” folder is cleaned up, I know it’s NOT cleared at a physical reboot. I’m tempted to build a cron job to just do it at reboot regardless, but just haven’t gotten there.
Wow! Seems to be a problem still. Just found 85% of my 500GB space filled up with those tmp-files.
Thanks for the hint.
Same here. I had to go to the developer docs to even figure out how to get root access to the underlying hypervisor. It required putting authorized_keys on a usb drive and plugging into the machine and running some commands. I have 25GB of my 100GB of available space in the tmp folder. You really shouldn’t be locking down a system so hard unless said system “just works” and doesn’t have common issues like this one.
It also appears that the home assistant OS doesn’t have any methods in place to clean out journal logs. I had 8GB of logs in the /mnt/data/logs folder. Here is a listing to help anyone who might be trying to figure out what it taking up all their storage.
8G /mnt/data/logs
18G /mnt/data/supervisor/backup (seem to be orphaned backups that don’t show in the UI)
5G /mnt/data/supervisor/addons/data/5c53de3b_esphome/cache (this can probably be cleared by cleaning the build files in the esphome UI)
For anyone else who needs to actually access your full storage:
hallo, via the terminal i cannot see anything under MNT , is that possible? i am using LS command and seems folder is empty , but i have the same issue of tons of GB of files should be somewhere, can someone halp me ? i am currently with ZERO space left on my 128gb ha blue , and should be 30/40gb free…
thanks for help
