Hey everyone.
I have a 10TB unraid system that I started backing up to a seperate system with a 22TB drive. The repository is a sftp provider.
After the first few days it was done, and from then I ran a backup every 6 hours with “low” retention.
I mistakenly also backed up some backup files from my computers that write to the unraid system.
So my drive filled up and I had no spare room to manouver, even after deleting logs and whatnot.
I did the bad thing and deleted some kopia backup files to free up space.
Ever since then, I’m getting errors during a full maintanance run. “quick” runs have no errors and backups still run smoothly.
I tried fixing this with everything I found online, from
kopia snapshot fix invalid-files --commit
or KOPIA_IGNORE_MAINTENANCE_REWRITE_ERROR=1
Didnt help, but thats not the point of the problem.
My problem is that yet again my drives are filling up for no apparent reason.
I again ran kopia maintenance run --full after attempting snapshot fix again and it fixing a few blobs.
However, the output is as follows:
GC found 2017937 unused contents (9.1 TB)
GC found 753348 unused contents that are too recent to delete (3.2 TB)
GC found 1755085 in-use contents (4.7 TB)
GC found 197 in-use system-contents (254.2 KB)
GC undeleted 4516 contents (20.8 GB)
it then does something (?) and will continue writing more data.
I then stopped the command and tried it with safety=none
(I know for sure there is nothing else writing backups so should be fine?) which does the same thing.
I even deleted my placeholder file (that I created after the last "incident”) which bought me 100GB, but Kopia filled that up quickly and now I’m at 3.4GB space left!
Before I run into the same issue I cancelled the command… and thats the current state.
As it seems the issues where actually fixed, as maintanance did not throw an error, and kopia snapshot verify also says everything is good.
What should I do now? Maintanance just keeps writing more data