Re: BTRFS errors following bad SATA connection

[Roman Mamedov <rm@romanrm.net>]

On Fri, 3 Jan 2025 09:21:45 +0100
Victor Banon <banon.victor@gmail.com> wrote:

> `find . -type f -exec cat {} >/dev/null +` returns over 5,000 entries 
> now. After `echo check > /sys/block/md0/md/sync_action`, `cat 
> /sys/block/md0/md/mismatch_cnt` returns over 3,900.
> 
> When is it time to call it quits and reformat everything?

When it stops mounting :)

As is, surely it is easier to restore 3900 files from the backup rather than
everything?

The problem is, it could be that the transid mismatch errors won't go away
even if you replace all the files, or you might not be able to do so. Attempt
to delete or otherwise manipulate some of them might fail with the same errors
in dmesg.

> By the way, once I inevitably restore from a backup, is there any risk 
> that I backed up invisibly corrupted files? I'm pretty sure my latest 
> complete backup was after the SATA issues had started.

Should not be, unless your backup system silently stores half-copied files
into the backup until the point it got an I/O error, and does not warn you of
the error, or you miss the warning.

-- 
With respect,
Roman