I just checked the scrub status (btrfs scrub status /data) for an ongoing scrub and see this:
total bytes scrubbed: 19.68TiB with 144 errors
error details: csum=144
corrected errors: 0, uncorrectable errors: 144, unverified errors: 0
This is on one of my backup NAS, and a few backup jobs were run while the scrub was ongoing, if that might have anything to do with it. I've never seen scrub errors before.
Is there a way to see where these errors occurred? I would think an rsync incremental backup job (pull) would fix any corrupted files, or is there a reason I need to manually tell it to do a complete backup next time?
Is there a way to see where these errors occurred? I would think an rsync incremental backup job (pull) would fix any corrupted files, or is there a reason I need to manually tell it to do a complete backup next time?
There are some suggestions here maybe worth trying:
Thanks for the pointer. I was able to find the location of the errors. Turns out I have one pretty messed up share. It contains files that were deleted from the primary system some time ago and not some that were added, even though the log shows the incremental rsync backups completed just fine. It appears rsync was not identifying there were any changes. It appears that this happened some time before the previous scrub, so it's also concerning that nothing showed up in the main log. There probably are entries in the downloaded logs, but who looks at those except when something else points top a problem?
The scrub had also ground to a crawl, so I stopped it. I've renamed the old share and created a replacement, with the backup now ongoing. Once that completes, I'm going to delete the old share; and I hope that doesn't give me any issues.
The scrub had also ground to a crawl, so I stopped it. I've renamed the old share and created a replacement, with the backup now ongoing. Once that completes, I'm going to delete the old share; and I hope that doesn't give me any issues.
When the dust settles you might also want to run the disk test.
