btrfs scrub only covers /mnt/external #124

New issue

Open

opened 2026-08-17 17:49:56 -04:00 by mysticalsoap · 0 comments

mysticalsoap commented

2026-08-17 17:49:56 -04:00

Owner

btrfs-scrub@ has exactly one timer instance: mnt-external. The LUKS root filesystem — all of /opt/docker, every app database, the restic repo's source data — and /mnt/support (btrfs, nvme) are never scrubbed. Checksums are only verified on read, so cold data can sit corrupt until the day a restore needs it.

Enable monthly btrfs-scrub@-.timer (root fs) and btrfs-scrub@mnt-support.timer
Wire results into the textfile-collector pattern (like smart.prom) so scrub errors and missed runs alert; a scrub that finds errors but tells no one is barely better than none

/mnt/disk1 (ext4, mergerfs member) has no checksumming at all — accepted for replaceable media, SMART covers the drive itself. Worth revisiting only if the pool grows (snapraid is the usual companion at 3+ data disks).

`btrfs-scrub@` has exactly one timer instance: `mnt-external`. The LUKS root filesystem — all of `/opt/docker`, every app database, the restic repo's source data — and `/mnt/support` (btrfs, nvme) are never scrubbed. Checksums are only verified on read, so cold data can sit corrupt until the day a restore needs it. - Enable monthly `btrfs-scrub@-.timer` (root fs) and `btrfs-scrub@mnt-support.timer` - Wire results into the textfile-collector pattern (like `smart.prom`) so scrub errors and missed runs alert; a scrub that finds errors but tells no one is barely better than none `/mnt/disk1` (ext4, mergerfs member) has no checksumming at all — accepted for replaceable media, SMART covers the drive itself. Worth revisiting only if the pool grows (snapraid is the usual companion at 3+ data disks).

mysticalsoap added the

general-admin

label

2026-08-17 17:49:56 -04:00

mysticalsoap referenced this issue

2026-08-17 17:50:17 -04:00

Stack analysis: feature gaps and divergence from other homelabs #116