Specific dataset snapshots corrupted, but pool scrub shows no errors?

TrueNAS General
3

The backup NAS is clean with no errors. My HBA has always been in IT mode. I’ve tested all of the files on the dataset and found no errors. This smells like a bug.

BACKUP NAS OUTPUT

└>[$] zpool status -v
  pool: POOL
 state: ONLINE
status: Some supported and requested features are not enabled on the pool.
        The pool can still be used, but some features are unavailable.
action: Enable all features using 'zpool upgrade'. Once this is done,
        the pool may no longer be accessible by software that does not support
        the features. See zpool-features(7) for details.
  scan: scrub repaired 0B in 18:57:28 with 0 errors on Mon Mar  2 20:42:31 2026
config:

        NAME                                      STATE     READ WRITE CKSUM
        POOL                                      ONLINE       0     0     0
          mirror-0                                ONLINE       0     0     0
            90859f98-9b79-4257-8f4a-e451f92a6ff0  ONLINE       0     0     0
            72ed84e6-b668-4f08-8496-88799d08ded8  ONLINE       0     0     0

errors: No known data errors

This user appears to have had the same issue, but maybe a little different.

I discussed this with Claude and went through some testing. Here’s what it had to say.

it’s a reporting bug, not real corruption. The <0x1> permanent errors in zpool status -v are phantom entries that ZFS is incorrectly flagging on encrypted datasets in TrueNAS SCALE 25.04. The actual data and metadata are completely intact.

The evidence is conclusive:

  • :white_check_mark: Scrub with dataset unlocked = 0 errors
  • :white_check_mark: All files readable
  • :white_check_mark: Replication working
  • :white_check_mark: Matches known bug pattern in 25.04
  • :white_check_mark: Errors vanish to “permission denied” when locked