Permanent errors in boot pool - dodgy debug file?

[dazmatic]

Hi all,

I've been having recurring issues with errors in a particular file in my boot-pool.

Code:

action: Restore the file in question if possible.  Otherwise restore the
        entire pool from backup.
   see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-8A
  scan: scrub repaired 0B in 00:00:55 with 1 errors on Sun May 29 10:03:12 2022
config:

        NAME        STATE     READ WRITE CKSUM
        boot-pool   ONLINE       0     0     0
          ada2p2    ONLINE       0     0     1

errors: Permanent errors have been detected in the following files:

        boot-pool/ROOT/12.0-U8.1@2021-07-18-16:21:09:/boot/modules/openzfs-debug.ko.debug

Daft thing is, it doesn't affect the system as I understand it's an issue with the debug file within the debug.ko file.
It would also appear to be an error in the previous version also. This appeared with an earlier version which I've since deleted but has now appeared with the later version.

Not really sure what to do about this, I'm sure it can be fixed without the need to reset the system and restore from config, any ideas?

Thanks

 

[sretalla]

What hardware are you using for your boot pool?

Some models of SSDs use a controller that doesn't play well with FreeBSD and you'll find errors after each scrub because of the way TRIM is handled by FreeBSD and that controller together. (e.g. WD Blue/Green 3D NAND SSD)

You can either replace it with something like a Kingston or Samsung SSD (don't use those controllers in any of their models to my knowledge) or turn off TRIM (applies to all SSDs in the system, so maybe not an attractive option.

Since those errors are checksum, you may instead find that it's an issue with the cabling to that drive.

 

[dazmatic]

Originally it was some Adata p600 or something like that, 32gb SSD but it had this error so being a moron mirrored it to a it 64gb Chinese SSD.

Despite several upgrades, the error is persistent with that one file.

I'd rather try to replace the file with a known good than have to start again.

 

[sretalla]

It's going to be far simpler to just back up your config, reinstall and restore your config.

 

[dazmatic]

So I think I might have figured something out.
Looks like this error is actually in an old snapshot

Code:

config:

        NAME        STATE     READ WRITE CKSUM
        boot-pool   ONLINE       0     0     0
          ada2p2    ONLINE       0     0     2

errors: Permanent errors have been detected in the following files:

        boot-pool/ROOT/13.0-RELEASE@2021-07-18-16:21:09:/boot/modules/openzfs-debug.ko.debug
root@theexpanse[~]# zfs list -r -t snapshot -o name,creation boot-pool/ROOT
NAME                                             CREATION
boot-pool/ROOT/13.0-RELEASE@2021-07-18-16:21:09  Sun Jul 18 16:21 2021
boot-pool/ROOT/13.0-RELEASE@2021-08-28-14:15:12  Sat Aug 28 14:15 2021
boot-pool/ROOT/13.0-RELEASE@2022-01-13-11:29:34  Thu Jan 13 11:29 2022
boot-pool/ROOT/13.0-RELEASE@2022-02-24-14:30:55  Thu Feb 24 14:30 2022
boot-pool/ROOT/13.0-RELEASE@2022-05-12-12:50:04  Thu May 12 12:50 2022
boot-pool/ROOT/13.0-RELEASE@2022-05-29-10:28:26  Sun May 29 10:28 2022
root@theexpanse[~]# zfs destroy boot-pool/ROOT/13.0-RELEASE@2021-07-18-16:21:09cannot destroy 'boot-pool/ROOT/13.0-RELEASE@2021-07-18-16:21:09': snapshot has dependent clones
use '-R' to destroy the following datasets:
boot-pool/ROOT/default
root@theexpanse[~]#

Tried to delete the snapshot and got the above error, dependent clones.

So how do I go about deleting that snapshot?

 

[dazmatic]

Think I've got it - the Default boot-pool was still there back from last year, checked the date and it was older than the dodgy one.

Deleted the 'default' boot environment and now the snapshot has been deleted.

Running a scrub now, hopefully it'll be cleared.