L2ARC Degraded, Which Device?

[jxdking]

Just found out my L2ARC is degraded, as below. The question is how to find out which device is faulty. My L2ARC consists 2 NVME drives in stripe.
I did zpool clear couple days ago, but as far as I could recall, I didn't see any error on cache devices.

kevin@svr:~$ arc_summary -s l2arc

------------------------------------------------------------------------
ZFS Subsystem Report                            Tue Sep 09 09:22:48 2025
Linux 6.8.0-79-generic                                  2.2.2-0ubuntu9.3
Machine: svr (x86_64)                                   2.2.2-0ubuntu9.3

L2ARC status:                                                   DEGRADED
        Low memory aborts:                                            43
        Free on write:                                               169
        R/W clashes:                                                   0
        Bad checksums:                                                 4
        Read errors:                                                   0
        Write errors:                                                  0

L2ARC size (adaptive):                                         160.0 GiB
        Compressed:                                    86.9 %  139.1 GiB
        Header size:                                    0.5 %  842.8 MiB
        MFU allocated size:                            36.9 %   51.4 GiB
        MRU allocated size:                            60.0 %   83.4 GiB
        Prefetch allocated size:                        3.1 %    4.3 GiB
        Data (buffer content) allocated size:          98.4 %  136.8 GiB
        Metadata (buffer content) allocated size:       1.7 %    2.3 GiB

L2ARC breakdown:                                                   25.8M
        Hit ratio:                                     29.8 %       7.7M
        Miss ratio:                                    70.2 %      18.1M

kevin@svr:~$ sudo zpool status -xv
  pool: tank
 state: ONLINE
status: One or more devices has experienced an error resulting in data
        corruption.  Applications may be affected.
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: resilvered 6.45T in 15:14:53 with 0 errors on Sat Aug 16 04:01:50 2025
remove: Removal of vdev 2 copied 78.3G in 0h14m, completed on Fri Oct 28 20:00:07 2022
        1.04M memory used for removed device mappings
config:

        NAME                                  STATE     READ WRITE CKSUM
        tank                                  ONLINE       0     0     0
          mirror-0                            ONLINE       0     0     0
            ata-WDC_WUH721414ALE604_9JGJVK4T  ONLINE       0     0     0
            ata-WDC_WUH721414ALE6L4_Y5JT8RBC  ONLINE       0     0     0
        logs
          nvme0n1p5                           ONLINE       0     0     0
          nvme1n1p5                           ONLINE       0     0     0
        cache
          nvme0n1p8                           ONLINE       0     0     0
          nvme1n1p8                           ONLINE       0     0     0

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

PS: It is a home lab with used parts. I also have a full backup. I don't really mind if it falls apart.

[amotin]

I'd say just ignore it. It is reported if any errors were detected there since reboot:

    l2_errors = int(arc_stats['l2_writes_error']) +\
        int(arc_stats['l2_cksum_bad']) +\
        int(arc_stats['l2_io_error'])

While it may mean that something is not perfect there, but I would not be worried unless you experience any actual problems:

[jxdking]

Thanks for replying.

One question, in zpool status, it states there are permanent errors, but nothing is actually listed below the statement. Is that a false alarm?

[amotin]

L2ARC errors are never persistent and are always retried to the main storage. I don't know why you don't see the files, but if you believe your hardware is sane, I'd run zpool clear and then zpool scrub to find out what am I looking on.

[jxdking]

Got it. Thanks. Will try zpool scrub.