My ZFS pool can't show up. ZFS import says no pools available to import.

[prolab]

I had a Proxmox VE server with 9 of 10TB hard drives. The 9 drives has been created as a raidz pool. It was running well for months, but recently I upgraded this server's motherboard (also with the cables to the hard drives). Then there were always READ/WRITE/CHECKSUM errors on some of the disks randomly when I check with zpool status. These errors could go away when I reboot the server or use zpool clear.

But today, after reboot, my ZFS pool can't show up. The pool's name is zfs52, when I try to import it, it shows I/O error:

root@pve52:~# zpool import zfs52
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.

root@pve52:~# zpool import
   pool: zfs52
     id: 8526622789643366080
  state: ONLINE
status: Some supported features are not enabled on the pool.
 action: The pool can be imported using its name or numeric identifier, though
        some features will not be available without an explicit 'zpool upgrade'.
 config:

        zfs52                                   ONLINE
          raidz1-0                              ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_QGG3AB2T  ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_QGGDS91T  ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_QGGE07HT  ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_QGGL1ZVT  ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_QGH5V04T  ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_X1G502KL  ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_X1G6H9LL  ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_X1G9TYHL  ONLINE
            ata-WDC_WD120EMFZ-11A6JA0_XHG0J1MD  ONLINE

Then I searched online, and tried these two commands trying to import it:

root@pve52:~# zpool import -f zfs52
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.
root@pve52:~# zpool import -Ffn zfs52
cannot import 'zfs52': no such pool available

Then it becomes wired that sometimes it says "I/O error", sometimes it says "no pools available". What's the reason? Is it because of the hardware (unstable hardware/unstable cable) or because of data corrupted? May I still recover the data? What to do?

root@pve52:~# zpool import zfs52
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.
root@pve52:~# zpool import zfs52
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.
root@pve52:~# zpool import -Ffn zfs52
cannot import 'zfs52': no such pool available
root@pve52:~# zpool import zfs52
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.
root@pve52:~# zpool import zfs52
cannot import 'zfs52': no such pool available
root@pve52:~# zpool import zfs52
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.
root@pve52:~# zpool import zfs52
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.
root@pve52:~# zpool import -Ffn zfs52
root@pve52:~# zpool import -Ffn zfs52
root@pve52:~# zpool import zfs52
cannot import 'zfs52': no such pool available
root@pve52:~# zpool import zfs52
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.
root@pve52:~# zpool import -a
no pools available to import
root@pve52:~# zpool import -a
cannot import 'zfs52': I/O error
        Destroy and re-create the pool from
        a backup source.
root@pve52:~# zpool import -D
no pools available to import
root@pve52:~#

 

[leesteken]

Maybe you have bad cables or connections or disks, because you got many (non-fatal) errors when you used the pool, and the redundancy/parity was enough to keep it going.
Maybe now finally, it cannot read the metadata (which was always cached) that is the start of the internal structure of the filesystem.
Maybe switching/replacing cables can help if the metadata (and checksums) are correct on (most of) the disks, but it could be written incorrectly because of bad cables/disks and then I don't know how to repair it.
There are some options with zpool import that make it scan for (older) metadata to repair and import (an earlier but consistent version of) the filesystem. You could try -F and -X, but please read man zpool import first.

 

[Dunuin]

And I think a raidz1 if 9x 10TB is a very bad idea. If one of your drives dies the resilvering might take weeks where the drives are tortured 24/7. Its likely that while doing the resilvering another disks might fail or cause errors and then you will loose your complete pool. With so much storage I would atleast use Raidz2 or maybe 3x 3 disk raidz1s striped together.