Pool Offline After Every Reboot

TrueNAS General
25

Am I correct that these drives where previously used in a TrueNAS Core system? And then you created a new pool in TrueNAS Scale?

To elaborate the steps for each option in more detail:

Option 1: Recreate the pool

This is useful if you don’t have any data in the pool. This does destroy all data in the pool.

  1. Export the pool in the UI. Do not select the “Destroy data on this pool”. Tick “Confirm Export/Disconnect” checkbox and press “Export/Disconnect”
  2. Go to the disks screen. Click on each disk you want to use in your pool (sda,sdb,sdc,sdd) and press the wipe button, it’s sufficient to select the “quick” method in the dropdown.
  3. After all four disks has been wiped, create the pool as you did initially. You’ll have no more issues rebooting

Option 2: Erase file system marker

This will fix the reboot issue without deleting any data.

Basically you do a careful, targeted wipefs on each partition. Your probe output showed that an old raid signature is detected.

You can first issue the following command to identify file system markers:

  • wipefs --no-act /dev/DISKPART

This should give you an output consisting of mostly zfs_member entries and a single linux_raid_member entry.

**DEVICE                                              OFFSET       TYPE              UUID                                 LABEL
xxx 0x1000       linux_raid_member cf174fb4-6f7b-943b-c217-ec7b7cbbea5c truenas:swap0
xxx 0x3f000      zfs_member        2099482002614675549                  boot-pool
xxx 0x3e000      zfs_member        2099482002614675549                  boot-pool

You’d then proceed by wiping the linux_raid_member signature.

  • wipefs --backup --all -t linux_raid_member /dev/DISKPART

But it is safer if you post the wipefs --no-act .. output here so we can verify that it indeed safe to erase.

1 Like