zfs raid1 does not boot with replaced failed disk

[rpereyra]

Hi all

Proxmox versión: 6.3-2

I've replaced a failed ssd disk on a Proliant DL380 Gen10, and seems all well syncronized .

--------------
root@pm4:~# zpool status
pool: rpool
state: ONLINE
scan: resilvered 698G in 0 days 00:43:11 with 0 errors on Sat Oct 23 12:22:43 2021
config:

NAME STATE READ WRITE CKSUM
rpool ONLINE 0 0 0
mirror-0 ONLINE 0 0 0
scsi-331402ec015377e62-part3 ONLINE 0 0 0
scsi-331402ec015377e63 ONLINE 0 0 0

errors: No known data errors
root@pm4:~#

-----------------

However, when I disconnect the disk that was in the previous raid and leave only the new disk, the server does not boot.

It only boots when I plug the disk from the previous raid.

I guess it's something to do with Grub. Can anyone give me any help with this issue?

thanks

 

[Dunuin]

You can't just replace the drive. You need to partition it first, copy over the bootloader and so on and let ZFS use only partition 3 and not the whole drive.
There was a guide on how to to that but proxmox changed the bootloader with PVE 6.4 so the guide you now find in the wiki/documentation sin't working for your PVE6.3 anymore. Wasn't able to find a old 6.3 documentation.

 

[rpereyra]

Thanks

I've follow this guide but seems something old:

https://edmondscommerce.github.io/replacing-failed-drive-in-zfs-zpool-on-proxmox/

 

[rpereyra]

My server seems using systemd-boot

Then I should run:

pve-efiboot-tool refresh

to syncronize boot loader on both drives ?

It's correct and safe ?

Thanks