Being that it is 2023 DSI uses technology other than basic HTML for its web presence.

Your browser is not enabling JavaScript

To truly use our website this will need to be corrected.
Or, use an alternative browser.

DSI Globe Logo




DSI -> University (Help you do it) -> Hardware

Proxmox Zfs Drive Failure Replacement

Finalized by: Stephen T on September 5, 2023

As of 8.0.3.

Proxmox does not support software raid outside of ZFS. ZFS was developed by Oracle for OpenSolaris and is an advanced file system featuring pooled storage, snapshots, data integrity with auto repair, RAID-Z, and supports up to 256 quadrillion zettabytes of storage.

Running Proxmox it makes sense to consider and budget for data storage with redundancy and protection. Outside of a really good hardware raid controller with a battery backup for the controller the next best thing would be ZFS.

Drives will fail eventually. Here is how to deal with that.

For a failed drive first identify the failed drive and its serial number.

NODE -> Disks -> ZFS – Double click failed pool.

Also make note of the failed device.


Replace the faulted drive with a new one. Use the same size drive. Try and use the same make, model drive. If you can’t make sure it is the same size and speed.

NODE -> >_Shell

zpool status

zpool replace -f (pool name) (Old HD ID) (New HD ID)

So, it would be something like…

zpool replace -f ssd1tb 14732964903956238993 /dev/sdb

This will start the process of resilvering the new drive which just means to initialize, add the new drive to the pool, and synchronizing the data in the pool. Depending on the speed of the system, size of the pool, and the drive speed this should not take long. Typically 5 to 10 minutes.

You can check the status to see where it is at in the process.

When complete should look similar to this…