You can't extend the existing raidz1 vdev by adding another disk but you can add another vdev to the pool to increase the pool's capacity. You'll need more than one additional disk if you want to retain redundancy. For example, you could use two disks to add a mirror vdev or 3+ disks for another raidz1 vdev.
Answer from Mike Fitzpatrick on Stack ExchangeAdd drive to ZFS pool
debian - expanding ZFS raid-z vdev with newly added disks? - Unix & Linux Stack Exchange
Adding new disks to Raidz | Proxmox Support Forum
ZFS - Expanding zfs pool with another one raid | The FreeBSD Forums
You can't extend the existing raidz1 vdev by adding another disk but you can add another vdev to the pool to increase the pool's capacity. You'll need more than one additional disk if you want to retain redundancy. For example, you could use two disks to add a mirror vdev or 3+ disks for another raidz1 vdev.
I know this question is old, but I just came across it. It looks like the ability to extend a RAIDZ one drive at a time is in the works! No ETA that I've seen though.
https://www.freebsdfoundation.org/blog/openzfs-raid-z-online-expansion-project-announcement/
I have four disks three of which are configured in raidz1-0. I'd like to expand my raidz using the fourth disk but it throws an error message I cannot figure out how to address. I run the following command zpool attach testpool raidz1-0 /Dev/sda It states "cannot attach /Dev/sda: can only attach to mirrors and top level disks ". I run Ubuntu 22.04 and freeBSD 13.1 but both behave in a similar fashion. Someone guide me how to successfully perform a raidz expansion. Thanks
Hi, I have build a storage for my backups (thus no high IO requirements) using old 3x 4TB drives in a raidz1 pool. Works pretty well so far: backup data is copied to the system, then a snapshot is created etc
Now I came to have another 4x 3TB drives and I'm thinking of adding them (or maybe only 3 as I currently have only 6 SATA ports on the MB) to the exiting pool instead of building a separate pool.
Why ? Because I'd rather extend the size of the pool rather than have to think about which pool I would copy the data to (why have /backup1 and /backup2 when you could have big /backup ?)
How ? I've read that a clever partitioning way would be to create 3TB partitions on the 4TB disks, then out of these and the 3TB disks create a 6x3TB raidz1. The remaining 3x1TB from the 4TB disks could be used as a separate raidz1, and extended in case I come to more 4TB disks.
Problem: the 4TB disks currently have a single 4TB partition on them, are in an existing raidz1. Means I would have to resize the partitions down to 3TB *w/o* loosing data.
Question: Is this somehow feasible in place ("in production"), meaning without copying all the data to a temp disk, recreating the zraid1, and then moving the data back ?
Many thanks
PS : it's about recycling the old HDDs I have. Buying new drives is out of scope