4 ms·

Btrfs is a different use-case. ZFS doesn't support different sized hard drives. I could also be wrong, but ZFS doesn't expand as easily, and you can't remove d

Btrfs is a different use-case. ZFS doesn't support different sized hard drives. I could also be wrong, but ZFS doesn't expand as easily, and you can't remove drives from it. Synology NAS just added support for Btrfs in their software

stewartbutler 10y ago
ZFS does support different sized drives, and you can add disks to a pool easily. It does not support different sized drives in a raidz (equivalent to an mdadm array), as far as I know, but you can mix raidz and singular drives in a pool, though that is a bit silly. It also does not support adding disks to a raidz, which is equivalent to growing a raid5 or raid6 array, which is admittedly a pain in the ass.
- mbreese 10y ago
  You can also upgrade an entire raidz vdev if you replace each drive in the vdev with a larger one (one at a time) and enable "autoexpand" on the zpool. It's a manual and time consuming process, waiting for each drive to resilver. Still an pain, but at least possible.
- Annatar 10y ago
  > It does not support different sized drives in a raidz % mkfile -v 64m /var/tmp/d0 /var/tmp/d0 67108864 bytes % mkfile -v 96m /var/tmp/d1 /var/tmp/d1 100663296 bytes % mkfile -v 128m /var/tmp/d2 /var/tmp/d2 134217728 bytes % sudo zpool create -f pool0 raidz /var/tmp/d0 /var/tmp/d1 /var/tmp/d2 Password: % zpool status pool: pool0 state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM pool0 ONLINE 0 0 0 raidz1-0 ONLINE 0 0 0 /var/tmp/d0 ONLINE 0 0 0 /var/tmp/d1 ONLINE 0 0 0 /var/tmp/d2 ONLINE 0 0 0 errors: No known data errors
  - stewartbutler 10y ago
    You can build it, but it is limited to the smallest device in the vdev, so the array will appear to only have three devices of 64m available. Though you can replace all smaller disks and grow it.
    - Annatar 10y ago
      That is not how RAIDZ works. Because every write is a dynamic width stripe, the capacity is an aggregate minus parity, not linear. The smallest device capacity applies to RAID1+0, which is mirroring, but it does not apply to RAIDZ.
4ad 10y ago
Of course that ZFS supports any sized drives in a pool. ZFS expands very easily, in fact, the user doesn't have to do anything, it automatically expands. Of course you can remove drives in ZFS. How else could you replace disks? I downvoted your post because every statement you made about ZFS is wrong.
- wtallis 10y ago
  You're describing operations that can be done on a pool, while conveniently ignoring that they can not be done to a single RAID array (vdev), which is the more useful comparison against btrfs. Yes, you can add drives of arbitrary size to a ZFS pool, but when put into the same array, they all get treated as the size of the smallest one. Yes, you can expand a ZFS pool but only by adding a new array or by incrementally upgrading all the drives in the new array. Yes, you can remove drives, but only for replacement purposes; you can't do an in-place rebuild to make the reduced number of drives the new normal instead of degraded mode.
  - simoncion 10y ago
    I know nothing about ZFS, so sorry if these are dumb questions. Can you have an array that contains a single disk? And if you can, then * Can you add as many single drive arrays to a pool as you like? * Can you migrate data off of a given single-drive array on to other single-drive arrays in the pool and then permanently remove the array from the pool?
    - lmz 10y ago
      Yes, yes, no. Plus a single disk array has no redundancy.
      - X86BSD 10y ago
        Minor nit: depending on your definition of redundancy, you may set copies=n, on a single vdev pool. You won't get hardware redundancy but will get data redundancy.

stewartbutler 10y ago

ZFS does support different sized drives, and you can add disks to a pool easily. It does not support different sized drives in a raidz (equivalent to an mdadm array), as far as I know, but you can mix raidz and singular drives in a pool, though that is a bit silly. It also does not support adding disks to a raidz, which is equivalent to growing a raid5 or raid6 array, which is admittedly a pain in the ass.

mbreese 10y ago
You can also upgrade an entire raidz vdev if you replace each drive in the vdev with a larger one (one at a time) and enable "autoexpand" on the zpool. It's a manual and time consuming process, waiting for each drive to resilver. Still an pain, but at least possible.
Annatar 10y ago
> It does not support different sized drives in a raidz % mkfile -v 64m /var/tmp/d0 /var/tmp/d0 67108864 bytes % mkfile -v 96m /var/tmp/d1 /var/tmp/d1 100663296 bytes % mkfile -v 128m /var/tmp/d2 /var/tmp/d2 134217728 bytes % sudo zpool create -f pool0 raidz /var/tmp/d0 /var/tmp/d1 /var/tmp/d2 Password: % zpool status pool: pool0 state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM pool0 ONLINE 0 0 0 raidz1-0 ONLINE 0 0 0 /var/tmp/d0 ONLINE 0 0 0 /var/tmp/d1 ONLINE 0 0 0 /var/tmp/d2 ONLINE 0 0 0 errors: No known data errors
- stewartbutler 10y ago
  You can build it, but it is limited to the smallest device in the vdev, so the array will appear to only have three devices of 64m available. Though you can replace all smaller disks and grow it.
  - Annatar 10y ago
    That is not how RAIDZ works. Because every write is a dynamic width stripe, the capacity is an aggregate minus parity, not linear. The smallest device capacity applies to RAID1+0, which is mirroring, but it does not apply to RAIDZ.

4ad 10y ago

Of course that ZFS supports any sized drives in a pool. ZFS expands very easily, in fact, the user doesn't have to do anything, it automatically expands. Of course you can remove drives in ZFS. How else could you replace disks? I downvoted your post because every statement you made about ZFS is wrong.

wtallis 10y ago
You're describing operations that can be done on a pool, while conveniently ignoring that they can not be done to a single RAID array (vdev), which is the more useful comparison against btrfs. Yes, you can add drives of arbitrary size to a ZFS pool, but when put into the same array, they all get treated as the size of the smallest one. Yes, you can expand a ZFS pool but only by adding a new array or by incrementally upgrading all the drives in the new array. Yes, you can remove drives, but only for replacement purposes; you can't do an in-place rebuild to make the reduced number of drives the new normal instead of degraded mode.
- simoncion 10y ago
  I know nothing about ZFS, so sorry if these are dumb questions. Can you have an array that contains a single disk? And if you can, then * Can you add as many single drive arrays to a pool as you like? * Can you migrate data off of a given single-drive array on to other single-drive arrays in the pool and then permanently remove the array from the pool?
  - lmz 10y ago
    Yes, yes, no. Plus a single disk array has no redundancy.
    - X86BSD 10y ago
      Minor nit: depending on your definition of redundancy, you may set copies=n, on a single vdev pool. You won't get hardware redundancy but will get data redundancy.