Hey all,
I am running a 4 x RAIDZ2 VDEV. Each disk is 16TB.
I’ve purchased an additional 16TB disk and extended the VDEV, which should increase the capacity from 28.1TB to 43.6TB, however the capacity has remained the same.
When I go to manage devices, it shows all 5 disks are online and that there are no ZFS errors, and the storage pool is reported to be healthy, but the extra disk space is not showing up.
Could someone please provide me with some assistance?
Hmm so it looks like it is still in progress and estimated to take around 20 hours, am I correct in assuming the extra space will show up once this is completed?
Even though steps have been taken to speed up the raid vdev expansion, 20 hours seems about right for a 16Tb drive (depends on amount of data to be written). Give it the day before worrying.
Some extra space will show up in the GUI once expansion is complete, but there’s an upstream issue with space reporting after raidz expansion. The extra space is there for use.
Hey, I can confirm some extra space has shown up now that the pool expansion has finished.
I have 5 x 16TB disks, meaning 2 are for parity and the other 3 for usage space.
3 x 14.55 = 43.65. So I should have a total of 43.65TB usable, where as the capacity shown is just 35.11TB.
Do you know how I could go about addressing this? I have daily snapshots enabled, does this have anything to do with it?
EDIT - I missed the comment directly above mine. How can I verify the space is actually there to use? My SMB share for example shows just 29.9TB usable space.
Short of recreating the pool without using raidz expansion, there is no solution.
No.
sudo zpool list in the shell will provide better information.
Note that the extra space is there, and can be used; what’s borked is accounting and reporting.
You may wish to look at the in place rebalancing script: zfs-inplace-rebalancing/zfs-inplace-rebalancing.sh at master · markusressel/zfs-inplace-rebalancing · GitHub
This is what I get from: zpool list
NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT
MainStorage 72.8T 18.7T 54.0T - - 0% 25% 1.00x ONLINE /mnt
This is what I get from: zfs list
NAME USED AVAIL REFER MOUNTPOINT
MainStorage 9.08T 26.0T 407K /mnt/MainStorage
Has reputable is that rebalance script?
I understand you are saying that the space is there, but hopefully there is some way to verify this?
The rebalancing script will force ZFS to rewrite old data to the new raidz stripe width, which will reclaim some space. It does not preserve snapshots, but is not known to cause data loss.
Rebalancing, however, will NOT fix space reporting. As discussed here, the issue is deep into the OpenZFS code.
