ZFS USED twice as LUSED

mailinglists

Renowned Member
Mar 14, 2012
641
67
93
Hi,

i noticed that my two of my backup VMs use internally less space than is seen on hipervizor.

Here is an example.
Inside VM:
Code:
/dev/sdb        4.0T  1.8T  2.2T  44% /backup
Outside VM:
Code:
rpool/data/vm-140-disk-1                                  2.70T  1.58T     2.70T  -
VOLSIZE  LUSED   USED  VOLBLOCK  COMPRESS  VOLBLOCK
  3.91T  1.66T  2.70T        8K        on        8K

Seems to me like the VM uses lots of small files (ext4 is default 4k) which in turn use 8k blocks on host and use twice as much disk space.
I might be wrong. It might also be related to compression. Does anyone care to explain?

I guess I could stop the VM, send rpool/data/vm-140-disk-1 elsewhere, recreate it (just this zvol) with 4k block size, send data back and start VM.
What do you guys think? Will this work and also decrease disk usage?

I also forgot why again we went with 8k default. :) I guess I could google.
 
What do you guys think?
I am using my magic crystal ball and can see you have a raidz based zfs pool
I also can see in there that the default ZVOL blocksize of 8k is used.

If that's wrong, my crystal ball needs re-calibration.

If it is true you need to create the zvol with a larger block size (64k, 128k) to avoid padding overhead.

;)
 
  • Like
Reactions: mailinglists

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!