Different disk image size and actual amount of space

[djkobraz]

Hello, I have a problem with the real disk space in my containers. This bug is present in LVM and raw images, on ZFS all work correctly.

LXC container have disk image size 100Gb but in the container:

df -h
Filesystem Size Used Avail Use% Mounted on
/dev/loop1 98G 1.1G 92G 2% /
none 492K 0 492K 0% /dev
tmpfs 126G 0 126G 0% /dev/shm
tmpfs 126G 8.2M 126G 1% /run
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 126G 0 126G 0% /sys/fs/cgroup

At first - 98G, not 100G... but this is not so critical how 92G available space ( 98 - 1.1 != 92 ) Where other disk space. I thought that the problem was on LVM-Thin, but on the clean ext4 partition and raw image this bug still present.

 

[Tapio Lehtonen]

If you made the filesystem with 5 % reserved for root use, then the calculation would be correct (i.e. 5 GB is reserved and not available).

 

[djkobraz]

How I can change this option? In the container creation manager, this option is not available. This is a custom LXC parameter?

 

[djkobraz]

Thank you!
tune2fs -m 0 /mnt/pve/nvme_0/images/415/vm-415-disk-0.raw give needed result and now disk allow use all space but still present only 98G space, maybe you know how to fix this also?

 

[Tapio Lehtonen]

I believe there is nothing to fix. The disk image may be 100 GB, but there is some overhead when you put a file system there, so the usable space may well be 98 GB.

 

[djkobraz]

I think that ZFS also has overhead but there this problem is not present... This is not a problem for personal use, but difficult to explain for a client why he takes 100Gb disk space but receives only 92Gb. ZFS works very good but has very big RAM overhead... Tune every node after creation is not looked like a production solution and can create problem after migration from ZFS to other solution as LVM or RAW. So, I think that this is not a small specific problem and can't be ignored, also I hope that this is my mistake and incorrect configuration. I can't believe that developers accept this as normal when you create 150Gb node but in dashboard only 146.65Gb ... look very strange http://i.imgur.com/r6Ydxw8.png