Re: Wrong df Avail space on three disk RAID1 when empty

[Hugo Mills <hugo@carfax.org.uk>]

[-- Attachment #1: Type: text/plain, Size: 3164 bytes --]

On Tue, Feb 02, 2016 at 03:16:40PM +0100, Marco Lorenzo Crociani wrote:
> Hi,
> df on btrfs RAID1 fs on 3 disks  (they are RAID0 hardware, two disk,
> arrays but its not important right?) show wrong avail space when
> empty
> 
> # uname -a
> Linux kvm4.prisma 4.4.0-2.el7.elrepo.x86_64 #1 SMP Tue Jan 26
> 13:06:01 EST 2016 x86_64 x86_64 x86_64 GNU/Linux
> 
> # btrfs --version
> btrfs-progs v4.4
> 
> #   btrfs fi show
> Label: 'data'  uuid: 4421464d-479c-44c4-9f01-8df0f7d05bd2
>         Total devices 3 FS bytes used 640.00KiB
>         devid    1 size 7.28TiB used 2.00GiB path /dev/sdb
>         devid    2 size 7.28TiB used 2.01GiB path /dev/sdc
>         devid    3 size 7.28TiB used 2.01GiB path /dev/sdd
> 
> # btrfs fi df /data
> Data, RAID1: total=1.00GiB, used=512.00KiB
> System, RAID1: total=8.00MiB, used=16.00KiB
> Metadata, RAID1: total=2.00GiB, used=112.00KiB
> GlobalReserve, single: total=16.00MiB, used=0.00B
> 
> 
> I expect to get from df -h some like
> Filesystem            Size  Used Avail Use% Mounted on
> /dev/sdb                       11T   17M    11T   1% /data
> but I get:
> Filesystem            Size  Used Avail Use% Mounted on
> /dev/sdb                       11T   17M    7,3T   1% /data

   I've noticed this kind of error, too, although not to such a great
degree. The space estimation is definitely broken, although not to
such a bad degree as it used to be.

   There's a relatively complex (but implementable) algorithm that can
be used to determine the remaining usable space quite precisely. I'm
guessing that that's not in use here. :)

   I'll try to dig out the actual formula this evening and post it to
the list for reference. (It's the formula that's used in the
btrfs-usage tool[1], if anyone wants to see the JavaScript
implementation).

   Hugo.

[1] http://carfax.org.uk/btrfs-usage/


> If I fill the fs
> 
> # ls -laht
> totale 11T
> -rw-r--r--.  1 root root 921G  1 feb 16.17 prova10
> drwxr-xr-x.  1 root root  134  1 feb 15.23 .
> -rw-r--r--.  1 root root 1,0T  1 feb 14.00 prova9
> -rw-r--r--.  1 root root 1,0T  1 feb 13.07 prova8
> -rw-r--r--.  1 root root 1,0T  1 feb 12.20 prova7
> -rw-r--r--.  1 root root 1,0T 28 gen 21.50 prova6
> -rw-r--r--.  1 root root 1,0T 28 gen 21.08 prova5
> -rw-r--r--.  1 root root 1,0T 28 gen 20.28 prova4
> -rw-r--r--.  1 root root 1,0T 28 gen 19.50 prova3
> -rw-r--r--.  1 root root 1,0T 28 gen 19.13 prova2
> -rw-r--r--.  1 root root 1,0T 28 gen 18.37 prova1
> -rw-r--r--.  1 root root 1,0T 28 gen 18.02 prova0
> dr-xr-xr-x. 18 root root 4,0K 28 gen 15.30 ..
> 
> I get:
> 
> Filesystem            Size  Used Avail Use% Mounted on
> /dev/sdb                       11T   11T    256K 100% /data
> 
> # btrfs fi df /data/
> Data, RAID1: total=10.90TiB, used=10.90TiB
> System, RAID1: total=8.00MiB, used=1.50MiB
> Metadata, RAID1: total=13.00GiB, used=11.24GiB
> GlobalReserve, single: total=512.00MiB, used=0.00B
> 

-- 
Hugo Mills             | "He's a nutcase, you know. There's no getting away
hugo@... carfax.org.uk | from it -- he'll end up with a knighthood"
http://carfax.org.uk/  |
PGP: E2AB1DE4          |                         Lexy, The League of Gentlemen

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]