[tlug] Re: Mysterious LVM issue

[John Fremlin <john@example.com>]

Birkir A. Barkarson wrote:
> The crazy values after that must be some kind of bug, as Francois noted.
> The 232GB reported doesn't add up no matter how you look at it.

That is certainly mysterious. It had me wondering. However, look at your
df logs in detail.


Here is the one with the insane 232 GB

birkirb@example.com df -h
Filesystem             Size   Used  Avail Use% Mounted on
/dev/hda2               11G   1.2G   8.9G  12% /
tmpfs                   66M      0    66M   0% /lib/init/rw
udev                    11M    66k    11M   1% /dev
tmpfs                   66M      0    66M   0% /dev/shm
/dev/hda1              104M    16M    83M  16% /boot
/dev/mapper/america_main-store
                       232G   230G      0 100% /srv
/dev/mapper/america_main-home
                       983M    25M   909M   3% /home
/dev/mapper/america_main-local
                       3.0G   473M   2.4G  17% /usr/local

and here is the next one with the normal values

<deleted about 8GB of stuff>

birkirb@example.com df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/hda2             9.9G  1.1G  8.3G  12% /
tmpfs                  62M     0   62M   0% /lib/init/rw
udev                   10M   64K   10M   1% /dev
tmpfs                  62M     0   62M   0% /dev/shm
/dev/hda1              99M   15M   79M  16% /boot
/dev/mapper/america_main-store
                      216G  211G     0 100% /srv
/dev/mapper/america_main-home
                      938M   24M  867M   3% /home
/dev/mapper/america_main-local
                      2.8G  451M  2.2G  17% /usr/local


In the one with the 232GB value all the other disks have bigger values too.

In fact
232*(1000^3) = 232000000000
216*(1024^3) = 231928233984

So the first one seems to have been done with base 1000 and the  others
with base 1024

This is an option to df --block-size=

Can you explain how this happened? ;-)