ABOUT "DU -SK" AND "DF"

This document describes why numbers from "du -s" and "df"
Disagree and is applicable to AIX version 3.2 and 4.1.

On AIX versions prior to 4.1 df reports its statistics in
1024 byte units and du reports in 512 byte units. On AIX 4.1
and later both df and du default to 512 byte units.  The
following discussion uses 4.1.2 df and du thus all units are
in 512-byte blocks.

THE PROBLEM

When you execute "du -s /filesystem_path" and subtract this
value from the total block count as reported by df to get a
free block value, the calculation yields a
value that is greater than df's free block value.  For
example:

   % du -s /tmp
   12920   /tmp

      % df /tmp
   Filesystem    512-blocks      Free %Used    Iused %Iused Mounted on
   /dev/hd3           57344     42208   26%      391     4% /tmp

      <total from df> - <used from du> = <false free block count>
      57344 - 12920 = 44424

      44424 is greater than 42208.  The reason for this discrep-
      ancy has to do with the implementation of du and df.

DU -SK

du -sk transverses the file tree adding up the number of
blocks allocated to each directory, symlink, and file as
reported by the stat() system call. This is how du arrives
at its total value.


DF

df looks at the file system disk block allocation maps to
arrive at its total and free values.

WHY THE NUMBERS DO NOT ADD UP

The file system allocates some of the disk blocks in the
file system to record its data; this data is referred to as
meta data. Meta data is not visible to most user level pro-
grams. Examples of meta data are inodes, disk maps, indirect
blocks, and super blocks.

du is an example of a user level program that is not aware
of file system meta data.  While df looks at the file system
disk allocation maps, and is aware of file system meta data.
df obtains the true file  system statistics whereas du only
sees a partial picture.  For example, n empty 4 MB JFS file
system created with frag=4096 and nbpi=4096 has the fol-
lowing meta data allocated:

     1 4k block for the LVM
     2 4k super blocks
     2 4k blocks for disk maps
     2 4k blocks for inode maps
     2 4k blocks for .indirect
    32 4k blocks for inodes
   -------------------------
    41 4k blocks for meta data on an empty 4M file system

       # df /foo
       8       /foo

The 8 512 byte blocks reported for du on this empty file
system are the blocks used by the root directory.

In order to get du's output to match df's output we must add
in the meta data. First, convert 42 4K blocks to 512 byte
units:

       41 * 8 = 328
       328(meta data) + 8(from du) = 336

so there are 336 512-byte blocks allocated on this empty
file system, thus:

   8192(total blocks) - 336(used from du + meta data) = 7856

This does match the output from df's free column.  This cal-
culation was easy to perform on an empty file system.
However, on a non-empty file system, the meta data for file
indirect blocks comes into play and such calculations are
very tedious and impractical.

In conclusion du -s produces a value that reflects the
number of disk blocks that are allocated to file and direc-
tories.  df reports on the actual allocation state of the
file system.  The true allocation state includes both user
data (files and directories) plus meta data.