If you look at the man page for mkfs_ufs, tunefs, fs_ufs (Solaris 2.6 in my
case) it gives an idea of why this is done.
mkfs_ufs states :
free=n The minimum percentage of free space to maintain in the file
system. This space is off-limits to normal users. Once
the file system is filled to this threshold, only the
superuser can continue writing to the file system. This
parameter can be subsequently changed using the tunefs(1M)
command. The default is 10%.
tunefs states :
tunefs is designed to change the dynamic parameters of a
file system which affect the layout policies. The file sys-
tem must be unmounted before using tunefs. When using
tunefs with filesystem, filesystem must be in /etc/vfstab.
The parameters which are to be changed are indicated by the
options given below.
Generally one should optimize for time unless the file sys-
tem is over 90% full.
-m minfree Specify the percentage of space held back
from normal users; the minimum free space
threshold. The default value used is 10%.
This value can be set to 0, however up to a
factor of three in throughput will be lost
over the performance obtained at a 10% thres-
hold. Note: If the value is raised above
the current usage level, users will be unable
to allocate files until enough files have
been deleted to get under the higher thres-
fs_minfree gives the minimum acceptable percentage of file
system blocks which may be free. If the freelist drops
below this level only the super-user may continue to allo-
cate blocks. fs_minfree may be set to 0 if no reserve of
free blocks is deemed necessary, however severe performance
degradations will be observed if the file system is run at
greater than 90% full; thus the default value of fs_minfree
Empirically the best trade-off between block fragmentation
and overall disk utilization at a loading of 90% comes with
a fragmentation of 8; thus the default fragment size is an
eighth of the block size.
You make your own conclusion sabout how correct F was when she said it was for
Hope this helps
> I recently went to an interview (hello F if you're subscribing!) and was
> asked a question as to why Unix sys admins always leave a partition of 10%
> on every disk. Well, the only thing I could think is that 10% space without
> quota would allow creation of log files etc to keep the system up, although
> I couldnt think why this would be on a separate partition. I therefore
> enquired whether the question pertained to data disks or the disk with the
> root partition. Apparently it doesnt matter. Then I turned my answer to
> inode tables and reserved space for file table data. Nope!
> So I asked F for the answer. She said that it was so that root could login
> to the disk to perform system maintenance. Now as you can imagine at first
> sight this didnt make sense at all (especially if the disk only stores data)
> and not wanting to cause confrontation in the interview I let it ride.
> Since then I have asked many colleagues and consulted the Solaris Admin
> manuals. FULLY! This is all I can find (page 10-27 SA-235) and I quote:
> Note - Approximately 10 percent of your disk capacity is reserved for file
> system proficiency. This is not reflected in the df -k output.
> Anyone have any ideas as to exactly what is this is about? I failed the
> interview apparently because my Unix skills were not strong enough although
> I have subsequently had two job offers!
> But this has really irritated me and I would like to know: Is she right?
> Slightly Disgruntled Unix Sys Admin.