Re: Fragmentation (was: XFS NFS server Oops)

[Simon Matter <simon.matter@ch.sauter-bc.com>]

Steeve McCauley schrieb:
> 
> On Wed, Mar 20, 2002 at 11:09:16AM -0600, Steve Lord wrote:
> > On Tue, 2002-03-19 at 19:37, Federico Sevilla III wrote:
> > Probably hold off for now on running fsr.
> >
> > Look at the other numbers on the output. The actual and ideal are
> > more interesting. If you look at these, the difference is the number
> > of extra extents you have above the ideal case. Then ask how much
> > data you have on the disk, dividing by the actual extent number
> > gives you the average length of the extents. It is also possible
> > that most of the fragmentation is restricted to a few files.
> 
> We were stressing some systems with 1.6TB drive arrays
> and ended up with very full, very fragmented filesystems.
> This led to fs corruption and/or sysstem instability when
> the system ran out of memory.  Note that these systems
> were running at 99.9% disk full and fragmentation factors
> of 99.8%.  Obviously any sane admin would not run a system
> in this state for long (insane admins, redundant?).
> 
> My question is can XFS can be configured to reserve
> a chunk of disk for root only access?  I vaguely remember
> setting up an HPUX system several years ago and specifying
> that 5% of the disk should be reserved for administrative
> purposes.  That might be a bit extreme with a 1.7TB drive
> array, but you get the idea.  Is there anything like this
> in xfs?

I was asking the very same question some month's ago and IIRC the answer
was no. You're right with HP/UX about the 5% reserved space but also
Linux with ext2 has this 5% limit as a default. Maybe this could go to
the feature list.

-Simon

> 
> Thanks,
> 
> steeve
> 
> --
> :wq