"Jared D. McNeill" writes:
> On Mon, 12 Mar 2001, Greg Oster wrote:
> > What does your disklabel for the RAID set look like?  I'm wondering
> > if this could be the "sucky performance due to bad default disklabel values
> "
> > problem reported in PR#11989...
> 
> Wups, I should probably send the whole thing :-)

Ah.. yes.. that's better ;)

> # /dev/rraid0d:
> type: RAID
> disk: raid
> label: default label
> flags:
> bytes/sector: 512
> sectors/track: 64
> tracks/cylinder: 1
> sectors/cylinder: 64
> cylinders: 78116

Ya... "sucky numbers" here... 
If this is a freshly created set (i.e. no real data on it yet) then
maybe try something like:

sectors/track: 64
tracks/cylinder: 16
sectors/cylinder: 1024
cylinders: 4882

for the disklabel, and fix up the other numbers to match.  (you'll need to 
newfs it again..)  In your other email you said:

> Hmm, I just ran '/sbin/disklabel raid0' and it gave me a message like it
> was creating a fresh label. That's not supposed to happen, and I'm pretty
> sure the following isn't mine (although I am using raid0d):

If it had a label before, then maybe something is eating that label!??!!?
That would be Bad... :(  Is this the first time you've tried to extract 
that file on this system, and/or the first time you've noticed this perf 
problem?  (I'm guessing it's the 'bad sector/track/cylinder' numbers, but
if something is eating the disklabel, then it could be something else too..)

> total sectors: 4999424
> rpm: 3600
> interleave: 1
> trackskew: 0
> cylinderskew: 0
> headswitch: 0		# microseconds
> track-to-track seek: 0	# microseconds
> drivedata: 0
> 
> 8 partitions:
> #        size   offset     fstype   [fsize bsize cpg/sgs]
>   c:  4999361       63     unused        0     0         # (Cyl.    0*- 78115
> )
>   d:  4999424        0     4.2BSD        0     0     0   # (Cyl.    0 - 78115
> )
>   e:  4999361       63     unused        0     0         # (Cyl.    0*- 78115
> )

Later...

Greg Oster