From owner-freebsd-scsi  Wed Nov 15 20: 8:36 2000
Delivered-To: freebsd-scsi@freebsd.org
Received: from gonzo.speakeasy.net (gonzo.speakeasy.net [216.254.0.5])
	by hub.freebsd.org (Postfix) with SMTP id CC67D37B4CF
	for <scsi@freebsd.org>; Wed, 15 Nov 2000 20:08:32 -0800 (PST)
Received: (qmail 11907 invoked from network); 16 Nov 2000 04:07:52 -0000
Received: from unknown (HELO aldan.algebra.com) (64.24.41.117)
  by gonzo.speakeasy.net with SMTP; 16 Nov 2000 04:07:52 -0000
Received: (from mi@localhost)
	by aldan.algebra.com (8.11.1/8.11.1) id eAG45gq05238
	for scsi@freebsd.org; Wed, 15 Nov 2000 23:05:42 -0500 (EST)
	(envelope-from mi)
From: Mikhail Teterin <mi@aldan.algebra.com>
Message-Id: <200011160405.eAG45gq05238@aldan.algebra.com>
Subject: same drives acting differently
To: scsi@freebsd.org
Date: Wed, 15 Nov 2000 23:05:38 -0500 (EST)
X-Face: %UW#n0|w>ydeGt/b@1-.UFP=K^~-:0f#O:D7w<gv/&E-lL7twZCT8B~/PA4|\t$ti+22K">
	 hJ5G_<5143Bb3kOIs9XpX+"V+~$adGP:J|SLieM31VIhqXeLBli"<kcG^EOVih
	 y+z3/UR{6SCQ
X-Mailer: ELM [version 2.4ME+ PL60 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-scsi@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.org

I have a pair of seemingly identical  old drives, which I wanted to turn
into a small raid array using ccd:

da1 at sym1 bus 0 target 0 lun 0
da1: <SEAGATE ST32550N 0021> Fixed Direct Access SCSI-2 device 
da1: 10.000MB/s transfers (10.000MHz, offset 15), Tagged Queueing Enabled
da1: 2047MB (4194058 512 byte sectors: 255H 63S/T 261C)
da2 at sym1 bus 0 target 1 lun 0
da2: <SEAGATE ST32550N 0022> Fixed Direct Access SCSI-2 device 
da2: 10.000MB/s transfers (10.000MHz, offset 15), Tagged Queueing Enabled
da2: 2047MB (4194058 512 byte sectors: 255H 63S/T 261C)

As I  was trying to  figure out the  best interleave factor,  I noticed,
that  systat reported  different figures  for  the drives  as the  rawio
benchmark was running. Something like this:

	Disks  ccd0   da0   da1   da2   da3  acd0   cd0
	KB/t  16.00  0.00  2.00  1.60  0.00  0.00  0.00
	tps      78     0   314   392     0     0     0
	MB/s   1.21  0.00  0.61  0.61  0.00  0.00  0.00
	% busy   97     0    34    97     0     0     0

See, the same  amount of MB/s, but seriously  different busy percentage.
This is quite consistent for reading  and writing parts of the rawio --
the da1 is always less busy than da2.

I'm wondering if the da2 is dying on me, or what else is wrong with it,
or is this because it is the second on the same SCSI chain, or what...

Any ideas? Thanks!

	-mi


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message