From owner-freebsd-current  Sun Aug  3 21:47:04 1997
Return-Path: <owner-freebsd-current>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id VAA12825
          for current-outgoing; Sun, 3 Aug 1997 21:47:04 -0700 (PDT)
Received: from pahtoh.cwu.edu (root@pahtoh.cwu.edu [198.104.65.27])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA12819;
          Sun, 3 Aug 1997 21:46:55 -0700 (PDT)
Received: from opus.cts.cwu.edu (skynyrd@opus.cts.cwu.edu [198.104.92.71])
	by pahtoh.cwu.edu (8.8.5/8.8.5) with ESMTP id VAA07345;
	Sun, 3 Aug 1997 21:46:54 -0700 (PDT)
Received: from localhost (skynyrd@localhost)
	by opus.cts.cwu.edu (8.8.6/8.8.5) with SMTP id VAA12297;
	Sun, 3 Aug 1997 21:46:54 -0700 (PDT)
Date: Sun, 3 Aug 1997 21:46:53 -0700 (PDT)
From: Chris Timmons <skynyrd@opus.cts.cwu.edu>
To: Stefan Esser <se@freebsd.org>
cc: current@freebsd.org
Subject: NCR behavior wrt spun-down disks at boot
In-Reply-To: <19970417114955.56763@x14.mi.uni-koeln.de>
Message-ID: <Pine.BSF.3.95.970803213137.12149A-100000@opus.cts.cwu.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


Hi Stefan,

I have a couple of DEC RZ26L drives on an SC200 controller (data only, not
the boot device nor swap.)  Presently I think they are jumpered to wait
for a start unit command before they spin up (I'll fix that the next time
I get into the box.)

With -current of July 22, 1997, the boot probe works fine, and when fsck
touches them I get a couple of complaints but thereafter everything is ok:

ncr1: <ncr 53c810a scsi> rev 0x11 int a irq 11 on pci0.12.0
ncr1: waiting for scsi devices to settle
scbus1 at ncr1 bus 0
sd1 at scbus1 target 1 lun 0
sd1: <DEC RZ26L    (C) DEC 440C> type 0 fixed SCSI 2
sd1: Direct-Access 
sd1: 10.0 MB/s (100 ns, offset 8)
1001MB (2050860 512 byte sectors)
sd2 at scbus1 target 2 lun 0
sd2: <DEC RZ26L    (C) DEC 440C> type 0 fixed SCSI 2
sd2: Direct-Access 
sd2: 10.0 MB/s (100 ns, offset 8)
1001MB (2050860 512 byte sectors)


(complaints until the drives come up... work fine thereafter)

sd1: error code 1, retries:2
sd1: error code 1, retries:1
sd1: error code 1, FAILURE
sd2: error code 1, retries:2
sd2: error code 1, retries:1
sd2: error code 1, FAILURE

With -current from today, things don't go as well.  I don't have the
machine on a serial console, so I will paraphrase what happens. 

The boot probe does not succede and the system hangs.  As in the log from
above, we get to:

	scbus1 at ncr1 bus 0

but then before we see sd1 (partially incomplete hand transcribed):

	scbus target 1 lun 0: command failed (9ff) @ fa043600
	
	ncr1 Aborting Job...
	ncr1:1: Error (10:0:) (8-0-0) (0/3) @ script 424


and the reset switch is my only friend :(

I think it is an acceptable solution for me to jumper the drives so that
they spin up on power, but I thought that this change in behavior was
noteworthy enough to call to your attention.

Regards,

-Chris