Conference star::wizards

1649.0. "Open: confused...will this work or not?" by STAR::JKEENAN () Tue Mar 04 1997 15:23

Return-Path: "VMS001::WWW"@vms001.das-x.dec.com
Received: by vmsmkt.zko.dec.com (UCX V4.1-12, OpenVMS V6.2 VAX);
	Tue, 4 Mar 1997 13:39:41 -0500
Received: from vms001 by mail12.digital.com (8.7.5/UNX 1.5/1.0/WV)
	id NAA20385; Tue, 4 Mar 1997 13:30:11 -0500 (EST)
Date: Tue, 4 Mar 1997 13:31:49 -0500
Message-Id: <[email protected]>
From: "VMS001::WWW"@vms001.das-x.dec.com (04-Mar-1997 1331)
To: [email protected], [email protected], [email protected]
Subject: Ask the Wizard: '[email protected]'
X-VMS-To: [email protected]

Remote Host: (null)
Browser Type: Mozilla/3.0 (Macintosh; I; PPC)
Remote Info: <null>
Name: Cindy Steinmetz
Email Address: [email protected]
CPU Architecture: VAX
Version: v 5.5-2
Questions: 

We are preparing to upgrade our VAX 4300 to a newer release of VMS.  We were going to go to release V6.2 until we started
doing some investigations into the 3-rd Party SCSI disk drive problems that have been discussed on the INFO-VAX Listserv.
We use Seagate SCSI drives with CMD Controllers.

I talked with someone at CSC and she told me that the problem was caused by the introduction of the tagged command queueing
feature in VMS 6.2 and that Seagate drives were the most vulnerable to problems.  She referred me to the VMS 6.2 New Features Manual
for more information.  The New Features Manaual only discussed tagged command queueing on an Alpha.  Does this mean that the feature
was not a part of the VAX 6.2 release?  And if so does that mean that our Seagate drives on our VAX
will work ok with VMS 6.2?  Also, if the feature does still exist for the VAX version of 6.2 do any of the ECO patches that are out there
fix the problems?

I am at this time very confused on this whole issue.  What we would like to do is upgrade a VAX 4300 that is running several Seagate SCSI drives and a couple of RF31 drives from VMS 5.5-2 to VMS 6.2 (via VMS 6.1).
Will this work or not?

Thanks.

Cindy Steinmetz
University of Indianapolis
1400 E Hanna Avenue
Indianapolis, IN 46227
(317) 788-3361
[email protected]

   This is the classic OpenVMS upgrade problem occasionally encountered
   when one is using a third-party system disk...

   These are not DIGITAL qualified drives, and the wizard is not in a
   position to predict the correct operation of these drives.  (And not
   knowing the particular Seagate drive model and firmware version...)

   If the SCSI implementation in the drive is correct, OpenVMS should
   operate with the drive -- most problems that were encountered with
   OpenVMS V6.2 were around a particular drive's mishandling of one or
   more of SCSI commands -- typically the command was the 10-byte SCSI
   mode sense command.

   You will want to contact Seagate or the vendor for support information
   on these disks, and for any applicable firmware upgrades. 

   OpenVMS Alpha V6.2 and later implements Tagged Command Queuing (TCQ),
   and various Seagate drives are not compatible with this SCSI feature.
   (TCQ is an OpenVMS Alpha V6.2 feature, and is not present on OpenVMS
   VAX.  TCQ is one of the underpinnings of the VMScluster dual-host SCSI
   support, a feature available only on OpenVMS Alpha...)

   For SCSI disk support information, see section 4.11.2.3 in the OpenVMS
   6.2 Release Notes.

   As you indicate OpenVMS VAX: I would perform the OpenVMS VAX upgrade,
   making the recommended full standalone BACKUP/IMAGE prior to the
   upgrade, and I would maintain a second bootable device -- either
   [SYS1] on the OpenVMS CD-ROM (the presence of this root depends on
   the particular CD-ROM release), standalone BACKUP on the OpenVMS
   CD-ROM, or an OpenVMS or standalone BACKUP root on another DIGITAL
   disk, such as one of the RF-series disks.

   Having a bootable OpenVMS root available is the best situation here,
   as it will allow direct diagnosis and direct repairs should there be
   any problems encountered with the Seagate system disk.