head	1.1;
access;
symbols
	with-solaris-support:1.1
	before-solaris-version:1.1
	plt-021:1.1;
locks; strict;
comment	@# @;


1.1
date	98.12.10.17.21.01;	author schaefer;	state Exp;
branches;
next	;


desc
@@


1.1
log
@Initial revision
@
text
@$Id$

From mortn@@dilog.com Wed Jul 31 02:00 MET 1996
Return-Path: mortn@@dilog.com
Received: from mail.eunet.ch (pc2.dilog.ch [193.72.186.154]) by sun1.dilog.ch (8.6.12/8.6.9) with ESMTP id CAA22933 for <marc@@dilog.ch>; Wed, 31 Jul 1996 02:00:39 +0200
Received: by mail.eunet.ch (8.7.5/1.34)
	id XAA21959; Tue, 30 Jul 1996 23:15:03 GMT
Received: from dilog.com by mail.eunet.ch (8.7.5/1.34)
	id XAA21953; Tue, 30 Jul 1996 23:14:58 GMT
Received: from 192.9.200.12.dilog.com ([192.9.200.19]) by dilog.com (8.7.5/8.7.3) with SMTP id QAA07156; Tue, 30 Jul 1996 16:15:41 -0700
Received: by 192.9.200.12.dilog.com with Microsoft Mail
	id <01BB7E32.EED4AB00@@192.9.200.12.dilog.com>; Tue, 30 Jul 1996 16:19:50 -0700
Message-ID: <01BB7E32.EED4AB00@@192.9.200.12.dilog.com>
From: "Mort K. Nistanaki" <mortn@@dilog.com>
To: "'Marc Schaefer'" <marc@@dilog.ch>
Cc: "'Paul Sutherland'" <paul@@dilog.ch>
Subject: Dual Libra test software
Date: Tue, 30 Jul 1996 16:19:48 -0700
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="us-ascii"
Content-Length: 1862
Status: RO

Marc,

We need to modify dualib diagnostics software so we can use it to =
qualify the dual libra.  Below is the specification of the program; =
could make this a priority and email a copy of the software to us ASAP.

The object of the software is to continuesly and concourently process =
the two tapes.  For each tape device, it will randomley select a tape =
from any AVAILABLE (use read element status data; don't assume a full =
pack) slot and insert it in the tape then write 2MB of data (w/t =
compare).  Once the data transfer is operation is done, it will =
randoumly select and empty slot and put the tape in the pack.  The =
object of this program will be to continuesly keep the stacker active =
and once a drive is empty, it will load a new tape in.

The idea here is to have a back to back move operations on both tape =
drives.  The randum selector must be intelligent enought were it will =
cover all slots for both insertion and retrieval.  What we are trying to =
do is qualify the mechanizem.  The test will also periodicly will unload =
and reload the pack.

The test will be program to run for ever or for a assigned period of =
time.  While it is testin it will keep track of number of =
insertion/retrieval for each drive and slot, pack load/unload, and pass =
count.

We need to come up with a test procedure where we can reliablity qualify =
the product and we need to define objectives and defenition for when and =
how the product is considered reliable.  This test will be the essential =
part of the above process and we need to have it as soon as possible.

As you are writting the software do use youre experience with the libra =
to interject new ideas in the test.  This test will not be interactive =
and the result will be pass/no pass.  Please disable all error recovery =
(except media errors from the tape).

Regards,
Mort.

@
