Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!vsi1!zorch!ditka!zinn!siia!wje
From: wje@siia.mv.com (Bill Ezell)
Newsgroups: comp.unix.aix
Subject: Re: Adding a 3rd party SCSI disk to an rs6000
Keywords: subclass disk rs6000 AIX
Message-ID: <1990Nov16.141718.11641@siia.mv.com>
Date: 16 Nov 90 14:17:18 GMT
References: <1149@cameron.egr.duke.edu> <531@comcon.UUCP>
Organization: Software Innovations, Inc.
Lines: 31

In <531@comcon.UUCP> tim@comcon.UUCP (Tim Brown) writes:

>I have been following this for a while.  I have also been talking to
>seagate engineers and experimenting.  Here is what works and why.

>... In other words you can't have a
>situation where the seagete (wren) is hooked up externally *and* have
>an IBM drive internally.  It is an either or situation.  
...

>3. If you already have one internal scsi, you must either forfiet a
>tape or cdrom OR wire the Wren up as an internal scsi chained with the
>first.

>I have proved this both at IBM on one of thier demo systems and on our
>system.  At IBM I hooked my WrenV up as another internal drive or
>second scsi (they had one in there).

Sorry to disagree, but we have an internal IBM disk, an internal Seagate
disk, AND an EXTERNAL 600 MB Wren AND a CDROM drive, AND a tape drive,
all at once, and all working (for a long time now).

The bit about the limit on the number of external devices is basically
IBM being paranoid. It really depends upon how much loading your external
SCSI devices place on the bus. Apparently, the line drivers aren't all
that robust. (Or perhaps they just want you to buy a 520?)
-- 
Bill Ezell
Software Innovations, Inc.
wje@siia.mv.com
(603) 883-9300