[SunRescue] Re: Solaris 7 and ADIC Scalar 218 DLT Library

[Ido Dubrawsky]

--BOKacYhQ+x31HxR3
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable

Well,=20

   Good news.  I finally got the E450 to recognize the tape drives.  It tur=
ns
out the System Admin for that box (a co-worker) had set the robot arm to SC=
SI
ID 4 (through the front panel menu) and the tape drives SCSI IDs (set throu=
gh
the "ID selector" wheels in the back) to 4 and 5.  I set the SCSI ID of the
robot to 3 but the drives still wouldn't show up.=20
  Then I took a close look at the cabling.  The ADIC scalar has a setup such
as:

					Centronic SCSI connectors
					3.    4.          5.     6.
					^      ^          ^      ^
                                        |      |          |      |
                                        |      |          |      |
                                        |      |          |      |
                                        |      |          |      |
                                        |      |          |      |


      1. <----wide SCSI Connector---->
      2. <----wide SCSI Connector---->

Now, when you've only got one SCSI bus on the system (which is not quite wh=
at
we have, but we've got this thing on it's own bus nonetheless), ADIC says t=
hat
the SCSI cabling should be: host->2, 1->3, 4->, 6-terminated (the cable from
the host connects into connector #2, then connector #1 uses a wide->centron=
ics
cable to connect 1->3, then this has a centronics->centronics cable to conn=
ect
4->5 and 6 has the active terminator).  Well, it appears that SCSI connecto=
r=20
#3 has been "damaged" somehow because with the cable attached from 1->3, the
drives don't show up in probe-scsi-all (I thought that one of them did since
I had set the SCSI ID of the drive to 4 and saw it show up, not realizing t=
hat
the other admin had set the SCSI ID of the robot to 4 as well...I was actua=
lly
seeing the SCSI ID of the robot, not the tape drive).  However, when I chan=
ged
the SCSI ID of the robot to 3, I realized that the device I was seeing under
probe-scsi-all was the robot (the SCSI ID of the device that did show up=20
changed to 3...big clue there...).  Anyway, even with the SCSI ID problem=
=20
resolved the tape drives still didn't show up.  I then plugged the wide-
centronics cable from connector #1 to connector #6 (1->6) and put the=20
terminator on #3.  Lo and Behold!  Both drives now show up and everything=
=20
works under Solaris!  However, this is technically "illegal cabling" accord=
ing
to the ADIC installation and configuration manuals.  I see no real reason a=
s to
why (and I know that it really isn't illegal).  There are times in which I=
=20
really can't stand the vendor's manuals/documentation....grumble, grumble,=
=20
grumble!!!.  Thanks to Mike Hebel though for pointing me to a great document
on sunsolve with regards to Solaris and Tape Libraries (I found it!).

Thanks Again,
Ido (It's not my job to be the sys admin)

On Fri, Feb 23, 2001 at 06:53:03PM -0600, rescue-request at sunhelp.org wrote:
> From: "Mike Hebel" <drone8of9 at crosswinds.net>
> To: <rescue at sunhelp.org>
> Subject: RE: [SunRescue] Solaris 7 and ADIC Scalar 218 DLT Library
> Date: Fri, 23 Feb 2001 17:18:29 -0600
> Reply-To: rescue at sunhelp.org
>=20
> Tape libraries are sometimes the hardest to configure.  All I can suggest=
 to
> you is that if you don't have a support contract with someone on this then
> try searching the newsgroups for some answers.  (I believe
> http://groups.google.com/ is the new interface to what used to be Deja
> News.)
>=20
> If you have a Sunsolve contract check out the notes on their private supp=
ort
> side of things.  http://sunsolve.sun.com   That site might be your best l=
ink
> to some info.  In fact it was one of the few places to find info on prope=
rly
> configuring st.conf - at least for me.  If you need it I'll see if I can't
> dig up the number of the Sun technical note on configuring the st.conf fi=
le.
>=20
> I wish I had more specific information for you.
>=20
> Don't give up.  The answers _are_ out there.  Just keep digging.
> www.Google.com is your friend. ;-)
>=20
> Mike Hebel
>=20

--=20
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
                        |Ido Dubrawsky               E-mail: idubraws at cisco=
.com
     |          |       |Network Security Engineer
    :|:        :|:      |Cisco Secure Consulting Services
   :|||:      :|||:     |Cisco Systems, Inc.
=2E:|||||||:..:|||||||:.  |Austin, TX. 78759
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D

--BOKacYhQ+x31HxR3
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBOpsMbG2NSFxvnJ/5AQGqWQP/ZgN44S5+5MyJwUfgiw3kLtpm43ZdBsRX
PPiw3+AllU1/d1h+KHE20nqeoTeagOn52p+r9YmyoCbcPYhqpzCDEdQV2PbghFlj
fQjurDxnU3fIIKQKZ+ct3IP/CtwcpEhQucbc+nycDK1DJungIrIjGrSrxIA7XdkN
j1fBnIAW9Ts=
=1NGt
-----END PGP SIGNATURE-----

--BOKacYhQ+x31HxR3--