Yesterday I described my trying to install an external tape device which had
failed. As it sometimes happens, the solution to the problem proved to be
trivial. As soon as I changed the external cable and terminator, the system
successfully recognized the tape device as well as all the others. Moreover,
putting the connectors in and out more than once did them good... and in the
end I found out they had come to good repair too... I arrived at a conclusion:
the issue was oxygenated contacts (the cable and terminator were never used).
Thanks to all who relplied:
ButcherSJB_at_aol.com,
Randy Rodgers <randy.rodgers_at_ci.ft-wayne.in.us>,
John E. Taylor jetaylor_at_slxcg01.csw.L-3com.com,
and "V.S. Glukhov" <glukhov_at_sequence.Stanford.EDU> whose remark might be
very useful for future readers of the arhives, though in my case it was the
connector at the *END* of the ribbon cable that was connected to the onboard
SCSI port, and it prevented from transformation of a chain SCSI to a star
topology. Here is his mail:
>I suspect that by connecting your tape drive you transformed your
>SCSI topology from a chain to a star. SCSI will not work at all if
>you have a star topology. I think the manual for the adapter should mention
>this. The scsi chain should be terminated on both ends.
>
>If you would draw a picture of your SCSI chain that would certainly help a
>lot. I suspect your have
>
>
>
>term --- tape ------ SCSI adapter ----- disk ---- disk ---- ... ---- term
> |
> +--------disk ----- ..... ---- term
>
My original mail describing some symptoms of wrong cable or terminator is at
the end of the message.
Thanks,
Irene
*************************************************************************
* *
* Irene A. Shilikhina e-mail: irene_at_alpha.iae.nsk.su *
* System administrator, *
* Institute of Automation & Electrometry, *
* Siberian Branch of Russian Academy of Sciences, *
* Novosibirsk, Russia *
*
http://www.iae.nsk.su/~irene *
*************************************************************************
* * *
* The road to hell is paved with * Every cloud has a silver lining. *
* good intentions. * *
* * *
*************************************************************************
On Thu, 17 Jun 1999, Irene A. Shilikhina wrote:
>
> Hello Managers,
>
> our system supports no more than four internal SCSI devices, and all of
> them are busy. Now I failed in my first try to install an external box
> with a (used) tape device (as far as I can suppose, TZK10 since it worked
> with the QIC-525 format DC6525) on our DEC 2000 model 300 with the only
> Adaptec AHA-1742A SCSI adapter. According to the documentation on the
> adapter, it must be rather simple, it isn't in reality though...
>
> I set an unused SCSI ID for the tape device and, following the documentation
> on the adapter, remove all three onboard SCSI terminators, then properly
> connected both ends of the SCSI cable, set the terminator on the second port
> of the external box, and ... the system hung during the "show dev" command.
>
> Since such a symptom may happen as a result of ID interference, I tested
> the real SCSI ID of the new device as follows. I disconnected the internal
> SCSI ribbon cable from the onboard port and connected an alternative ribbon
> instead, with the only device on it - the tape one. So, I made sure it has
> ID=3 whereas the disks occupy IDs=0,1,2,4, and HOST ID=7. It seems to be OK.
> Nevertheless, connected as an external device it confuses the adapter.
>
> Some more experiments. I disconnected the external box from the end of the
> external cable and set the terminator instead of it (the command "show dev"
> hung), then removed it, i.e. remained its end open ("show dev" showed all
> the disks exactly as they are).
>
> Is it possible behavior in the case if an external cable is wrong or too
> long? But it is no longer than 1m. Both of the connectors of the SCSI cable
> are 50-pin, both the cable and the terminator were shipped along with the
> Alpha itself. Or may it be the adapter wrong? Of course, there would be
> useful one experiment more - to try replacing one of the disks with the
> tape drive as an internal device, but it's not very convenient, and I don't
> know whether it's worth trying.
>
> In the end, I must have returned to the old situation, but ... what puzzled
> me most was that the system booted happily even WITHOUT ANY OF THE THREE
> ONBORD TERMINATORS ... (it wasn't until later that I remembered I should
> had set them again)...
>
> I would be grateful for any advice,
> Irene
Received on Fri Jun 18 1999 - 09:47:38 NZST