Re: System Errors and SCSI

[John Le Febvre <John@klatch.demon.co.uk>]

dempson@atlantis.actrix.gen.nz (David Empson) writes:
>     0911 cantSync
> 
> indicating that it is an error from toolset 9 (Apple Desktop Bus), and
> the specific error is "cannot synchronize".  (It means that the CPU
> was not able to synchronize communciation with the keyboard
> microcontroller, possibly due to the keyboard micro having gone nuts.)

Aha!  Distinctly possible as my keyboard is a bit faulty.  It tends to
put two 4's in for each key press every so often, also the 'down-arrow'
key is similarly affected.  A minor annoyance.

[SCSI probs]
> If you have two or more SCSI drives connected to an original Apple
> SCSI card, you must have two terminators: a "pass through" terminator
> connected between the computer and first drive (or internal
> termination on the first drive), and a second terminator after the
> last drive (or internal termination on the last drive).
> 
> It is probably best to use a passive terminator for the computer end.
> The other end could be active for better results.

I remembered about this pass-through terminator (good job I had one to 
hand!).  It didn't make any difference.  Anyway, I think it's a bit more 
serious than that.  When I had tried this prior to getting the active 
term, I tried to partition the drive but it failed and started reporting 
itself as a 6Mb drive (via SlotScan).  Eventually I took the drive to a 
PC and low-level formatted it with the built-in utilities of an Adaptec 
1542cf card.  Took it back to the Apple, hey presto 510Mb again.  Then I 
got the active term, tried it again, and now the drive once more reports 
itself as 6Mb (although it did say 10Mb once...).  I took it back to the 
PC again, but even the PC won't recognize it now...

> Another possibile problem is supply of termination power - at least
> one of the drives must be doing this.  It doesn't sound like this is
> the cause though.

Yep, had the 510Mb Seagate supplying the bus.

> Also double-check the SCSI IDs.  One of the drives might be lying
> (e.g. it might not have a connection to an external selector switch,
> so it would actually have an ID set by jumpers or defaulting to 0).

330Mb=ID5, 510Mb=ID0 (double-checked).  Actually, my drive Partition
software reported this the one time it recognised both drives together.

> I often find that peculiar SCSI problems are impossible to solve by
> obeying the rules.  You might have to over-terminate or
> under-terminate the bus to get it to work, for example.  There are
> also cases where the order of the drives affects the behaviour.  These
> are usually due to poor SCSI implementations in one of the drives.

Ok, thanks (again!) for your help, David.  I shall experiment some more
and see what I can do.
-- 
John.
john@klatch.demon.co.uk