This was a thread I started around the first of January and I have
good but cautionary news to report.
When I first tried the Maxtor drive, it was on a 256-MHZ Dell
Pentium mother board with a single IDE controller for the hard drive and
CDROM, the usual IBM-style 1.4-meg floppy and a Yamaha sound chip that
never quite behaved well under Linux. There was also a 3C509 10Baset
Ethernet card and the built-in USB1.1 port.
It is a good, but rather ordinary type system by today's standards.
The USB driver assigned IRQ9 to service the USB system and maybe
that is where things went wrong. I still don't have a good way to know
that for sure.
All I know is that I got to see failure modes like I had never seen
them before. Some disk operations would succeed normally, but most would
end up in Never-Never land and one couldn't even tell the system to reboot
when that happened because root couldn't even kill the processes.
I was in the process anyway of moving all the Linux activity from
that system to a newer Dell mother board so I thought I would give the
Maxtor another try on the new system. I didn't hold out a lot of hope
because the newer system was basically much like the old one except faster
and with two IDE controllers and a different Ethernet card.
It also turned out the sound system was a better fit under Linux
and actually works as it should.
It appears that the USB driver chose a better selection of IRQ this
time because the Maxtor drive appears to work like a good disk drive should
on the new system. The USB port is still a nice slow USB1.1 port so the
Maxtor doesn't get to really scream, but file transfers happen at a rate
one would expect at a 12-Mb bus speed. It is about like Ethernet over a
good LAN using IRQ11 on this system.
What this all amounts to is that if you decide to try this type of
a set-up, look out for the warning signs that something might not be right.
I was lucky for once in that the system I was moving to appears to
be working while the one I am retiring (actually returning to its rightful
owner) had problems, but it could just as easily be the other way around.
In other words, have a plan B.
I think Linux represents an incredible human effort and the people
who have written these drivers and who do their best to make them
automatically do the right thing when faced with an unlimited level of
chaos can't be thanked enough.
Martin McCormick WB5AGZ Stillwater, OK
OSU Center for Computing and Information Services Network Operations Group
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
