[We are trying this again, last time I got a "rice.edu not found for mailer
ddn" error -Brad]
Folks,
After a number reciving a number of requests to post further
information on ``sticktion'' and Seagate/Imprimis/Control Data Corporation
(CDC) drives, I have compiled the following. It is being cross-posted to
Sun-Managers and Sun-Spots, in hopes that it will find it's way to some
interested folks that are on one, but not the other.
I would first recommend that you seriously re-consider slapping the
drives. I have had private responses from people who have said (in
effect) "Yah, I slapped (or kicked!) the drive pretty hard, it still
didn't work and had to be sent back to the manufacturer." I think I can
tell you what they found inside the drive afterwards -- Head material
splattered all over the inside of the drive, the heads ripped off the
actuator arms, perhaps even scratches and gouges all over the media
because the actuator arms got bent and came into physical contact with the
media. All in all, if you ever saw the insides of a drive that got
slapped too hard, you might loose your cookies -- it's *not* a pretty
sight!
It is pretty easy to detect the drive having been slapped too hard,
all you have to do is look for external indications of having been
dropped, or whatever (looking for bent sheet metal, etc...). If you don't
find any, then you can be pretty well assured that the drive was slapped
intentionally, as opposed to being accidentally dropped. This is a very
good reason for voiding the waranty of the drive in question.
What I recommend instead is that you follow these instructions (at
your own risk, of course):
Parts Needed: Hair dryer or space heater, accurate and fairly
small thermometer, tool(s) to open the OEM drive enclosure
(shoebox) that the Seagate/Imprimis/CDC drive is in.
1. Turn off the power to your shoebox.
2. Disconnect all power and data cables to the drive.
3. Remove the drive from the shoebox.
4. Reconnect all data and power cables to the drive.
5. Start a program running that constantly tries to access the
drive. Anything simple will do, so long as it loops forever
(until stopped by a signal that you intentionally give it).
6. Place the thermometer on top of the drive.
7. Start up the hair dryer or space heater at the lowest
setting. Make sure that the drive is heated as evenly as
possible.
8. Slowly increase the setting of the heater, watching the
temperature as you go. Make sure the temperature does not
exceed 110 degrees Fahrenheit.
9. Once the drive becomes accessible, keep the temperature fairly
constant for at least five to ten more minutes.
10. Slowly decrease the setting of the heater, thus slowly
decreasing the ambient temperature near the drive (watching
the thermometer as you go).
11. Stop the program that has been accessing the drive.
12. Power back down, putting the drive back the way you found it.
Remember, Seagate/Imprimis/CDC Wren drives do not like cold
temperatures -- If you feel completely comfortable in the room that they
are in, while you are wearing shorts and short-sleeve shirts, then the
drives will also be comfortable. If the computer room is kept at 50-60
degrees Fahrenheit, then you are almost guaranteed to have problems,
perhaps chronic ones.
Also, the above is primarily intended for the Wren III - VI drives.
The Wren VII is a different kind of beast, and may or may not be amenable
to this kind of treatment. The above may actually be harmful to drives
from other manufacturers, or other models from the same manufacturer(s).
I seem to remember that the Wren IV and Wren V models were most
susceptible to sticktion, but that may be my memory playing tricks on me.
Finally, if my procedures don't work for you, then I would recommend
that you send the drive back to the manufacturer -- they can send it back
to Seagate, and have the folks in the clean room fix your problem for you,
without causing any damage to the drive or data on the drive.
I take no responsibility for anything you may do while trying to fix
your sticktion problems, whether you are using my procedures or not -- the
above should work fine for most folks, but you use whatever procedures you
choose, entirely at your own risk.
Please respond via e-mail. I will summarize and re-post, if appropriate.
________________________________________________________________________
| Brad Knowles | Internet: blknowle@frodo.jdssc.dca.mil |
| System Administrator | or: blknowle@wis-cms.dca.mil |
| DCA/JDSSC/JNSL | Ph: (703) 693-5849 Fax: (703) 693-7329 |
| The Pentagon, Room BE685 |_________________________________________|
| Washington, D.C. 20301-7010 | my opinions != DCA's opinions or policy |
|______________________________|_________________________________________|
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:13 CDT