Hardware Reference
In-Depth Information
terms that you may not be able to save anything. This does two things: first it allows the
customer to prepare himself for the worst outcome (no data), and secondly it allows you to
proceed with more confidence in that you are not going to be held responsible for the data that
was (or still may be) on the hard drive. The technical stuff is basic. You should be carrying with
you a working boot disk. I like to use my Windows 95/98 setup disk, the one that gives me
several options on how to proceed after the initial boot process. I first look at the drive
(physically) and get as much information off of the case as I can. Usually it will include how
many cylinders, heads and sectors the drive has. I then go to Setup and input this information
in the BIOS. Saving that information I boot to my trusty boot disk and hopefully get an A:
prompt. Using Fdisk I look at the drive and see if it still has a DOS partition. If it does then I
switch to the C: prompt (if it comes up, I then breath a sigh of relief) and look at the directory
tree. This only means that I can see the FAT (File Allocation Table), I still may not be able to
actually get data off of the drive. Remember that while you are working in DOS, your file
names are in the 8.3 format and any information transferred in that environment will be
rendered difficult if not impossible to use. So let's say that we have been blessed with a good
Directory Tree and we can see the desired information buried somewhere on
the platters. Knowing I will most certainly replace the drive, I need to attempt to get the data
off of it in its most usable form, which will be in a Windows long filename format. At this point,
I'm excited enough about becoming the local computer hero that I pull the drive out and hook
it up as a secondary drive to a working system. I then boot the second system and see if it will
recognize the drive. It does, and I happily transfer the data to a directory on a good drive
where I can either burn a CD for the customer or restore it to the new drive when it becomes
available. If you have no second drive available to install the failing disk drive into you are now
going to have to make a tough decision. This premise is far too vague for actual fieldwork but
we will say that the customer has no idea where his data resides on the drive (the usual case).
Being familiar with the 8.3 format you then ask him what the last 3 letters of the filename was.
He states .doc or .wpd or some other format. You do a file search such as "dir *.wpd/s" which
will then search the drive and its subdirectories for the data. You then can move the data off of
the drive to a floppy and even though you may have to rename all of the files back to whatever
they were originally named, you have the praise and honor of being the guru of
the moment. At best you can hope to get some data for the customer, at worst you now have
a working knowledge of
the system and its integrity (i.e. controllers, peripheral cards, etc.) You also have the
opportunity to discuss valid backup options with the customer. But that is another story.
From: Joe Blackledge
Recently I had a machine that would lock up at various times. This was a critical PC attached to
a piece of measuring equipment. There was a lot of time invested in writing the measurement
programs stored on the hard drive. Of course, the programs were not backed up. I discovered
that if it was cold (had been off for some time), it would run for about 10 minutes. To make a
long story short, it was a heat-related problem with the hard drive. The warmer the machine
got, the shorter the period of time the hard drive would work. I opened the case up, took the
hard drive out, stretched the cables out and replugged them so the drive was outside the
machine. I found two large zip-lock bags and filled them with ice. I laid the drive on one of
them and placed the other on top. I waited about 30 minutes for it to get real cool. I connected
