From: Paul Curtayne (niche at domain tinet.ie)
Date: Mon 30 Aug 1999 - 22:47:49 IST
OK, I'm a little stuck and it's all M$'s fault.
My question is general and is about how bad blocks might develop on a
hard-disk, but I feel like ranting at MS!! If I didn't have to patch
MS Office none of this would have happened!!!
A client has a custom Access database, and thanks to known major bugs
in Access their data became corrupted irrevocably. (The problem
apparently lies in the 'repair database' utility!!!)
The MS recommended route is to patch Office with SR1, SR2, and patches
for some DLL's and the JET and use a backup as the data is gone for
good in this case.
So, I went to do this on a 486 with 36MB RAM, and an almost new 2GB
disk. This install of Office didn't even have SR-1 installed. Then the
installer for SR-1 crashed.
When I restarted Scandisk started to report bad blocks everywhere,
more than 150 of them. Mostly they were at the very end of the used
part of the HD, but other places too.
Scandisk started marking them as bad. This was unbelievably slow, and
when it got to the latter parts of the disk which contained no data, I
aborted it.
Incredibly Win95 booted, and all programs seem to run ok, even though
there are so many bad blocks.
My clients are naturally perturbed as it is a new-ish HD, and they
have had lots of miscellaneous problems recently. They are pressuring
me heavily for an explanation, and I need reassurance that my patches
didn't cause damage.
I've told them that I don't believe a crashed program could really
physically damage 150+ blocks at several places on the disk.
Or can it?????
Question: apart from the disk being a factory dud, what explanations
can I offer? What causes bad blocks?
Dud IDE controller? Head crashes? (what causes these? Power surges??
banging the case?)
I should add that I have now removed and recreated the partition. I
re-formatted it with the /C option in DOS to test blocks already
marked bad. This format seemed to detect even more bad blocks, and
after that I ran scandisk again, and it is detecting even more bad
blocks!!
I think Scandisk is corrupting the disk!!!
AAAGHHGHHH!! Please help!
This archive was generated by hypermail 2.1.6 : Thu 06 Feb 2003 - 13:04:31 GMT