[comp.sys.3b1] Hard Drive problems...

sardella@tdw254.ed.ray.com (01/28/91)

I been having some problems with my unix-pc.  First the smgr would die after
running for a little while and the phone manager would not run properly.
I reformatted the hard and ran that check that takes about 20 minutes.  In
the check that took 20 minutes it gave some bad block(5) and supposely fix them.
I installed the founddation set (3.5) , unix utilities, phone manager and other 
systems software.  Then after a while I use a command like ls and
system would come back with illegal instruction and other weird message.  I sync 
the disk and ls would work again.  
   I guest the question is, Is my hard drive trashed and is there any way to get
around this?  Before I buy a new hard drive I want to make sure this is the
problem.  I going going to reformat my hard drive and try installing software
again but I don't think that will solve the problem.

-- 
Norman "STORMIN" Sardella 
Work: sardella@sud509.ed.ray.com
Leisure: sardella@world.std.com 
Home: uunet!world!stormy!norm

n138ct@tamuts.tamu.edu (Brent Burton) (04/25/91)

As I posted yesterday, my drive came in.  I'm not worrying about the
preservation of the old software - I've that figured out now.

I'm having a problem getting the drive formatted and creating the filesystem.

Drive details:   Heads: 7   Cyl: 1314   (80Mb MFM)
Bad Blocks (from factory):
    CYL   HEAD   BYTE LOCATION
   208     4      1635
   515     6      257
   693     0      2385
   791     5      4410
   792     5      4571
   793     5      4572
   947     6      6595
   1068    0      6745

I format the drive and enter the above info into the bad block table.  (BTW,
I'm using the 3.51 Diagnostics disk.)

The surface test reveals CYL 207, H 4, Sector 2 is bad and it's relocated.
This is all the surface test finds.

After the formatting and tests, the drive has 73584 total blocks (1K logical)
with Partition 0 having 52, Par 1 having 5000 (swap), and the userspace
has 68528 blocks.  Are these numbers supposed to sum to 73584?  If so,
why don't these?  Hints?

The formatting appears to work OK - once the bad blocks are entered, the
read tests work fine, as do the next random seek tests.

The main problem is with the Floppy Boot disk and it's Filesystem disk.  When
these disks are used to install the new OS, mkfs fails about a minute into
its work with a "write error at 68328".  This block, which corresponds to
Cyl 1220, H 1, Sector 0, is entered (by me) into the bad block list via the
dianostics.  However, when I try the mkfs again, it fails at the same spot.

Can anyone offer suggestions or other knowledge I might need to know to pull
this off?  I've formatted raw drives before on other systems, but this 7300
is just some new territory.  Please email any responses.
Brent Burton
  n138ct@tamuts.tamu.edu