ISC 2.0.2 Installation Help Needed

Tom Yager tyager at maxx.UUCP
Fri Jun 29 02:32:43 AEST 1990


In article <4524 at cvl.umd.edu>, liuqing at cvl.umd.edu (Larry Liuqing Huang) writes:
> I am trying to install ISC 2.0.2 on a 386 machine with a 120 meg hard disk.
> After the "surface analysis" it says "too many bad sectors" and terminated
> the installation procesure.  I can not cut the hard disk into pieces of
> small partitions.  
> 
> 1).  Is there any way to get around this problem?
> 2).  Is this fixed in the new release of ISC 2.2?
> 
> Thanks for your help.
> liuqing at cvl.umd.edu

It's hard to diagnose this because you don't mention what type of disk
controller and disk drive you're using.

The surface analysis utility will report as bad any sector that it can't
write with a known value and read back. The inability to write a sector is
not necessarily a symptom of a disk drive bad spot. Here's a few things you
might check:

  - Make sure you're using a controller supported by 386/ix. That includes
    MFM, most ESDI and RLL controllers, and Adaptec and Future Domain
    SCSI controllers. If you're using ESDI or RLL, the controller must
    provide exact emulation of a Western Digital MFM controller.
  - Take the time to check (again) for loose cables. Even if they look tight,
    pull the connectors off and put them on again. If they don't fit snugly
    over the pins on the controller or drive, replace the cable. Same thing
    for any visible damage to the cable itself. Also look for bent pins on
    the controller board.
  - Double-check the interleave at which you formatted the drive. It must
    match the interleave factor you provided during 386/ix installation. If
    you didn't do a low-level format of the drive, consider doing it and
    specifying a known interleave.
  - Check your system's bus speed. Many disk controllers expect to be run in
    a standard AT bus: 8 MHz or less. If your system has jumpers or SETUP
    parameters controlling bus speed, make sure you're running at a standard
    speed.

If you want to narrow the problem down, there's two things you might try.
First, install DOS on the system. Doing a DOS format on a partition does a
rudimentary bad track check. When the format finishes, use CHKDSK to find
out how much space is taken up by bad tracks. If it's high (say, more than
10%), be very suspiscous of your hardware.

Lastly, next time you try to do a surface analysis with the 386/ix install,
write down the first few sectors reported bad. Run the analysis again, and
compare the new numbers with your list. If they match, it's likely you really
do have a bad drive. Unless, that is, the analysis simply reports all the
sectors as bad.

If it's any consolation, I've run into problems like yours before (hence the
long list of suggestions), but it's always turned out to be something easy
to fix.

Good luck!
(ty)

-- 
+--Tom Yager, Technical Editor, BYTE----Reviewer, UNIX World---------------+
|  NET: decvax!maxx!tyager     -or-     uunet!bytepb!maxx!tyager           | 
|  I speak only for myself           "UNIX: It's not a job,                |
+-------------------------------------it's a Jihad!" -co-worker------------+



More information about the Comp.unix.i386 mailing list