Subject: Re: Command 0xNN completed with result code 0xNN workaround?
To: Gunnar Helliesen <gunnar@bitcon.no>
From: Boris Gjenero <bgjenero@undergrad.math.uwaterloo.ca>
List: port-vax
Date: 01/28/1998 00:03:56
Gunnar Helliesen wrote:
> First of all I did the "low-level format dance":

Hey, at least it's a VS2000.  The formatter is nice and straightforward
and it actually works... unlike the RQDX3 stuff which may refuse to
format unformatted drives and even convince you to format the wrong
drive.

<some t 70 output snipped>
> 
> VSfmt_STS_BBRvec := 5
> 
> VSfmt_RES_Succ
> >>>
> 
> If I understand this correctly the formatter found 5 new bad blocks not
> in the manufacturer's bad block table? And the problem is that NetBSD
> doesn't understand this bad block table that the formatter has created?

I think that 5 blocks is the total number of blocks it has redirected. 
However, I could be wrong because I haven't been paying close attention
to what happens when the formatter actually *does* find the
manufacturer's bad block table.  

Yes, the problem is that NetBSD doesn't understand the table.  What's
worse, NetBSD can overwrite the table because it can write to the whole
physical disk including the table itself.

> Ugh. Linux? OK, if it'll help. How's that done? I can't find anything
> about specifying the fs type in man newfs. Do I enter it in the disktab
> entry?

I have absolutely no clue as to whether this would work.  It was just an
idea.  First of all, there doesn't seem to be any NetBSD program that
makes ext2fs filesystems.  Maybe e2fsprogs will compile on NetBSD/vax. 
Then there's the problem of kernel support.  I do have mount_ext2fs, but
I don't think I have support in the kernel.  It isn't even in GENERIC,
and I don't know if it works with NetBSD/vax.
(If anybody has any useful comments on how insane any of this sounds,
please do tell me. :-)

> man badsect talks about identifying the bad sectors through messages on
> the console. How? Is that part of the "command 0xnn completed..."
> message?

You'd could:
- Read the bad sector list to find bad sectors
- Find or write a program that tests each sector

IMHO, those "command 0xnn completed..." error messages really suck. 
They should tell you the location but they don't.  They don't even tell
you the what is causing them.  They really need improvement (hint to
whoever is responsible).

> Thanks for all your help!

You're welcome.

-- 
|  Boris Gjenero <bgjenero@undergrad.math.uwaterloo.ca>              |
|  Home page:  http://www.undergrad.math.uwaterloo.ca/~bgjenero/     |
|  "Luke, you're going to find that many of the truths we cling to   |
|  depend greatly on our own point of view." - Obi-Wan Kenobi, ROTJ  |