Disk error and recovery

[Leroy Dorman]

In article <1090 at maynard.BSW.COM>, campbell at maynard.BSW.COM (Larry Campbell) writes:
> In article <180 at focsys.UUCP> larry at focsys.UUCP (Larry Williamson) writes:
> <>I saw this disk error on my console this morning. Now I'm worried.
> <>
> <>Disk I/O Error: Fun: 30  Cy: 220  Hd 4 Sec 12  Status: 51 Estat: 80 Drstat: A4
> <>
> <>Which drive had the I/O error? What do I do to fix the file that is now
> <>corrupt? How do I find the file that is corrupt? I assume that I should
> <>add this location into the bad sector map, but can I do this now without
> <>re-building the system?
> 
> I had the same problem and called Microport.  Their answer (this is no joke!)
> was "Just use dd to read the entire disk -- chances are you can reproduce
> the error this way, and that'll tell you which disk had the error."

> They also had nothing to say about how to find out which file might have
> been affected, and whether the bad sector would be remapped or not.
> 
dd will tell you what disk block (in logical blocks) that has the
problem. Then, fsdb can be used to reattach the badspot someplace else
(like .Badblock in the root directory)  This is, of course, not
trivial.
Why the drive number (even the minor number) isin't there 
is really pointless. You might as well not report the error!.

> I am serious.  This is what they said.  This is what they call "support".

They once suggested that I recompile regexp() with 'all the ints changed
to longs' to fix a large model bug.  Highly professional. This warrants
a real scream from the users. This is the '80s, and errors like this and
'Bad Code' diagnostics shouldn't need be tolerated.

------------------------------------------------------------------
Eric Dorman				Kirk- "Wheels, Mr. Spock"
University of California, San Diego	Spock- "A flivver, Captain"
Scripps Institution of Oceanography	  From- A Piece of the Action"
 siolmd!eric at sdsioa.ucsd.edu
 mp1 at sdcc12.ucsd.edu  Attn: eric
 dorman at mplvax.nosc.mil  Attn: eric