[rescue] netbsd error

[Joshua D Boyd]

On Wed, Feb 13, 2002 at 08:55:32PM -0500, George Adkins wrote:
> Sure, you could use the scsi control commands to force sparing of the 
> affected blocks, which if the automatic sector sparing has already used the 
> spares for that track would cause the whole track to be spared, which if the 
> spare track for that zone has already been used will cause the entire 
> cylinder to be spared, yadda, yadda, yadda...

I have no idea what what you just said means.  Time for some googling I guess.
 
> How important is the project?  I mean once those drives start with that 
> stuff, it only gets worse.  About nine months ago, I had a production machine
> that started throwing one of those every month or so, then it one or two 
> every month, then the machine would lock up and have to be powercycled 
> occasionally, and the bad blocks reassigned, then it was happening once every
> week or two, and then within the space of the last month, the machine just 
> became unbootable, even hammering on it from a VT...
 
> Good thing I had tape backup, and the machine wasn't too important (NAT box 
> at work, so the secretaries can surf the web...)

I seem to be getting a lot of these errors, although they all have the info
field and command info entrys.  

No data of importance is going to be stored on this box since it's job is
also to be a NAT.  However, my options are to make do with this drive, use
a 200meg drive, or wait at least a month till I can buy a new drive.  The
200meg drive idea wouldn't be too awefull if I could better configure my NFS
server.  Sure, as a normal user I can do compiles on the NFS shares, but when
I switch to root to do the make install step, a lot of packages try to create
files in that directory, and I can't seem to convince the NFS server that
I mean no_root_squash when I say it.  If only I had more spare drives.

-- 
Joshua D. Boyd