Node cloning

Robert G. Brown rgb at phy.duke.edu
Fri Apr 6 13:27:34 PDT 2001


On Fri, 6 Apr 2001, Greg Lindahl wrote:

> On Fri, Apr 06, 2001 at 10:11:27AM -0400, Josip Loncaric wrote:
>
> > BTW, we found that about 10% of our IDE hard drives (mostly early
> > Seagate 7200rpm UDMA models) had either too many bad blocks or bad
> > blocks in unacceptable locations (like the swap partition) and had to be
> > replaced.  We now use a combination of Seagate and IBM drives, and over
> > the past two years about 20% of them have developed at least some bad
> > blocks that we had to map out using the 'e2fsck -c ...' command.
>
> How interesting. Centurion I and II have 2 TB of disk (mostly 5400 rpm
> IDE), and we've never had to manually do that.
>
> BTW you can "mkswap -c" to mark bad blocks in swap. I don't know why
> you'd find a bad block in swap less acceptable than a bad block in the
> filesystem.

I think that this is a matter of luck more than anything else.  I've had
plenty of bad luck with IDE drives (for a while I thought I was
literally cursed or that the building needed a monopole-sweeping).  Of
the 7-8 drives that failed (out of only 20 or so) a number just died but
several not only had bad blocks that needed to be mapped but
continually, gradually, slowly, developed more.  I'd run for eight weeks
and have a hard crash and find that more bad blocks had developed.  This
went on until I got all the drives finally replaced.

In more recent years, I haven't had this experience, but I suspect that
it is strongly related to "lemon"-type distributions of bad hardware.

   rgb

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu







More information about the Beowulf mailing list