[Not quite] Dead hard drives [COLUG]

[Jim]

Kent Broestl wrote:

[> Jim wrote:]

> >    dd if=/dev/hdx | tee hdx.0 | md5sum
> >    dd if=/dev/hdx | tee hdx.1 | md5sum
> >    dd if=/dev/hdx | tee hdx.2 | md5sum

> > Do it several times and compare them. 
> > If the copies are identical, then your hardware is likely OK. 
> > If the copies are different, then your drive likely has hardware 
> > problems. 
> 
> Luckily the size of partitions is just right... 

I don't know what you mean by the size of partitions 
being "just right". 

> I did the dd if=/dev/hdf1 | tee hdf.0 | md5sum a couple of times, 
> and the sums match. 

That's good. That means that the data on the drive is stable. 
That's also, _all_ that it means. 

By the way, I was talking about doing the whole _drive_ (/dev/hdf), 
not just a mere _partition_ (/dev/hdf1) of the drive. 

> Looks like the data is still good. Yay!

You do not know that the data is good or ever was good. 
The dd/md5sum stuff just tells you that the data is 
_stable_, tending to indicate that the _hardware_ is OK. 

> > Those high drive letters give me the willies. 
> > I worry about wierd RAID stuff when I see /dev/hd[e-z]. 
> > RAID can complicate things just a "little bit". 
> > Is any RAID stuff anywhere near this drive? 
> >   
> Good question... The motherboard "supports" raid (ABIT KT7A with 
> Highpoint RAID -- 

"Highpoint" sounds like a marketing jive name. 
Look on the motherboard. What is the chip number? 

> I don't know if anybody knows any caveats with 
> Highpoint) but I never designated the drive as part of an array either 
> with Linux or through the Highpoint interface at boot time.

Good. 

> Oddly enough though, I noticed that when I compiled my kernel back in 
> spring, the box wouldn't mount hdf1 at boot time without having raid 
> modules compiled for the kernel.

Hmmm. I don't know what to make of that. 
Others who have wrangled with cheap RAID will have to comment. 

> > Kent Broestl wrote:
> >> And here is the partition table for /dev/hdf:
> >>
> >> Disk /dev/hdf: 16 heads, 63 sectors, 79656 cylinders
> >>
> >> Nr AF Hd Sec Cyl Hd Sec Cyl Start Size ID
> >> 1 00 1 1 0 15 63 1023 63 80293185 83
> >> 2 00 0 0 0 0 0 0 0 0 00
> >> 3 00 0 0 0 0 0 0 0 0 00
> >> 4 00 0 0 0 0 0 0 0 0 00

> This is the output of...
> fdisk /dev/hdf
> x (expert commands)
> p (print partition table

What version of fdisk did you use? 
What version of Linux did you run that fdisk under? 
What command line options did you specify to fdisk? 
How did you get the option of fdisk into email? 
Did you (re)type it or did you copy and paste it? 

Jim