raid 0 problems

[Technomage]

On Monday 27 March 2006 07:11, Alan Dayley wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Nathan England wrote:
> > You are right, the raid software in linux wrote 1 drive then the next. I
> > think this is partially why the LVM2 stuff took over and that is being
> > used, while it seems to me the software raid has almost dissappeared.
>
> That is different.  Strictly speaking, what you describe here is *not*
> RAID 0.  It is some kind of logical volume thing.
>
> The good news is that, in this case, all the data on the good drive
> should be contiguous and intact except for the end portions that are
> "completed" on the bad drive.  If you can get the filesystem structure
> to no longer "point" at the bad drive, you would only loose the data on
> the bad drive.

yep. thats the rub. getting it to "think" it was truncated....

>
> I don't know how to do that though so maybe this whole comment is a time
> waster for you.  ;^)
>
> When buying torx driver bits, spend the extra for one with good quality
> steel.  I have purchased cheap ones only to have them twist and break in
> the screw.

that is a given. I know about the cheap tools (caused no end of problems when 
being used).

harbor freight tools is a pretty good source for good tools at *lower* 
prices. :)