Michael Hennebry wrote:
> What would dd do?
> I've never had much of a clue what, besides some reformatting,
> dd does that cp does not.
> E.g. if there is no file, there is nothing to which one can set
> if and of, hence dd would copy from standard input to
> standard output.
> I assume there is something wrong with that logic,
> but I don't know what.
Ermm... oops?
I seem to have been labouring under a misapprehension.
I assumed that when you cp a device node, what you get is a copy of the
device node. (And if you use cp -R, you get just that).
cp -R /dev/hda7 /tmp
ls -l /tmp/hda7
brw-r----- 1 root root 3, 65 May 20 16:49 hda7
I thought that was what normal cp did, too.
Instead, cp opens the device node for reading, and copies all the
output. So cp /dev/hda7 /scratch will create a *very* large file in
/scratch. If the appropriate partition contains a filesystem, you'll get
a copy of that filesystem which you should be able to loopback mount.
Thanks for making me check that and learn something,
James.
--
E-mail address: james | ... more holes in Internet Explorer than
@westexe.demon.co.uk | Blackburn, Lancashire...
| -- http://theinquirer.net/?article=17235
