"Derek D. Martin" wrote:
> At some point hitherto, Benjamin Scott hath spake thusly:
> > On Sat, 9 Feb 2002, mike ledoux wrote:
> > > # tar cBlf - . | (cd /mnt/tmp && tar xBspf -)
> >
> > Why not "cp -a"? Are you just used to old-school Unix, or is there
> > something wrong with "cp" we should know about?
>
> Even on newer Unix systems, IIRC, cp can't handle certain file types,
> like sockets/fifos, and device files. For that reason, unless one is
> positive there are none of the above, tar is safer.
I've used cp -a with no problems from several years now. I used to use the
tar pipe construct until I confirmed cp worked to my satisfaction. I've
copied /dev with no problems at all.
If copying /, do remember to exclude /proc by cp -ax (-x is equivalent to
the one file system -l of tar).
>From the man page for cp:
-a = same as -dpR
-d = preserve links
-p = preserve file attributes
-R = recursive
Specifically, there is a note about -r saying:
-r copy recursively, non-directories as files WARNING:
use -R instead when you might copy special files
like FIFOs or /dev/zero
While I've never encountered a problem, sparse files might be created
differently.
Again from man cp:
By default, sparse SOURCE files are detected by a crude
heuristic and the corresponding DEST file is made sparse
as well. That is the behavior selected by --sparse=auto.
Specify --sparse=always to create a sparse DEST file when�
ever the SOURCE file contains a long enough sequence of
zero bytes. Use --sparse=never to inhibit creation of
sparse files.
begin:vcard
n:Jenkins;Dan
tel;fax:1-603-627-7513
tel;work:1-603-627-0443
x-mozilla-html:TRUE
url:http://www.rastech.com
org:Rastech Inc.;*** Technical Support Excellence for a Quarter Century
adr:;;21 Curtis Lane;Bedford;NH;03110;USA
version:2.1
email;internet:[EMAIL PROTECTED]
title:Technical Director
fn:Dan Jenkins
end:vcard
