Re: ShadowFS (status)

[Marcus Brinkmann]

On Fri, Aug 03, 2001 at 03:53:17PM +0200, Neal H Walfield wrote:
> Let us take a look at what others have done.  E.g.:
> 
>  - http://www.cs.helsinki.fi/linux/linux-kernel/Year-2000/2000-23/0791.html

Have you actually read the discussion?  It doesn't seem to be helpful to us.
Busy is a concept that is defined by translators themselve.  We hold (or
should hold) ports to the underlying filesystems, so they will all be
"busy".

If you just wanted to point out that there are other union mounts
implementations, well, that's true.  The BSD guys have one, too.
It is certainly interesting to look what they can done, but intuitively I
would guess that they have different problems and are more limited by kernel
constraints.

Of course, what we really are interested in are union mounts from a posix
point of view (not the implementation details, but the conceptuation).
If there is an article or report on that, this would be good.  Maybe plan9
has something about it? (They have user level filesystems, as far as I
know).  Some web search should be done.

Something very interesting is deleting files.  Related to this:
"whiteout" filesystem.  All files appearing in this filesystem result into
clearing out the files with the same name in the shadowfs (so they negate
the effect of a file).  So there are some very nice things that can be done.
Later, probably ;)

Thanks,
Marcus

-- 
`Rhubarb is no Egyptian god.' Debian http://www.debian.org brinkmd@debian.org
Marcus Brinkmann              GNU    http://www.gnu.org    marcus@gnu.org
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de