[RFC,PATCH 0/4] Dynamic Pseudo Root

[Trond Myklebust]

On Mon, 2007-12-10 at 14:37 +1100, Neil Brown wrote:
> On Friday December 7, trond.myklebust at fys.uio.no wrote:
> > 
> > Agreed. You are almost forced to do something like this if you want the
> > filehandles to remain valid when the administrator changes the export
> > options to allow access to path components that were previously part of
> > the pseudo-fs. Otherwise, you have to use volatile filehandles, which
> > are poorly supported by existing clients.
> 
> I must confess that that is an argument that really bothers me.
> 
> If we do not have multiple independent working implementations of
> volatile filehandles, then they should not be in the spec.
> And if we do, then the server should be allowed (even expected) to use
> them.

They should never have been in the spec: the whole idea of volatile
filehandles is broken beyond belief. However, the main point is that
whether or not we plan on supporting the damned things in the future,
they are currently _not_ supported by the entirety of clients that are
out there.

Trond