Re: Proposal: a consistent mount interface

Matthias Urlichs (
27 Nov 1996 23:27:19 +0100

In, article <>,
Maciej Stachowiak <> writes:
[ 2. Have a different mount program for each remote file system ]

> However, the second solution does have some advantages. Here they are,
> along with my objections to them. First, it allows arbitrary
> interesting things to be done in user space at mount time. However,

There are cases where this won't work. For instance, let's say my superFS
mount call requires some public-key authorization, using my personal
authentication daemon on my secure machine (look at ssh).

I do NOT think it's be a good idea to pass what's essentially an open file
descriptor (ssh can do it that way) through a mount program and kerneld to
a mountd thing which doesn't have access to the user's environment and thus
may not be able to determine whether the user is authorized to access that
particular file system.

> using kerneld, this can be achieved anyway. Second, if each fs type

Since mount is the only program which does this, kerneld is unnecessary.

If practice makes perfect, and nobody's perfect, why practice?
Matthias Urlichs         \  noris network GmbH  /  Xlink-POP Nürnberg 
Schleiermacherstraße 12   \   Linux+Internet   /   EMail:
90491 Nürnberg (Germany)   \    Consulting+Programming+Networking+etc'ing
   PGP: 1024/4F578875   1B 89 E2 1C 43 EA 80 44  15 D2 29 CF C6 C7 E0 DE
       Click <A HREF="">here</A>.    42