[OpenAFS-devel] AFS/DFS Mount points

Christer Bernérus bernerus@medic.chalmers.se
Sun, 10 Apr 2005 20:08:39 -0700


On 10 apr 2005, at 18.50, Jeffrey Hutzelman wrote:

>>
>
> Fortunately, at least for AFS, no transition is required at all, 
> because the problem simply does not exist.  While it is true that AFS =

> mount points are represented as symlinks whose target begins with # or =

> % (denoting the type of mount point), the thing that makes them mount =

> points and not symlinks is their mode.  It is possible to create 
> symlinks to targets starting with '#' or '%' by making an 
> RXAFS_Symlink() call, and clients will treat such links correctly.  It =

> is simply not possible to create such links using the symlink() 
> syscall on an OpenAFS client, because the cache manager interprets 
> such requests as requests to create mount points.

This, IMHO, seems to be a very odd way of implementing things. If I get 
this right, first you create a new type of object in the file system, 
then you make a special RXAFS_api to be able to create the old type of 
object that would otherwise have been created, then you map over an 
existing API to create the new type of object, and finally create an 
administration application to use for creating the new type of object 
that internally will have to use the *old* API. Phew!

Anyone left with DFS running that can check if it works the same there, 
or was this yet another AFS weirdness that was straightened out when 
DFS was created ? (See Dean's mail for example).
>

----
Christer Bernérus, SM6FBQ
IT services
Infrastructures support group
Chalmers University of Technology
bernerus@medic.chalmers.se