[OpenAFS-port-darwin] Status?

[Derrick J Brashear]

On Tue, 29 Jun 2004, Steve Lidie wrote:

>>> r.e. the above installer:  it added an afsd.options file which totally 
>>> hosed/confused me: AFS didn't know which realm it was in, hence there 
>>> were many missing mount points as well as a hundred or so strange new 
>>> ones.  Does .dynroot ring a bell?
>> 
>> Sure, it replaced your CellServDB and then used it to populate a fake 
>> root.afs volume. All of the cells you care about published themselves, 
>> right? ;-)
>> 
>
> It didn't replace CellServeDB, but I've just learned that it uses CellServDB 
> to make root.afs ;)  The *real* problem was we have an alias for the cell 
> name and it was that missing short name that broke things.  Apparently w/o 
> the dynamic root option the symbolic link would be present .... if I 
> understand the AFS admin's explanation...

Yes, that's true. However, if you drop a CellAliases file in 
/var/db/openafs/etc you can make it put readwrite symlinks in the fake 
root volume

The fake root volume is very useful if you have mobile clients (on those 
clients)