[OpenAFS] Timeouts, Client-Configuration & AFS-Filespace on Client-PCs

[FBO]

Hi AFS-Fans,

3 little questions:

1. When one of my fileservers is down and I try to access a volume
   on it ("ls -l"ing a directory with a mountpoint to that volume)
   ls waits and shows something like "ls : 'volume': connection timed out".
   Is it possible to reduce that timeout till that message is
   shown? 5 seconds would be very good...

2. As far as i understood AFS, putting a key (with asetkey) on a host
   means that if someone manages to be root on that host,
   my whole Cell is compromised. Is that correct?
   Is there another way to put available disk space on client-PCs
   (=pc's on regular user's desktops)
   into AFS, maybe with a second cell?

3. How to I estimate, what afs-client-configuration
   ( -stat xxx, -dcache xxx, -daemons xxx, -volumes xxx, -files xxx)
   is best for me?
   I would be interested in expressions like:
   
    "The more ''-files'' the better"
    
   or
   
    "less ''-dcache''s if local hdd is slow, more if it's fast"
   

   How much cache size (I mean the last field in /etc/openafs/cacheinfo)
   is reasonable for Clients accessing only the local cell (100MBit-Lan).

Thank you very much for any answer and suggestion.

FBO