Re: AFS - The Ultimate Sulution? -- What is the point?

["Brandon S. Allbery KF8NH" <allbery () ece cmu edu>]

On Sep 17, 2006, at 10:03 , Valdis.Kletnieks () vt edu wrote:
> Go back and re-read the last few batches of AFS updates, and ask  
> youself
> for each bugfix "Could this *potentially* have been leveraged by a  
> clued
> hacker?".

I haven't noticed many issues beyond potential denial of service  
attacks --- which are mitigated to some extent by replication (of  
course, someone could go after *all* the servers...).  The biggest  
problems at this point are:

- if you get the afs/cell@REALM key, you've got the entire cell
- no data encryption to speak of (fcrypt?  it is to laugh)

Work is being done on both fronts, although I'm not the right person  
to speak to about either.

In any case, you need to lock up your DB and file servers as tight as  
you can if you want the cell to be at all secure.

(Unfortunately, I don't think anyone has, other than inadvertently,  
tested how AFS reacts to invalid packets.  One of those things I'd  
love to do if I ever got a few round tuits....)

-- 
brandon s. allbery    [linux,solaris,freebsd,perl]     allbery () kf8nh com
system administrator [openafs,heimdal,too many hats] allbery () ece cmu edu
electrical and computer engineering, carnegie mellon university    KF8NH



_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/