[Infrastructures] AFS in an infrastructure

[Kyle Moore]

Is Campbell's AFS book still relevant and current to bother with?

If AFS is so great, which it sounds to be, why is it not very popular? 
It sounds like it has a lot to offer but this list is the first place I 
have heard of it.


Steve Traugott wrote:
> On Mon, Mar 14, 2005 at 10:36:15AM -0600, Sean Kelly wrote:
> 
>>Now I'm trying to vision how all the pieces mentioned on
>>Infrastructures.Org fit into an AFS world.
>>
>>* Version Control could be done over AFS. You could store the
>>  CVS (or Subversion fsfs) repository on AFS.
>>* Host install images could be stored on AFS and thus be made available through 
>>  many different machines running BOOTP/DHCP/whatever.
>>* Directory Services could be done using LDAP or Hesiod, depending on which 
>>  route you went down.
>>* Authentication would be done through Kerberos.
>>* Network File Servers are AFS file servers
>>* File Replication Servers are just copying things off AFS
>>* Client File Access is AFS
>>* Client O/S Update could be done with tools pulling patches off AFS
>>...
>>
>>Am I going overboard with the AFS thing here? Where does the "Gold Server"
>>fit in when you have AFS and can just store everything in a
>>replicated/backed up distributed common namespace?
> 
> 
> You're not going overboard -- the work that we were doing in the early
> 90's that led eventually to the "bootstrapping" paper and
> Infrastructures.Org would have taken a very different tack if OpenAFS
> had been available back then.  As it was, we evaluated AFS and found it
> lacking in terms of popularity, manageability, compatibility, and just
> didn't get a warm fuzzy feeling from it.  And it was expensive.  And
> then there was the 2G filesize limit...  
> 
> Since the gold server is just a passive thing that serves files via
> CVS/SVN, rsync/SUP, and NFS, a lot of that can be replaced.  The trick
> is avoiding the circular dependencies so you can manage the AFS servers
> themselves, and getting the client authenticated to Kerberos in early
> boot so you can do pulls at that time.
> 
> The main reason we need to write ISFS is because we still can't count on
> AFS being in very many infrastructures, and to prevent circular
> dependencies for those shops that do have AFS servers and want to use
> ISconf4 to manage them.
> 
> Steve

-- 
Kyle Moore