Re: [nfsv4] [FedFS] Multiple FSL resolution

["Everhart, Craig" <[e-mail]>]

As a quick shot, I believe that section 3.2, step 5, is in error, at
least for the case of NFSv4{,.1}, which is the domain of discourse here.
All the FSL's for a given FSN should be converted and sent to the file
server client, and the client can choose among them.

 

There are possible opportunities for introduction of policy-controlled
behavior if there are more FSL's than a given server can convert or than
it can communicate to the file server client, but the idea is that the
file servers take all locations from the NSDB and present them all to
the client for its choice.

 

                                Craig

 

From: [e-mail] [mailto:[e-mail]] 
Sent: Wednesday, December 09, 2009 9:37 PM
To: [e-mail]
Subject: [nfsv4] [FedFS] Multiple FSL resolution

 

Either an easily answered question or a possible topic for discussion in
tomorrow's FedFS call.

 

The requirements spec
(http://www.ietf.org/id/draft-ietf-nfsv4-federated-fs-reqts-06.txt)
talks about associating multiple FSL's with an FSN where the file sets
referred to by the FSL's are replicas of one another.  Maintenance of
the replicas is outside the scope of FedFS.    

 

When a client traverses a junction that refers to an FSN which has
multiple FSL's associated with it the file server queries the relevant
NSDB for the FSLs associated with the FSN (section 3.2, step 4).    But
only one of the potentially multiple FSLs is returned to the client in
the form of a protocol specific referral (section 3.2, step 5).      

 

How does the server decide which of multiple FSLs to return to a client?

 

Thanks,

Chris

-- 
Chris Stacey - Senior Technologist, Celerra Unified Storage Engineering,
EMC Corporation (New Zealand).  Phone: [phone]