Re: [nfsv4] Asking for details about OPEN_DOWNGRADE

["J. Bruce Fields" <bfields at fieldses.org>]

On Wed, Jun 25, 2008 at 04:17:23PM -0400, Rick Macklem wrote:
> Are you saying that servers will say NFS4ERR_ACCESS or NFS4ERR_IO in
> reply to an Access Op for ACCESS4_EXECUTE? Or, are you simply noting
> that the client will have to locally decide if a file still opened
> against the server rw can be executed? (The latter seems to be a
> client side implementation issue. In my client code, the relationship
> between local opens and Open state on the server might be descibed as
> "loose" and this is not an issue. As you might have guessed, my client
> never does Open_downgrades.)

At least in the case of the linux server, if an nfsv4 client has a file
open for write, then the server holds that file open.  So if a client
has a file open for write on the server, then a user on the server that
tries to execute that file will get ETXTBUSY.

So in theory failing to open_downgrade could lead to spurious
ETXTBUSY's.

--b.
_______________________________________________
nfsv4 mailing list
nfsv4 at ietf.org
https://www.ietf.org/mailman/listinfo/nfsv4