Subject: Re: nfs servers and 5 minute VOP_READ's
To: Frank van der Linden <frank@wins.uva.nl>
From: Bill Studenmund <wrstuden@nas.nasa.gov>
List: tech-kern
Date: 03/16/1999 15:18:10
On Wed, 17 Mar 1999, Frank van der Linden wrote:
> On Tue, Mar 16, 1999 at 01:55:26PM -0800, Bill Studenmund wrote:
> >
> > How did the patches to add server support look? Obviously they aren't
> > right for a jukebox as-is, but they let the fs say "try later."
>
> The patch looks okay. I just noticed that there's a bug in the server code
> that doesn't even allow it to return the JUKEBOX error at the moment..
> lemme fix that first.
Ok.
Jason actually had some concerns. Mainly that the patch does nothing for a
v2 mount.
I think a better thing to do would be something like
error = VOP_READ/VOP_WRITE( ..., IO_NODELOCKED | IO_NDELAY, ...)
if (error == EAGAIN) {
if (v3)
error = NFSERR_TRYAGAIN;
else {
error = drop me;
}
}
I'm not sure what "drop me" would consist of though...
His point was that a client shoudn't be able to hose an nfs server just by
making a v2 mount.
What would happen to a v2 client if we sent it a v3 error?
Take care,
Bill