Re: [PATCH] NFSD: fix decode_cb_sequence4resok

["J. Bruce Fields" <bfields@redhat.com>]

On Tue, Feb 22, 2011 at 04:15:22PM -0800, Benny Halevy wrote:
> On 2011-02-22 16:11, J. Bruce Fields wrote:
> > On Tue, Feb 22, 2011 at 02:43:22PM -0800, Benny Halevy wrote:
> >> Fix bug introduced in patch
> >> 85a56480 NFSD: Update XDR decoders in NFSv4 callback client
> >>
> >> Although decode_cb_sequence4resok ignores highest slotid and target highest slotid
> >> it must account for their space in their xdr stream when calling xdr_inline_decode
> > 
> > Thanks, applying for 2.6.38.  (How come you caught this, and I didn't?
> > I guess it's just that the object code depends more on the callback
> > returns?)
> 
> That's a great question.
> I caught it by testing cb_layoutrecall, both with the objects layout
> and with the files layout with the patch I sent for PNFSD_LEXP to recall the
> layout on truncate.
> cb_recall decoding should be broken too at the server.
> It's possible nothing really depends on its status.

Yeah, the server may take a failure as a sign it should retry the
request, but the only "reply" that really matters is the eventual
{deleg,layout,whatever}return.

On the client side a failure here would be visisble at most as a retry,
so pynfs won't catch this kind of thing.

--b.