[pnfs] [PATCH 01/10] pnfsd: initialize layoutget return_on_close

[Marc Eshel]

Yes, I did. I found the problem on the server side, we never initialized 
pattern offset to zero so get layout failed, after setting it to zero it 
now works.
Thanks, Marc.

pnfs-bounces at linux-nfs.org wrote on 05/08/2008 01:57:08 PM:

> Marc, did you merge the following patch?
> http://linux-nfs.org/pipermail/pnfs/2008-May/003544.html
> 
> Benny
> 
> On May. 08, 2008, 13:32 -0700, marc eshel <eshel at sbcglobal.net> wrote:
> > Hi Andy,
> > A quick test of these set of 10 patches shows that all the traffic 
goes 
> > to the MDS. A session is created with each DS but it is the last thing 

> > that the DS gets from the client, all read/write activities goes to 
the 
> > MDS. I will try to look at it later.
> > Marc.
> > 
> > 
> > andros at netapp.com wrote:
> >> From: Andy Adamson <andros at umich.edu>
> >>
> >> Signed-off-by: Andy Adamson<andros at netapp.com>
> >> ---
> >>  fs/nfsd/nfs4xdr.c |    4 +++-
> >>  1 files changed, 3 insertions(+), 1 deletions(-)
> >>
> >> diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
> >> index 40c25b8..b6b8ac8 100644
> >> --- a/fs/nfsd/nfs4xdr.c
> >> +++ b/fs/nfsd/nfs4xdr.c
> >> @@ -3469,7 +3469,9 @@ nfsd4_encode_layoutget(struct 
> nfsd4_compoundres *resp,
> >>     int maxcount, leadcount;
> >>     struct super_block *sb;
> >>     unsigned int *p_start = resp->p;
> >> -   struct pnfs_layoutget_arg args;
> >> +   struct pnfs_layoutget_arg args = {
> >> +      .return_on_close = 0,
> >> +   };
> >>     ENCODE_HEAD;
> >> 
> >>     dprintk("%s: err %d\n", __func__, nfserr);
> >> 
> > 
> > _______________________________________________
> > pNFS mailing list
> > pNFS at linux-nfs.org
> > http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs
> 
> _______________________________________________
> pNFS mailing list
> pNFS at linux-nfs.org
> http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs