Return-Path: linux-nfs-owner@vger.kernel.org Received: from mx2.netapp.com ([216.240.18.37]:53763 "EHLO mx2.netapp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754169Ab2FDNtW convert rfc822-to-8bit (ORCPT ); Mon, 4 Jun 2012 09:49:22 -0400 From: "Adamson, Andy" To: Boaz Harrosh CC: "Myklebust, Trond" , "" Subject: Re: [PATCH 3/3] NFSv4.1 fence all layouts with file layout data server connection errors Date: Mon, 4 Jun 2012 13:49:05 +0000 Message-ID: <2D678724-1ED3-433F-AE4D-BCBC93D01783@netapp.com> References: <1338571178-2096-1-git-send-email-andros@netapp.com> <1338571178-2096-3-git-send-email-andros@netapp.com> <4FCA94A6.3070105@panasas.com> In-Reply-To: <4FCA94A6.3070105@panasas.com> Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Sender: linux-nfs-owner@vger.kernel.org List-ID: On Jun 2, 2012, at 6:33 PM, Boaz Harrosh wrote: > On 06/01/2012 08:19 PM, andros@netapp.com wrote: > >> From: Andy Adamson >> > > > The tittle does not match at all the message text nor the patch. > > More like: Don't send LAYOUTCOMMIT if data is rewritten through MDS Well, the meat of this patch is the LAYOUTRETURN for fencing. -->Andy > > I think. > Boaz > >> Clear the inode LAYOUTCOMMIT flag as we will resend writes through the MDS. >> >> Signed-off-by: Andy Adamson >> --- >> fs/nfs/nfs4filelayout.c | 4 ++-- >> 1 files changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c >> index e134029..96a75d1 100644 >> --- a/fs/nfs/nfs4filelayout.c >> +++ b/fs/nfs/nfs4filelayout.c >> @@ -205,8 +205,8 @@ static int filelayout_async_handle_error(struct rpc_task *task, >> case -EPIPE: >> dprintk("%s DS connection error %d\n", __func__, >> task->tk_status); >> - if (!filelayout_test_devid_invalid(devid)) >> - _pnfs_return_layout(inode); >> + clear_bit(NFS_INO_LAYOUTCOMMIT, &NFS_I(inode)->flags); >> + _pnfs_return_layout(inode); >> filelayout_mark_devid_invalid(devid); >> rpc_wake_up(&tbl->slot_tbl_waitq); >> nfs4_ds_disconnect(clp); > >