linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: Christoph Hellwig <hch@infradead.org>,
	Fred Isaman <iisaman@netapp.com>,
	linux-nfs@vger.kernel.org
Subject: Re: [PATCH 11/12] NFSv4.1: layoutcommit
Date: Wed, 23 Mar 2011 17:21:57 -0400	[thread overview]
Message-ID: <20110323212157.GA14281@infradead.org> (raw)
In-Reply-To: <1300914915.11677.88.camel@lade.trondhjem.org>

On Wed, Mar 23, 2011 at 05:15:15PM -0400, Trond Myklebust wrote:
> We had a spec clarification about this recently. The result is that for
> 'files' layout types:
> 
> 1) DATA_SYNC writes and unstable writes with COMMIT must store enough
> data to disk to allow the server to recover the data if it crashes (i.e.
> it provides the equivalent of O_DSYNC and/or fdatasync()). FILE_SYNC
> writes must commit all data + metadata (i.e. they do the equivalent of
> O_SYNC writes).
> 2) This means that layoutcommit is only needed when the layout specifies
> COMMIT to the data servers, or if the writes to the data servers return
> a DATA_SYNC. All it does is commit the time stamps + file size to disk
> on the metadata server (and so you avoid the need to do an expensive
> recovery process on crash).
> 
> IOW: the above should be correct.

Okay, care to add a comment explaining this fact?  The name layoutcommit
doesn't quite suggest that it doesn't commit anything but timestamps
and size.

Can you recover the size from the data servers with the code above?  The
size should also be on disk for an fdatasync.


  reply	other threads:[~2011-03-23 21:21 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-23 13:27 [PATCH 00/12] NFSv4.1 pnfs wave5 submission (try 2) Fred Isaman
2011-03-23 13:27 ` [PATCH 01/12] NFSv4.1: don't send COMMIT to ds for data sync writes Fred Isaman
2011-03-23 13:27 ` [PATCH 02/12] NFSv4.1: rearrange nfs_commit_rpcsetup Fred Isaman
2011-03-23 13:27 ` [PATCH 03/12] NFSv4.1: add callback to nfs4_commit_done Fred Isaman
2011-03-23 13:27 ` [PATCH 04/12] NFSv4.1: pull error handling out of nfs_commit_list Fred Isaman
2011-03-23 13:27 ` [PATCH 05/12] NFSv4.1: pull out code from nfs_commit_release Fred Isaman
2011-03-23 13:27 ` [PATCH 06/12] NFSv4.1: shift filelayout_free_lseg Fred Isaman
2011-03-23 13:27 ` [PATCH 07/12] NFSv4.1: alloc and free commit_buckets Fred Isaman
2011-03-23 13:27 ` [PATCH 08/12] NFSv4.1: add generic layer hooks for pnfs COMMIT Fred Isaman
2011-03-23 19:42   ` Benny Halevy
2011-03-23 19:45     ` Benny Halevy
2011-03-23 20:30     ` Fred Isaman
2011-03-23 20:39       ` Benny Halevy
2011-03-23 20:52         ` Trond Myklebust
2011-03-23 13:27 ` [PATCH 09/12] NFSv4.1: remove GETATTR from ds commits Fred Isaman
2011-03-23 19:51   ` Benny Halevy
2011-03-23 20:32     ` Fred Isaman
2011-03-23 20:35       ` Trond Myklebust
2011-03-23 13:27 ` [PATCH 10/12] NFSv4.1: filelayout driver specific code for COMMIT Fred Isaman
2011-03-23 13:27 ` [PATCH 11/12] NFSv4.1: layoutcommit Fred Isaman
2011-03-23 20:33   ` Benny Halevy
2011-03-23 21:00   ` Christoph Hellwig
2011-03-23 21:15     ` Trond Myklebust
2011-03-23 21:21       ` Christoph Hellwig [this message]
2011-03-23 21:26         ` Trond Myklebust
2011-03-23 13:27 ` [PATCH 12/12] NFSv4.1 remove temp code that prevented ds commits Fred Isaman
  -- strict thread matches above, loose matches on Subject: below --
2011-03-24 13:57 [PATCH 11/12] NFSv4.1: layoutcommit William A. (Andy) Adamson
2011-03-24 16:37 ` Benny Halevy
2011-03-24 16:48   ` Trond Myklebust
2011-03-24 16:54     ` Fred Isaman
2011-03-24 16:58     ` Benny Halevy
2011-03-24 17:15       ` Trond Myklebust
2011-03-25  9:39         ` Benny Halevy
2011-03-24 14:45 William A. (Andy) Adamson
2011-03-24 17:06 ` Benny Halevy

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110323212157.GA14281@infradead.org \
    --to=hch@infradead.org \
    --cc=Trond.Myklebust@netapp.com \
    --cc=iisaman@netapp.com \
    --cc=linux-nfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).