linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Trond Myklebust <trond.myklebust@fys.uio.no>
To: "Halevy, Benny" <bhalevy@panasas.com>
Cc: Jeff Layton <jlayton@poochiereds.net>,
	Mikulas Patocka <mikulas@artax.karlin.mff.cuni.cz>,
	Arjan van de Ven <arjan@infradead.org>,
	Jan Harkes <jaharkes@cs.cmu.edu>,
	Miklos Szeredi <miklos@szeredi.hu>,
	linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	nfsv4@ietf.org
Subject: RE: Finding hardlinks
Date: Wed, 03 Jan 2007 00:46:44 +0100	[thread overview]
Message-ID: <1167781604.6090.120.camel@lade.trondhjem.org> (raw)
In-Reply-To: <E472128B1EB43941B4E7FB268020C89B149CF0@riverside.int.panasas.com>

On Sun, 2006-12-31 at 16:19 -0500, Halevy, Benny wrote:

> Even for NFSv3 (that doesn't have the unique_handles attribute I think
> that the linux nfs client can do a better job.  If you'd have a filehandle
> cache that points at inodes you could maintain a many to one relationship
> from multiple filehandles into one inode.  When you discover a new filehandle
> you can look up the inode cache for the same fileid and if one is found you
> can do a getattr on the old filehandle (without loss of generality you should 
> always use the latest filehandle that was returned for that filesystem object,
> although any filehandle that refers to it can be used).
> If the getattr succeeded then the filehandles refer to the same fs object and
> you can create a new entry in the filehandle cache pointing at that inode.
> Otherwise, if getattr says that the old filehandle is stale I think you should
> mark the inode as stale and keep it around so that applications can get an
> appropriate error until last close, before you clean up the fh cache from the
> stale filehandles. A new inode structure should be created for the new filehandle.

There are, BTW, other reasons why the above is a bad idea: it breaks on
a bunch of well known servers. Look back at the 2.2.x kernels and the
kind of hacks they had in order to deal with crap like the Netapp
'.snapshot' directories which contain files with duplicate fileids that
do not represent hard links, but rather represent previous revisions of
the same file.

That kind of hackery was part of the reason why I ripped out that code.
The other reasons were
        - that you end up playing unnecessary getattr games like the
        above for little gain.
        - the only servers that implemented the above were borken pieces
        of crap that encoded parent directories in the filehandle, and
        which end up breaking anyway under cross-directory renames.
        - the world is filled with non-posix filesystems that frequently
        don't have real fileids. They are often just generated on the
        fly and can change at the drop of a hat.

Trond


  parent reply	other threads:[~2007-01-02 23:47 UTC|newest]

Thread overview: 100+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-12-20  9:03 Finding hardlinks Mikulas Patocka
2006-12-20 11:44 ` Miklos Szeredi
2006-12-20 16:36   ` Mikulas Patocka
2006-12-20 16:50     ` Miklos Szeredi
2006-12-20 19:54       ` Al Viro
2006-12-20 20:12         ` Mikulas Patocka
2006-12-31 15:02         ` Mikulas Patocka
2006-12-21 18:58   ` Jan Harkes
2006-12-21 23:49     ` Mikulas Patocka
2006-12-22  5:05       ` Jan Harkes
2006-12-23 10:18       ` Arjan van de Ven
2006-12-23 14:00         ` Mikulas Patocka
2006-12-28  9:06           ` Benny Halevy
2006-12-28 10:05             ` Arjan van de Ven
2006-12-28 15:24               ` Benny Halevy
2006-12-28 19:58                 ` Miklos Szeredi
2007-01-02 19:15                   ` Pavel Machek
2007-01-02 20:41                     ` Miklos Szeredi
2007-01-02 20:50                       ` Mikulas Patocka
2007-01-02 21:10                         ` Miklos Szeredi
2007-01-02 21:37                           ` Mikulas Patocka
2007-01-03 11:56                       ` Pavel Machek
2007-01-03 12:33                         ` Miklos Szeredi
2007-01-03 12:42                           ` Pavel Machek
2007-01-11 23:43                             ` Denis Vlasenko
2007-01-03 12:45                           ` Martin Mares
2007-01-03 13:54                           ` Matthew Wilcox
2007-01-03 15:51                             ` Miklos Szeredi
2007-01-03 19:04                               ` Mikulas Patocka
2007-01-04 22:59                               ` Pavel Machek
2007-01-05  8:43                                 ` Miklos Szeredi
2007-01-05 13:12                                   ` Pavel Machek
2007-01-05 13:55                                     ` Miklos Szeredi
2007-01-05 14:08                                       ` Mikulas Patocka
2007-01-05 15:09                                         ` Miklos Szeredi
2007-01-05 15:15                                           ` Miklos Szeredi
2007-01-08 11:27                                             ` Pavel Machek
2007-01-08  5:57                                           ` Mikulas Patocka
2007-01-08  8:49                                             ` Miklos Szeredi
2007-01-08 11:29                                               ` Pavel Machek
2007-01-08 12:00                                                 ` Miklos Szeredi
2007-01-08 13:26                                                   ` Martin Mares
2007-01-08 13:39                                                     ` Miklos Szeredi
2007-01-09 16:26                                                   ` Steven Rostedt
2007-01-09 19:53                                                     ` Frank van Maarseveen
2007-01-09 20:11                                                       ` Steven Rostedt
2007-01-11 10:07                                                       ` Pádraig Brady
2007-01-05 17:30                                   ` Frank van Maarseveen
2006-12-28 18:14               ` Mikulas Patocka
2006-12-29 10:34                 ` Trond Myklebust
2006-12-30  1:04                   ` Mikulas Patocka
2007-01-01  2:30                     ` Nikita Danilov
2007-01-01 22:58                       ` Mikulas Patocka
2007-01-01 23:05                         ` Nikita Danilov
2007-01-01 23:22                           ` Mikulas Patocka
2007-01-04 13:59                             ` Nikita Danilov
2007-01-02 23:14                     ` Trond Myklebust
2007-01-02 23:50                       ` Mikulas Patocka
2006-12-28 13:22             ` Jeff Layton
2006-12-28 15:12               ` Benny Halevy
2006-12-28 15:54                 ` Jeff Layton
2006-12-28 16:26                   ` Jan Engelhardt
2006-12-28 18:17                 ` Mikulas Patocka
2006-12-28 20:07                   ` Halevy, Benny
2006-12-29 10:28                     ` [nfsv4] " Trond Myklebust
2006-12-31 21:25                       ` Halevy, Benny
2007-01-02 23:21                         ` Trond Myklebust
2007-01-03 12:35                           ` Benny Halevy
2007-01-04  0:43                             ` Trond Myklebust
2007-01-04  8:36                             ` Trond Myklebust
2007-01-04 10:04                               ` Benny Halevy
2007-01-04 10:47                                 ` Trond Myklebust
2007-01-05  8:28                                   ` Benny Halevy
2007-01-05 10:29                                     ` Trond Myklebust
2007-01-05 16:40                                 ` Nicolas Williams
2007-01-05 16:56                                   ` Trond Myklebust
2007-01-06  7:44                                   ` Halevy, Benny
2007-01-10 13:04                                   ` Benny Halevy
2006-12-29 10:12                 ` Trond Myklebust
2006-12-31 21:19                   ` Halevy, Benny
2007-01-02 23:20                     ` Trond Myklebust
2007-01-02 23:46                     ` Trond Myklebust [this message]
2007-01-11 23:35             ` Denis Vlasenko
2006-12-29 10:02           ` Pavel Machek
2007-01-01 22:47             ` Mikulas Patocka
2007-01-01 23:53               ` Jan Harkes
2007-01-02  0:04                 ` Mikulas Patocka
2007-01-03 18:58                   ` Frank van Maarseveen
2007-01-03 19:17                     ` Mikulas Patocka
2007-01-03 19:26                       ` Frank van Maarseveen
2007-01-03 19:31                         ` Mikulas Patocka
2007-01-03 20:26                           ` Frank van Maarseveen
2007-01-12  0:00                             ` Denis Vlasenko
2007-01-03 22:30                           ` Pavel Machek
2007-01-03 21:09                     ` Bryan Henderson
2007-01-03 22:01                       ` Frank van Maarseveen
2007-01-03 23:43                         ` Mikulas Patocka
2007-01-04  0:12                           ` Frank van Maarseveen
2007-01-08  6:19                             ` Mikulas Patocka
     [not found] <7x5mR-2wX-3@gated-at.bofh.it>
     [not found] ` <7x9Ad-18O-35@gated-at.bofh.it>
     [not found]   ` <7yXEy-UI-39@gated-at.bofh.it>
     [not found]     ` <7yYKa-2Ds-3@gated-at.bofh.it>
     [not found]       ` <7zcWP-7ET-5@gated-at.bofh.it>
     [not found]         ` <7zdzA-jc-27@gated-at.bofh.it>
     [not found]           ` <7zeP5-2ic-15@gated-at.bofh.it>
     [not found]             ` <7zgH9-5my-17@gated-at.bofh.it>
     [not found]               ` <7zJSM-14t-9@gated-at.bofh.it>
     [not found]                 ` <7zSW5-6cj-9@gated-at.bofh.it>
     [not found]                   ` <7zX9l-4rS-7@gated-at.bofh.it>
     [not found]                     ` <7zXMb-5g5-27@gated-at.bofh.it>
2007-01-05 23:54                       ` Bodo Eggert

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=1167781604.6090.120.camel@lade.trondhjem.org \
    --to=trond.myklebust@fys.uio.no \
    --cc=arjan@infradead.org \
    --cc=bhalevy@panasas.com \
    --cc=jaharkes@cs.cmu.edu \
    --cc=jlayton@poochiereds.net \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=mikulas@artax.karlin.mff.cuni.cz \
    --cc=nfsv4@ietf.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).