From: Nathan Scott <nathans@sgi.com>
To: Norberto Bensa <norberto+linux-kernel@bensa.ath.cx>,
L A Walsh <lkml@tlinx.org>
Cc: linux-kernel@vger.kernel.org, linux-xfs@oss.sgi.com
Subject: Re: XFS: how to NOT null files on fsck?
Date: Thu, 29 Jul 2004 11:30:49 +1000 [thread overview]
Message-ID: <20040729013049.GE800@frodo> (raw)
In-Reply-To: <40EEC9DC.8080501@tlinx.org>
On Fri, Jul 09, 2004 at 09:37:48AM -0700, L A Walsh wrote:
> It's a feature! :-)
>
> It's been in the code for years to randomly write nulls to some files
Pfft, nonsense. The problem relates to an updated inode size
being flushed ahead of the data behind it (hence a size update
can make it out before delayed allocate extents do, and we end
up with a hole beyond the end of file, which reads as zeroes).
> Apparently not easily reproduced, no one has a clue why it does it.
> Just does.
No, its actually well known why it behaves this way.
We are looking into ways to address this, and have some
ideas - the trick is fixing it without hurting write
performance - which we will do, its just not trivial.
There are several techiques to reduce the impact of this
behaviour, as others have described (or see the linux-xfs
archives).
cheers.
--
Nathan
next prev parent reply other threads:[~2004-07-29 0:40 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-07-05 5:47 XFS: how to NOT null files on fsck? Norberto Bensa
2004-07-09 16:37 ` L A Walsh
2004-07-09 21:59 ` Chris Wedgwood
2004-07-10 18:33 ` L A Walsh
2004-07-10 18:43 ` Chris Wedgwood
2004-07-10 21:24 ` Bernd Eckenfels
2004-07-11 21:54 ` Helge Hafting
2004-07-12 17:56 ` H. Peter Anvin
2004-07-12 19:59 ` Chris Wedgwood
2004-07-12 20:32 ` H. Peter Anvin
2004-07-12 22:29 ` Bernd Eckenfels
2004-07-12 23:03 ` Bernd Eckenfels
2004-07-12 23:14 ` Chris Wedgwood
2004-07-10 18:43 ` Jan Knutar
2004-07-10 18:46 ` Chris Wedgwood
2004-07-10 18:55 ` Norberto Bensa
2004-07-10 19:19 ` Chris Wedgwood
2004-07-12 21:20 ` Chris Wedgwood
2004-07-12 22:40 ` L A Walsh
2004-07-12 22:53 ` Chris Wedgwood
2004-07-13 1:44 ` Bernd Eckenfels
2004-07-13 5:24 ` Chris Wedgwood
[not found] ` <2hgxc-5x9-9@gated-at.bofh.it>
2004-07-13 7:25 ` Anton Ertl
2004-07-13 8:09 ` Chris Wedgwood
2004-07-13 9:34 ` Anton Ertl
2004-07-13 9:53 ` Chris Wedgwood
2004-07-13 10:27 ` Tim Connors
2004-07-13 10:38 ` ismail dönmez
2004-07-13 11:16 ` Nick Piggin
2004-07-13 12:52 ` ismail dönmez
2004-07-13 10:58 ` Chris Wedgwood
2004-07-13 13:33 ` Anton Ertl
2004-07-13 20:32 ` Chris Wedgwood
2004-07-13 22:42 ` Bernd Eckenfels
2004-07-14 18:49 ` Anton Ertl
2004-07-14 19:00 ` Chris Wedgwood
2004-07-13 22:24 ` Helge Hafting
2004-07-13 22:39 ` Chris Wedgwood
2004-07-13 23:23 ` Bernd Eckenfels
2004-07-14 18:53 ` Anton Ertl
2004-07-10 19:33 ` Andreas Schwab
2004-07-10 19:40 ` Chris Wedgwood
2004-07-10 19:46 ` Norberto Bensa
2004-07-10 20:03 ` Chris Wedgwood
2004-07-11 1:21 ` Gopikrishnan Sidhardhan
2004-07-29 1:30 ` Nathan Scott [this message]
2004-08-03 18:31 ` L A Walsh
2004-08-04 0:48 ` Andi Kleen
2004-08-04 6:37 ` L A Walsh
2004-08-05 8:16 ` Helge Hafting
2004-08-06 1:10 ` Nathan Scott
2004-08-06 1:34 ` Andrew Morton
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=20040729013049.GE800@frodo \
--to=nathans@sgi.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-xfs@oss.sgi.com \
--cc=lkml@tlinx.org \
--cc=norberto+linux-kernel@bensa.ath.cx \
/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