xfs
[Top] [All Lists]

Re: different behaviour between XFS and ext3

To: Blair Barnett <bbarnett@xxxxxxxxxx>
Subject: Re: different behaviour between XFS and ext3
From: Steve Lord <lord@xxxxxxx>
Date: 04 Aug 2003 17:09:35 -0500
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <3F2ED423.9E7F227D@xxxxxxxxxx>
Organization:
References: <3F283E36.778CC934@xxxxxxxxxx> <1059662873.2837.56.camel@xxxxxxxxxxxxxxxxxxxx> <3F2ED423.9E7F227D@xxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
On Mon, 2003-08-04 at 16:46, Blair Barnett wrote:
> Steve Lord wrote:
> 
> >
> >
> > Its a bug, not sure what it is yet, those subsequent syncs should
> > be working. In testing sync and reboot I was untaring a linux kernel
> > doing sync followed by immediate reset of the box. I can build the
> > kernel afterwards, so that works. When I run the test, there is more
> > data on the disk than number 10, but the inode size is back at where
> > it was at the first sync.
> >
> > Steve
> 
> Need any help chasing this down? Let me know.

Thanks, I have some code which fixes this, causes some other strange
behavior though (including corruption in other ways), so it needs work.
Basically the file data is on disk, and before remounting the
filesystem, the inode size is correct on disk, its just that recovery
is rolling the inode size back again.

Steve


-- 

Steve Lord                                      voice: +1-651-683-3511
Principal Engineer, Filesystem Software         email: lord@xxxxxxx


<Prev in Thread] Current Thread [Next in Thread>