xfs
[Top] [All Lists]

Re: md + xfs (fwd)

To: Mihai RUSU <dizzy@xxxxxxxxx>
Subject: Re: md + xfs (fwd)
From: Simon Matter <simon.matter@xxxxxxxxxxxxxxxx>
Date: Thu, 11 Jul 2002 14:02:15 +0200
>received: from mobile.sauter-bc.com (unknown [10.1.6.21]) by basel1.sauter-bc.com (Postfix) with ESMTP id 9047B57306; Thu, 11 Jul 2002 14:02:16 +0200 (CEST)
Cc: Seth Mos <knuffie@xxxxxxxxx>, Linux XFS List <linux-xfs@xxxxxxxxxxx>
Organization: Sauter AG, Basel
References: <Pine.LNX.4.33.0207111441130.17973-100000@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
Mihai RUSU schrieb:
> 
> On Thu, 11 Jul 2002, Seth Mos wrote:
> 
> >
> > I think that the raid device has a older state in which the filesystem was
> > clean. But this one is rather awkward. I don't think it is normally 
> > possible.
> >

When booting a RedHat system with softraid, the linuxrc script looks
something like this:

#!/bin/nash
 
echo "Loading raid1 module"
insmod /lib/raid1.o
mount -t proc /proc /proc
echo Mounting /proc filesystem
echo Creating root device
mkrootdev /dev/root
raidautorun /dev/md0
echo 0x0100 > /proc/sys/kernel/real-root-dev
umount /proc
echo Mounting root filesystem
mount --ro -t xfs /dev/root /sysroot
pivot_root /sysroot /sysroot/initrd

The mount command is a builtin command of /bin/nash, maybe this explains
the different behaviour. I hope one of the gurus can explain it to us.

Simon

> 
> That is why I asked here in the first place. I was not sure about how
> RAID1 on linux works but it seems to me that when a crash occured it
> choses one of the RAID1 images and rebuilds from that, but I am afraid
> that maybe in that state only the databytes that tell XFS it is clean,
> where set right, and other fs structures are corupt.
> 
> I should xfscheck that partition but I would really prefer to do that when
> all other fails (beeing a production server).
> 
> ----------------------------
> Mihai RUSU
> 
> Disclaimer: Any views or opinions presented within this e-mail are solely
> those of the author and do not necessarily represent those of any company,
> unless otherwise specifically stated.



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