All of lore.kernel.org
 help / color / mirror / Atom feed
From: Francis SOUYRI <francis.souyri@wanadoo.fr>
To: linux-lvm@sistina.com
Subject: Re: [linux-lvm] Some suggestions about lvminitrd_create
Date: Fri May 17 00:04:02 2002	[thread overview]
Message-ID: <3CE48E38.40605@wanadoo.fr> (raw)
In-Reply-To: 20020516152200.GP1248@tykepenguin.com

Patrick Caulfield wrote:

>On Thu, May 16, 2002 at 05:10:58PM +0200, Bas wrote:
>  
>
>>>>Just a couple of suggestions firstly it would be nice if it checked
>>>>for /boot/System.map-'uname -r' rather than for just /boot/System.map.
>>>>When you have more than one kernel floating about this is handy.
>>>>        
>>>>
>>>Yep, that's what the "-F" option aims to support.
>>>
>>>      
>>>
>>>>Second suggestion it would be nice if a -t option existed to allow a
>>>>choice in the type of filesystem the initrd is created in. Not everyone
>>>>is using ext2 these days :)
>>>>        
>>>>
>>>Well, why do you want to use a different filesystem on a relatively tiny
>>>and little populated ram disk?
>>>      
>>>
>>Because once you choose to use XFS, you don't need ext2 utils anymore, so
>>why would you want to install ext2utils ?
>>    
>>
>
>You really want to use XFS for an initrd??
>
><fx tests:>
>
># dd if=/dev/zero of=disk bs=1024 count=8192
># mkfs.xfs disk
># mount -oloop disk /mnt
># df /mnt
>Filesystem           1k-blocks      Used Available Use% Mounted on
>/root/disk                3392        32      3360   1% /mnt
>
>OK you could tune that up a bit but I think ext2 is more approriate for an
>initrd and the ext2 utils don't take up /that/ much space: probably much less
>than your XFS journals in fact :-)
>
>patrick
>
>
>_______________________________________________
>linux-lvm mailing list
>linux-lvm@sistina.com
>http://lists.sistina.com/mailman/listinfo/linux-lvm
>read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html
>
>  
>
Hello,

    I use "romfs" for the 'initrd" it take less place in the kernel than 
"ext2" (I use "reiserfs" for "normal" filesystems), the only problem is 
that "romfs" is a read only filesytem but you can use the "tmpfs" like 
this in your "linuxrc" script:

#!/bin/sh
mount -t tmpfs none /etc
mount -t proc none /proc
echo "Scanning for Volume Group..."
/sbin/vgscan
echo "Activating 'root' Volume Group"
/sbin/vgchange -a y /dev/vgroot
umount /proc
umount /etc

I use the "nash" shell from redhat, it take few place and have builtin 
the mount/umount commands.

The LVM is compiled in the kernel and I have these files in my "initrd":

/bin/nash
/bin/sh -> nash
/lib/ld-linux.so.2
/lib/libc.so.6
/lib/liblvm-10.so.1
/sbin/vgchange
/sbin/vgscan

Best regards.

Francis

  reply	other threads:[~2002-05-17  0:04 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-05-15  4:03 [linux-lvm] Some suggestions about lvminitrd_create Jonathan Buzzard
2002-05-16  9:27 ` Heinz J . Mauelshagen
2002-05-16 10:09   ` Bas
2002-05-16 10:22     ` Patrick Caulfield
2002-05-17  0:04       ` Francis SOUYRI [this message]
2002-05-17  2:19         ` Patrick Caulfield

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=3CE48E38.40605@wanadoo.fr \
    --to=francis.souyri@wanadoo.fr \
    --cc=linux-lvm@sistina.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.