Skip to Content.
Sympa Menu

sm-discuss - Re: [SM-Discuss] Initrd with udev and LVM; may be good for ISO

sm-discuss AT lists.ibiblio.org

Subject: Public SourceMage Discussion List

List archive

Chronological Thread  
  • From: David Kowis <dkowis AT shlrm.org>
  • To: SM-Discuss <sm-discuss AT lists.ibiblio.org>
  • Subject: Re: [SM-Discuss] Initrd with udev and LVM; may be good for ISO
  • Date: Mon, 29 Aug 2005 12:07:48 -0500

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Please feel free to file a feature request on bugzilla for us. Right now
we're more concerned with basic things like getting an install bug
free. This is an interesting thing, however. I think software RAID can be
done without the initrd having LVM support in it. It has to do
with partitions being marked a certain way.
using strategies listed here:
http://wiki.clug.org.za/clugwiki/index.php/RAID-1_in_a_hurry_with_grub_and_mdadm

notably "Make the hda partition types "RAID auto-start"..."
Anyways, we'll be getting to these "fancier" features eventually :)

Sergey A. Lipnevich wrote:
> Hi All,
>
> I've recently committed changes that make a static udevstart binary, as
> well as
> changes for device-mapper and lvm to make static binaries. I also committed
> a
> new scenario for a very good linux-initrd spell by David. This scenario uses
> the things above to create a working initrd image without devfs, that also
> allows a root partition to be a logical volume. This work was also based on
> Red
> Hat's LVM howto document and LVM2 docs.
> While my "itch" has been scratched, and my new server has / as l.v., I'd
> like to
> suggest that a new ISO includes these three static binaries mentioned
> above, and
> uses udev instead of obsolete devfs.
> Also, presently, LVM stuff is not on the ISO's initrd so it can't be used to
> troubleshoot a system with / as l.v. However, the fact that the latest ISO
> 0.9.4 beta itself *has* LVM on it, allows one to set up such a system (and
> this
> is VERY GOOD, thanks a lot guys!). It wouldn't start without an initrd
> though
> :-), hence my commits.
>
> One note: I had to use busybox's shell in this new initrd "scenario"
> because on
> amd64 with gcc 3.4 and 4.0, bash.static sefgaults in an easily reproducible
> way. The one built with 3.3 seems to be fine.
>
> Cheers,
> Sergey.
> _______________________________________________
> SM-Discuss mailing list
> SM-Discuss AT lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/sm-discuss

- --
David Kowis

ISO Team Lead - www.sourcemage.org
SourceMage GNU/Linux

Progress isn't made by early risers. It's made by lazy men trying to find
easier ways to do something.
- Robert Heinlein

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (MingW32)

iD8DBQFDE0DktgErhgxHMHsRAqOoAKCD4xCgg1yCK0NqbeyGqIOI5o1hAQCdFk2t
a5bOa42TPwBeRsgh3weoAkc=
=6aAP
-----END PGP SIGNATURE-----




Archive powered by MHonArc 2.6.24.

Top of Page