mdadm and raidtools

[Bill Rugolsky Jr.]

On Wed, Jan 14, 2004 at 05:22:22PM -0500, Jeremy Katz wrote:
> On Wed, 2004-01-14 at 10:22 -0500, Bill Rugolsky Jr. wrote:
> > With mdadm support in initscripts (and soon initrd, I hope), it would
> > not be terribly difficult to do this in Fedora. 
> 
> initrds currently just call the RAID autostart ioctl from within nash.
> Switching to mdadm here is going to make things a bit larger :/
 
I know.  That's why doing a live upgrade is a bit of a hassle. :-/

At this point, what fraction of users realistically are going to have a
kernel+early userspace (initrd or initramfs) config that fits on a floppy?
What fraction of users can't boot from a rescue CD?  Many systems don't
even have a floppy drive anymore.

nash made a lot of sense when the common kernel config could be
shoe-horned onto a floppy disk with a small libc.  At this point,
I'm inclined to use something like busybox + the full tools, compiled
against a lightweight libc: klibc, uClibc, or dietlibc.  That would
include things like mdadm, lvm2/evms, udev, etc.

Since the performance of DM and (LVM2/EVMS2) is now acceptable, I would
hope to see it very well supported.  Certainly RHEL4 needs to get this
right, and so it should be widely tested in Fedora.

Laptop users have been clamoring for good encryption support for years.
How nice it would be to just drop dm-crypt support into early userspace.

I'd like a mechanism to modify early userspace by:

   1. Building binaries against the correct libraries.

   2. Dropping a script into something like

	/etc/sysconfig/{initrd,initramfs}.d

   3. Somehow specifying the dependence on binaries, e.g.,
      by also putting them into a special directory, or naming
      them in comments in the script, etc.

This ought to be designed with HAL, an initscripts rework, etc.,
in mind.  I expect to do a rough cut of this before migrating
FC1 machines to FC2.

Regards,

	Bill Rugolsky