http://bugzilla.novell.com/show_bug.cgi?id=505670
User anniyka@googlemail.com added comment
http://bugzilla.novell.com/show_bug.cgi?id=505670#c41
--- Comment #41 from Anniyka Wandersmann 2009-09-29 07:01:31 MDT ---
Just when you get a workaround and you think you will came along ...
I had found a workaround for the not building of the initrd. just do "mkinitrd
-A" than it will not go through all scripts to figure out what modules to use
but just takes every module it finds.
A little bad but for a workaround in a testsystem ok.
Than I bootet my system and guess ... "waiting for /dev/vg00/root to appear
.."
I started the console and noticed to my liking that now mdadm and vgchange wher
again in the initrd. So I tried to mount /dev/sda2 since it is one off my /boot
devices in an RAID1.
It won't and I found out that it is not sda but hda (it's all SATA except this
one is PATA)
Ok nonetheless I used mdadm with -A --scan and got my RAIDs build. Why this
time again the md127 instead of md1 appears is beyond me.
I used "vgchange -a y" and all LVs appear, including /dev/vg00/root.
As I had all the older scripts from mkinitrd (the ones weher lvm and dm is
different) I copied the lvm boot script to /boot, umounted /dev/hda2 and tried
to boot by usning "init".
Then the devices /dev/vg00 went away and I ended where I started: "waiting for
/dev/vg00/root to appear ..."
Any further notions, at least, which bug I should report this too now?
Seems to me that the whole initrd is broken as long it hast to deal with MD and
LVM in some way. Perhaps even something in udev or kernel since we are back to
/dev/hda now.
--
Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.