Bug 490008 - mkinitrd if using md raid
Summary: mkinitrd if using md raid
Status: RESOLVED DUPLICATE of bug 445490
Alias: None
Product: openSUSE 11.1
Classification: openSUSE
Component: Basesystem (show other bugs)
Version: Final
Hardware: Other openSUSE 11.1
: P5 - None : Normal (vote)
Target Milestone: ---
Assignee: Michal Marek
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-03-28 11:53 UTC by Lars Müller
Modified: 2009-11-09 09:48 UTC (History)
1 user (show)

See Also:
Found By: Customer
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
Patch as suggested by Manfred on the openSUSE ML (759 bytes, text/plain)
2009-03-28 11:57 UTC, Lars Müller
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Lars Müller 2009-03-28 11:53:49 UTC
http://lists.opensuse.org/opensuse/2009-03/msg02086.html

i use this layout:
/boot is on /dev/md0 mirror at the beginning of the 2 disks.
lvm volumegroup is on /dev/md1 mirror
/ is a lvm volume
swap and other filesystems are also lvm volumes.

mkinitrd creates a /etc/mdadm.conf inside the initrd that contains only
the /dev/md1. The system stops when it it trying to check the /boot
filesystem - because it is not found. When i add the /dev/md0 into the
initrd mdadm.conf then booting works. Am i doing something wrong?
Comment 1 Lars Müller 2009-03-28 11:57:10 UTC
Created attachment 282688 [details]
Patch as suggested by Manfred on the openSUSE ML
Comment 2 Milan Vančura 2009-10-15 15:03:10 UTC
This is Michal's part. Michal, do you agree with the patch?
Comment 3 Michal Marek 2009-10-15 15:16:02 UTC
Thanks for sitting on another bug for more than half a year :(
Comment 4 Michal Marek 2009-11-02 14:28:15 UTC
Do I get it right that
* the root fs is on a lvm volume inside /dev/md1
* /dev/md0 contains /boot and is not needed to mount the root fs
* booting with just /dev/md1 doesn't work
* adding /dev/md0 to the initrd mdadm.conf magically fixes it?

That looks really strange, it *could* be that you are in fact hitting bug 445490 and adding the other md device avoids the race condition. Are you able to reproduce this bug with all updates applied?
Comment 5 flo gleixner 2009-11-05 15:25:34 UTC
Had the same problem. Problem was solved with patch below. Not sure, if this is officially fixed, but the script /lib/mkinitrd/scripts/setup-md.sh was patched later. I think it works now, because kernel updates worked.
Comment 6 Michal Marek 2009-11-05 15:56:56 UTC
That patch is not a solution, at best it hides some race. Are you able to reproduce this with all official updates applied and no local modifications ('rpm -V mkinitrd mdadm' should not print anything)?
Comment 7 flo gleixner 2009-11-07 09:39:07 UTC
'rpm -V mkinitrd mdadm' did not print anything and recent kernel updates worked - i think the bug is fixed somehow.
Comment 8 Michal Marek 2009-11-09 09:48:36 UTC
OK, thanks for testing. Let's hope it was the same race as in bug 445490.

*** This bug has been marked as a duplicate of bug 445490 ***