* dmraid - possible boot fail due to mdraid rules?
@ 2015-01-09 12:50 Miroslav Shaltev
[not found] ` <20150109135054.66bac1f1-h/dpFJZgkqBSq9BJjBFyUp/QNRX+jHPU@public.gmane.org>
0 siblings, 1 reply; 2+ messages in thread
From: Miroslav Shaltev @ 2015-01-09 12:50 UTC (permalink / raw)
To: initramfs-u79uwXL29TY76Z2rM5mHXA
Dear all,
i have a question regarding the dmraid module.
On a centos 7 system (but i am afraid the problem is distro independent)
kernel: 3.10.0-123.13.2.el7.x86_64
dracut: dracut-033-161.el7.x86_64
storage: dmraid 1 with lvm on top
with a root on a dmraid device, some mdraid udev rules get pulled into
the initramfs.
The output of dracut -M
bash i18n network ifcfg drm plymouth dm dmraid kernel-modules lvm
resume rootfs-block terminfo udev-rules biosdevname systemd usrmount
base fs-lib shutdown
gives no hint about that. However
grep -r "64-md-raid.rules" dracut/modules.d/*
dracut/modules.d/90dmraid/module-setup.sh: inst_rules 64-md-raid.rules
dracut/modules.d/90mdraid/module-setup.sh: inst_rules 64-md-raid.rules
Due to the mdraid rules, the actual dmraid device cannot be activated,
as a result the /dev/mapper links are pointing to wrong devices and the
boot fails.
Commenting out the line in dracut/modules.d/90dmraid/module-setup.sh
allows me to create bootable initramfs images.
Could it be, that the line in dracut/modules.d/90dmraid/module-setup.sh
was introduced just by a mistake or should i fill a bug report?
Cheers,
miroslav
--
Dr. Miroslav Shaltev
- Systemadministration -
Forschungszentrum L3S
Appelstrasse 9a
30167 Hannover
Fax: +49 (0)511 762 - 17779
E-Mail: shaltev-ng3u6l8Qjkg@public.gmane.org
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: dmraid - possible boot fail due to mdraid rules?
[not found] ` <20150109135054.66bac1f1-h/dpFJZgkqBSq9BJjBFyUp/QNRX+jHPU@public.gmane.org>
@ 2015-01-21 13:04 ` Harald Hoyer
0 siblings, 0 replies; 2+ messages in thread
From: Harald Hoyer @ 2015-01-21 13:04 UTC (permalink / raw)
To: Miroslav Shaltev, initramfs-u79uwXL29TY76Z2rM5mHXA
On 09.01.2015 13:50, Miroslav Shaltev wrote:
> Dear all,
>
> i have a question regarding the dmraid module.
>
> On a centos 7 system (but i am afraid the problem is distro independent)
>
> kernel: 3.10.0-123.13.2.el7.x86_64
> dracut: dracut-033-161.el7.x86_64
> storage: dmraid 1 with lvm on top
>
> with a root on a dmraid device, some mdraid udev rules get pulled into
> the initramfs.
>
> The output of dracut -M
>
> bash i18n network ifcfg drm plymouth dm dmraid kernel-modules lvm
> resume rootfs-block terminfo udev-rules biosdevname systemd usrmount
> base fs-lib shutdown
>
> gives no hint about that. However
>
> grep -r "64-md-raid.rules" dracut/modules.d/*
>
> dracut/modules.d/90dmraid/module-setup.sh: inst_rules 64-md-raid.rules
>
> dracut/modules.d/90mdraid/module-setup.sh: inst_rules 64-md-raid.rules
>
> Due to the mdraid rules, the actual dmraid device cannot be activated,
> as a result the /dev/mapper links are pointing to wrong devices and the
> boot fails.
>
> Commenting out the line in dracut/modules.d/90dmraid/module-setup.sh
> allows me to create bootable initramfs images.
>
> Could it be, that the line in dracut/modules.d/90dmraid/module-setup.sh
> was introduced just by a mistake or should i fill a bug report?
>
> Cheers,
> miroslav
>
Oh wow.. actually, that dates back to commit b2a843b
$ git show b2a843b4
commit b2a843b471931a8686fffc76cd0afeacabdc5b63
Author: Victor Lowther <victor.lowther-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date: Fri Mar 6 15:26:24 2009 -0600
Make 90dmraid load the dm-raid udev rules instead of 95udev-rules
And yes, this is a bug. Please file a bug report!
Thanks for your discovery!
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2015-01-21 13:04 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-01-09 12:50 dmraid - possible boot fail due to mdraid rules? Miroslav Shaltev
[not found] ` <20150109135054.66bac1f1-h/dpFJZgkqBSq9BJjBFyUp/QNRX+jHPU@public.gmane.org>
2015-01-21 13:04 ` Harald Hoyer
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox