Re: MD device not found on boot



On Friday, 02 March 2012 10:52:37 +0100,
tv.debian@xxxxxxxxxxxxxx wrote:

Currently I have a host with Debian GNU/Linux Lenny on which I've
compiled a 2.6.32 kernel (using the Backports source code) to add
support of Layer7 and IMQ.

The package with the Backports kernel binary works well, but when I boot
my compiled kernel, MD device on / is not found. The configuration file
that I took as a reference is the provided by the installed Backports
kernel binary, so I think it should include modules for MD, and then I
could confirm this in the configuration file used to compile my kernel.

Perhaps it is a problem in generating the initrd file? The process I
used to compile and install the kernel is the Debian way, creating a
package then installed with dpkg.

Hi,

Hi!

did you check if /etc/default/mdadm contains necessary options:

INITRDSTART='all'

AUTOSTART=true

Yes, /etc/default/mdadm contains these two options

Is mdadm.conf up to date ?

Can you tell us what kind of raid and metadata we are talking about.

This is the detail of the root device:

# mdadm --detail /dev/md2
/dev/md2:
Version : 00.90
Creation Time : Wed Sep 6 11:27:09 2006
Raid Level : raid1
Array Size : 19534976 (18.63 GiB 20.00 GB)
Used Dev Size : 19534976 (18.63 GiB 20.00 GB)
Raid Devices : 2
Total Devices : 1
Preferred Minor : 2
Persistence : Superblock is persistent

Update Time : Fri Mar 2 11:41:16 2012
State : clean, degraded
Active Devices : 1
Working Devices : 1
Failed Devices : 0
Spare Devices : 0

UUID : da5a0c19:c38ccc6d:325196eb:eb163021
Events : 0.5276104

Number Major Minor RaidDevice State
0 0 0 0 removed
1 8 3 1 active sync /dev/sda3

The /etc/mdadm/mdadm.conf used metadata=00.90. I changed it to 0.90, if
this may cause some inconvenience (metadata format 00.90 unknown,
ignored). But I don't think the problem is this because
2.6.32-bpo.5-amd64 boots without problems and my compiled kernel is
based on config-2.6.32-bpo.5-amd64; I only patch the kernel with support
of IMQ and Layer7.

This is the output with default kernel options:

Gave up waiting for root device. Common problems:
- Boot args (cat /proc/cmdline)
- Check rootdelay= (did the system wait long enought?)
- Check root= (did the system wait for the right revice?)
- Missing modules (cat /proc/mdoules: Is /dev)
ALERT! /dev/md2 does not exist. Dropping to a shell!

And using rootdelay=120:

Success: loaded module raid1.
done
Begin: Assembling all MD arrays... mdadm: metadata format 00.90 unknown ignored.
mdadm: metada format 00.90 unknown, ignored
mdadm: metada format 00.90 unknown, ignored
mdadm: metada format 00.90 unknown, ignored
mdadm: metada format 00.90 unknown, ignored
mdadm: No devices listed in conf file were found
Failure: failed to assamble all arrays.
done.
Begin: Waiting for udev to process events ... done
done
Begin: waiting for root filesystem ... done
Gave up waiting for root device. Common problems:
- Boot args (cat /proc/cmdline)
- Check rootdelay= (did the system wait long enought?)
- Check root= (did the system wait for the right revice?)
- Missing modules (cat /proc/mdoules: Is /dev)
ALERT! /dev/md2 does not exist. Dropping to a shell!



Thanks for your reply.


Regards,
Daniel
--
Fingerprint: BFB3 08D6 B4D1 31B2 72B9 29CE 6696 BF1B 14E6 1D37
Powered by Debian GNU/Linux Lenny - Linux user #188.598

Attachment: signature.asc
Description: Digital signature



Relevant Pages

  • Re: Distributions
    ... | general there are a lot of packages for people to use. ... kernel, have to run on a multitude of different systems, they tend to be ... and slower than if you compile those packages, ... can have that stability with virtually any distro. ...
    (Debian-User)
  • Re: Distributions
    ... Ubuntu is based on Debian, ... | general there are a lot of packages for people to use. ... kernel, have to run on a multitude of different systems, they tend to be larger ... and slower than if you compile those packages, ...
    (Debian-User)
  • Problems with custom kernel fbsd 6.2
    ... My kernel for freebsd 6.2 is not compiling but only with my custom config. ... I was able to compile it with the GENERIC kernel included, but my modified one will not compile. ... # Power management support ...
    (freebsd-questions)
  • Re: [PATCH] 2.6 workaround for Athlon/Opteron prefetch errata
    ... >optimisations independently of each other', is fairly simple, (in ... >concept), and elegant, (as it lets you compile the most finely tuned ... >Up to now, selecting a CPU to compile for basically means, "Use ... it's silly to include them all in a kernel for a 386. ...
    (Linux-Kernel)
  • Re: C++ pushback
    ... There are all sorts of macros that use member initialization of that form. ... This does not break the code at run time, this breaks the code at compile time, and should be less painful. ... The kernel relies really _really_ heavily on such structure initializers, and breaking them would effectively break the world as far as the kernel is concerned. ... If they were not, one could simply make a base class having members outlined, and which class does not enforce type safety and is for inheritance only. ...
    (Linux-Kernel)