Kernel panic installing RHEL 5.3

[Rick Stevens]

Brenda Radford wrote:
> 
> -----Original Message-----
> From: redhat-install-list-bounces at redhat.com
> [mailto:redhat-install-list-bounces at redhat.com] On Behalf Of Rick Stevens
> Sent: Monday, June 01, 2009 2:16 PM
> To: Getting started with Red Hat Linux
> Subject: Re: Kernel panic installing RHEL 5.3
> 
> Brenda Radford wrote:
>> My install of RHEL 5.3 is giving me a Kernel panic message. This is the
>> output on the screen.
>>
>>  
>>
>> Memory for crash kernel (0x0 to 0x0) notwithin permissible range
>>
>> Red Hat nash version 5.1.19.6 starting
>>
>>     Reading all physical volumes. This may take a while . . .
>>
>>     Found volume group "VolGroup00" using metadata type lvm2
>>
>>     /proc/misc: No entry for device-mapper found
>>
>>     Is device-mapper driver missing from kernel?
>>
>>     Failure to communicate with kernel device-mapper driver.
>>
>>     0 logical volume(s) in volume group "VolGroup00" now active
>>
>> Unable to access resume device (dev/VolGroup00/LogVol01)
>>
>> mount: Could not find filesystem '/dev/root'
>>
>> setuproot: moving /dev failed: No such file or directory
>>
>> setuproot: error mounting /proc: No such file or directory
>>
>> setuproot: error mounting /sys: No such file or directory
>>
>> switchroot: mount failed: No such file or directory
>>
>> Kernel panic - not syncing: Attempted to kill init!
>>
>> What do I do?
> 
> I have seen this on a number of systems under CentOS 5.3.  I'm not sure
> what's causing the problem, but it appears to be a driver in the kernel
> with 5.3 that's required in the initrd for the kernel but isn't being
> included by mkinitrd or anaconda.
> 
> I haven't seen this issue with 5.2 systems updated to 5.3--just raw 5.3
> installs--which leads me to believe it's an issue with anaconda as
> opposed to the kernel or mkinitrd.
> 
> So, install 5.2, update to 5.3 and you should be good to go.  I would
> file a bugzilla about this.  Be specific.  Include the computer model
> (or motherboard make/model), disk controller and disk types, video card,
> everything.  The more information they have, the better chance they have
> of fixing it.
> ----------------------------------------------------------------------
> - Rick Stevens, Systems Engineer                      ricks at nerd.com -
> - AIM/Skype: therps2        ICQ: 22643734            Yahoo: origrps2 -
> -                                                                    -
> -      Cuteness can be overcome through sufficient bastardry         -
> -                                         --Mark 'Kamikaze' Hughes   -
> ----------------------------------------------------------------------
> 
> 
> Rick,
> 
> While I was waiting for replies from the list, my looking for a job linux
> bigot friend Brad told me to redo the install, choosing a text install
> instead of a GUI install. It worked for some reason. There was no kernel
> panic on the second install. I used the same CDs I used the first time,
> although I chose different packages the second time (the GUI screen was
> wider than my wide monitor). I had burned the DVD and had planned to use
> that for the install, but that is a long story.
> 
> But I will file a bugzilla and include the fact that I did a GUI install and
> got a kernel panic, if you think it will help.

Absolutely.  File the bugzilla.  There can be some rather nasty
interactions between the X driver for the display and the kernel.  A
text install is much "easier" on the machine.

After your text installation (and if you're feeling adventuresome), you
can install the X stuff and see if the problem pops up again.  My guess
is that it won't and you've found a problem between the X driver,
installer kernel and anaconda.

This same thing was a big problem early in the Fedora 11 development
cycle with Intel graphics cards.  They've got it more-or-less sorted
now, but anything's possible.

Note that RHEL 5 is based on Fedora 6.  I don't know which Fedora will
be snapshotted and used as the guts for RHEL6, but with the teething
problems F11 has had, my guess it WON'T be Fedora 11.  Consequently, you
should let Red Hat know about this as there probably will be a RHEL 5.4
release before RHEL 6.0 and they don't want this sort of gotcha to be 
perpetuated.

> Thanks to you and to Engin for his reply.

Glad to try to help, but you seem to have solved it yourself.  Well
done!  And thanks for filing that bugzilla.  Not many people would take
the time.

Be as specific as you can as to the errors you saw and your hardware
configuration in the bugzilla.  As root, run "lspci", "lsusb" and
"dmidecode", save their output to a text file:

	echo "lspci output:" >/tmp/hwrpt.txt
	lspci >>/tmp/hwrpt.txt 2>&1
	echo "" >>/tmp/hwrpt.txt
	echo "lsusb output:" >>/tmp/hwrpt.txt
	lspci >>/tmp/hwrpt.txt 2>&1
	echo "" >>/tmp/hwrpt.txt
	echo "dmidecode output:" >>/tmp/hwrpt.txt
	dmidecode >>/tmp/hwrpt.txt 2>&1

then attach the /tmp/hwrpt.txt file to your report.
----------------------------------------------------------------------
- Rick Stevens, Systems Engineer                      ricks at nerd.com -
- AIM/Skype: therps2        ICQ: 22643734            Yahoo: origrps2 -
-                                                                    -
-           "I understand Windows 2000 has a Y2K problem."           -
----------------------------------------------------------------------