hawkeye11
Contributor
Contributor

RHEL 4 machine converted fine, but get kernel panic when I start the VM

So I have a bunch of old Red Hat 4 servers that I need to convert to VMs.   I am using the standalone converter 5.0 and the VMs range from Red Hat version 2.6.9.-11 to 2.6.9-42.   Some of the machines I can P2V with out any issues but a majority of them fail with

mount: error 6 mounting ext3
mount: error 2 mounting none
switchroot: mount failed: 22
umount /initrd/dev failed: 2 (see pic for more details)

error.jpg

So I get a 100% complete when I use the Converter but when I boot up the VM I get that error.   I have followed the directions from http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=100240... but I still not having any luck.   I made changes to /etc/fstab, /boot/grub/grub.conf, /boot/grub/device.map etc...   I am also having issues with rebuilding the ramdisk.   I ran the command with the existing version on the drive but I get an error about about no modules found.   I have also found another article saying the version of TAR might be an issue so i've upgrade the version to 1.14-15 and re-ran converter and it still failing.   Anyone have any ideas what is wrong?    I am windows admin with some linux background so please go detailed on the linux commands.

8 Replies
sparrowangelste
Virtuoso
Virtuoso

change the scsi controller type of the vm

--------------------- Sparrowangelstechnology : Vmware lover http://sparrowangelstechnology.blogspot.com
0 Kudos
hawkeye11
Contributor
Contributor

I have tried switch between all 4 options and went back to LSI Logic Parallel.   I have also tried to convert the VMs to vmware hardware 7 and it did not make a difference.   I went as far as doing a straight copy of the HDD with ghost and I still have the same problems.   Is there something I need to do on the OS before I start the copying?

0 Kudos
hawkeye11
Contributor
Contributor

Ok so I narrowed down my issue more to RedHat 4.2 32 bit versions.   Is there any know issues with RHEL 32 bit and ESXi 5.0?

0 Kudos
KlausK
Contributor
Contributor

I ran into a kernel panic with RHEL/CentOS 5 after upgrading to ESXi 5.0 on AMD CPUs which required moving that VM to an Intel cluster and upgrading to a newer kernel version. I don't know if that applies to your environment.

0 Kudos
markekraus
Contributor
Contributor

  1. Boot into the RHEL4 rescue environment
  2. chroot /mnt/sysimage
  3. Edit /etc/modprobe.conf and change the scsi_hostadapter entries to look like this:

    alias scsi_hostadapter mptbase
    alias scsi_hostadapter1 mptscsi
    alias scsi_hostadapter2 mptfc
    alias scsi_hostadapter3 mptspi
    alias scsi_hostadapter4 mptsas
    alias scsi_hostadapter5 mptscsih

  4. cd /boot
  5. move the current initrd images to another folder or rename them to .old
  6. run mkinitrd (example: mkinitrd -v -f initrd-2.6.9.-11.EL.img 2.6.9.-11.EL)
  7. Repeat the above step for each kernel (in case you ever decide to boot into them later)
  8. Exit the chroot environment
  9. Try booting the VM

We ran into this as well and this did the trick.

Digisaster
Contributor
Contributor

Thanks Markekraus

You helped me a lot with your post!


0 Kudos
kellyer
Contributor
Contributor

Thanks Markekraus

Unfortunately I spent 6 hours troubleshooting before I found your excellent instructions

0 Kudos
PericoPB
Contributor
Contributor

Thanks. I have the same problem. 

 

0 Kudos