cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1088
Views
20
Helpful
1
Replies

Kernel Panic after Context Directory Agent (CDA) Patches

Hello,

Ive installed the CDA on a Hyper-V Host. There is no choice for a VMWare Host. I know sadly, that this is not offical supported. Otherweise the CDA runs with just two problems which could be possible solved.

The NTP Service seems not running as it should. Ive tried several NTP Servers from LAN and WAN, but the clock runs 3-4 seconds per day to fast or to slow. Without NTP Servers (and manually set the clock) its the same. After a few days the CDA is useless.

My hope was a patch release, so I installed patch 1 without problems. But the clock error wasnt away. I installed patch 2 and the bootscreen said everytime:

mount: could not find filesystem /dev/root

[...]

kernel panic - not syncing: attepmted to kill init!

I installed and configured the CDA from beginning and tried patch 3 with the same result of kernel panic. Any suggestions what I can do? I need the CDA for my ASA.

Regards

Andreas

1 Accepted Solution

Accepted Solutions

phill.campbell
Level 1
Level 1

I appreciate this post is fairly old, but to anyone encountering this problem, you need to rebuild the Initial RAM Disk as follows:

  1. Download and boot a CentOS 5 ISO
  2. Type "linux rescue" at the boot prompt
  3. When asked, mount filesystems as read-write
  4. Chroot into the system using the following commands:
    1. mount --bind /proc /mnt/sysimage/proc
    2. mount --bind /dev /mnt/sysimage/dev
    3. mount --bind /sys /mnt/sysimage/sys
    4. chroot /mnt/sysimage
  5. Backup current boot image
    1. cp -p /boot/initrd-2.6.18-348.el5PAE.img /boot/initrd-2.6.18-348.el5PAE.img.bak
  6. Create new initial ram disk with Hyper-V modules
    1. mkinitrd -f -v /boot/initrd-2.6.18-348.el5PAE.img 2.6.18-348.el5PAE --preload hv_storvsc --preload hv_vmbus --preload hv_utils
  7. Reboot

Note that you may need to change the kernel version depending on your patch level. The above commands are based on patch 5.

References:

https://wiki.centos.org/TipsAndTricks/CreateNewInitrd

http://blog.earth-works.com/2013/05/14/fan-fully-automated-nagios-hyper-v-kernel-panic-switchroot-mount-failed-no-such-file-or-directory/

View solution in original post

1 Reply 1

phill.campbell
Level 1
Level 1

I appreciate this post is fairly old, but to anyone encountering this problem, you need to rebuild the Initial RAM Disk as follows:

  1. Download and boot a CentOS 5 ISO
  2. Type "linux rescue" at the boot prompt
  3. When asked, mount filesystems as read-write
  4. Chroot into the system using the following commands:
    1. mount --bind /proc /mnt/sysimage/proc
    2. mount --bind /dev /mnt/sysimage/dev
    3. mount --bind /sys /mnt/sysimage/sys
    4. chroot /mnt/sysimage
  5. Backup current boot image
    1. cp -p /boot/initrd-2.6.18-348.el5PAE.img /boot/initrd-2.6.18-348.el5PAE.img.bak
  6. Create new initial ram disk with Hyper-V modules
    1. mkinitrd -f -v /boot/initrd-2.6.18-348.el5PAE.img 2.6.18-348.el5PAE --preload hv_storvsc --preload hv_vmbus --preload hv_utils
  7. Reboot

Note that you may need to change the kernel version depending on your patch level. The above commands are based on patch 5.

References:

https://wiki.centos.org/TipsAndTricks/CreateNewInitrd

http://blog.earth-works.com/2013/05/14/fan-fully-automated-nagios-hyper-v-kernel-panic-switchroot-mount-failed-no-such-file-or-directory/

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: