CentOS - Oct 2016 - Cannot boot CentOS 7 VM after updating Host CentOS 7 Kernel

so, Just chroot to mountpoint:

http://www.cyberciti.biz/faq/unix-linux-chroot-command-examples-usage-syntax/

chroot /mounted/path /bin/bash and then .. mkinitrd (see man page for
documentation)

2016-10-30 22:57 GMT+02:00 Eero Volotinen <eero.volotinen at iki.fi>:
> A bit hard to say. Try chrooting into environment and rebuilding initrd?
>
> --
> Eero
>
> 2016-10-30 22:53 GMT+02:00 Paul R. Ganci <ganci at nurdog.com>:
>
>> On 10/30/2016 12:26 PM, Paul R. Ganci wrote:
>>
>>> <snip>I am thinking of putting the CentOS iso out and then
booting the
>>> VM into it just to poke around the file system. Otherwise my other
option
>>> is to just clone a twin VM on another server and then just change
the
>>> networking IPs/hostname. Anybody have any other ideas as to how to
debug
>>> this problem?
>>>
>> So I booted off the CentOS-7-x86_64-DVD-1511.iso and everything looks
>> just fine:
>>
>> > df
>> Filesystem                         1K-blocks        Used Available
>>  Use%   Mounted on
>> /dev/mapper/live-rw           2030899    949022    1077781     47% /
>> devtmpfs                           2004040              0 2004040
>>  0%   /dev
>> tmpfs                                 2023652              0 2023652
>>    0%   /dev/shm
>> tmpfs                                 2023652        8520 2015132
>>  1%   /run
>> tmpfs                                 2023652              0 2023652
>>    0%  /sys/fs/cgroup
>> /dev/sr1                             4227724  4227724 0   100%
>> /run/install/repo
>> tmpfs                                 2023652          200 2023452
>>  1%  /tmp
>> /dev/mapper/centos-root  10799104  3894196    6904908     37%
>> /mnt/sysimage
>> /dev/vda1                            508588     143516 365072     29%
>> /mnt/sysimage/boot
>> tmpfs                                2023652               0 2023652
>>    0%  /mnt/sysimage/dev/shm
>>
>> > ls /mnt/sysimage
>> bin   boot   dev   etc   home   lib   lib64   media   misc   mnt net
>>  opt   proc   root    run    sbin   srv   sys    tmp   usr var
>>
>> > ls -l /mnt/sysimage/boot
>> total 109424
>> -rw-r--r--.    1 root root       126431  Oct 10 23:18
>> config-3.10.0-327.36.2.el7.x86_64
>> drwxr-xr-x. 2 root root               26  Oct   2  2015  grub
>> drwx------.  6 root root             104  Oct 13 02:21  grub2
>> -rw-r--r--.    1 root root   40655493  Apr   3  2015
>> initramfs-0-rescue-6494b5d98adc4f66b0cf4c19a0f6ab66.img
>> -rw-------.    1 root root   29666884 Oct 13  01:25
>> initramfs-3.10.0-327.36.2.el7.x86_64.img
>> -rw-------.    1 root root   18119089  Oct 13 02:20
>> initramfs-3.10.0-327.36.2.el7.x86_64kdump.img
>> -rw-r--r--.    1 root root   10190975  Dec 19  2015 initrd-plymouth.img
>> -rw-r--r--.    1 root root      252739  Oct  10 23:20
>> symvers-3.10.0-327.36.2.el7.x86_64.gz
>> -rw-------.    1 root root    2965270  Oct  10 23:18
>> System.map-3.10.0-327.36.2.el7.x86_64
>> -rwxr-xr-x.   1 root root    4902656  Apr    3  2015
>> vmlinuz0-rescue-6494b5d98adc4f66b0cf4c19a0f6ab66
>> -rwxr-xr-x.   1 root root    5157936   Oct  10 23:18
>> vmlinuz-3.10.0-327.36.2.el7.x86_64
>>
>> So the CentOS DVD iso in linux rescue mode shows that everything is
there
>> and can be mounted. I guess that means somehow either grub itself is
>> corrupted or one of the boot images. So is there a way for me to
generate a
>> new initrd while booted in linux resuce mode or will re-installing grub
>> help? How would I attempt re-installing grub while booted in linux
rescue
>> mode?
>>
>> --
>> Paul (ganci at nurdog.com)
>> Cell: (303)257-5208
>> _______________________________________________
>> CentOS mailing list
>> CentOS at centos.org
>> https://lists.centos.org/mailman/listinfo/centos
>>
>
>

On 10/30/2016 03:05 PM, Eero Volotinen wrote:
> so, Just chroot to mountpoint:
>
>
http://www.cyberciti.biz/faq/unix-linux-chroot-command-examples-usage-syntax/
>
> chroot /mounted/path /bin/bash and then .. mkinitrd (see man page for 
> documentation)
Thank you for the hint. The way I fixed this problem was to do this:

1.) Booted into linux rescue mode using the CentOS-7-x86_64-DVD-1511.iso
2.) chroot /mnt/sysimage
3.) Used nmtui to setup a network
4.) yum update (installed the 3.10.0-327-36.3.el7.x86_64 kernel)
5.) grub2-mkconfig -o /boot/grub2/grub.cfg 6.) yum reinstall 
kernel*-3.10.0-327-36.3.el7.x86_64

Step 5 was necessary because of a grubby error indicating that there was 
no suitable template. Step 6 was necessary because according to the 
googled page I found regarding step 5 indicated that the default kernel 
would not be that found in step 4. I probably could have skipped step 4 
and just done steps 5 and 6. But the VM is now up and running the latest 
kernel.

Now the question is how did this happen. My best guess is that the VM 
was unhappy when I booted the host while it was running. Is rebooting 
the host without shutting down guests first a risky thing to do?
-- 
Paul (ganci at nurdog.com)
Cell: (303)257-5208

> On Oct 30, 2016, at 7:31 PM, Paul R. Ganci <ganci at nurdog.com>
wrote:
> 
> Now the question is how did this happen.
I've seen something similar when installing a kernel if /etc/fstab
didn't match df. Mkinitrd bombs out leaving the system unbootable. The
rescue .iso/mkinitrd path you followed was the fastest way to get the system up.