[CentOS] Unable to boot CentOS 6 - Segmentation Erorr

Sun May 29 00:27:02 UTC 2016
John Cenile <jcenile1983 at gmail.com>

Hi all,
I had an issue this morning with one of my virtual machines. It wouldn't
boot (into any runlevel), nor could I chroot into the root partition using
a rescue disk.

Unfortunately I didn't grab a screenshot, however the error(s) when booting
were:


/pre-pivot/50selinux-loadpolicy.sh: 14

<other messages>

init: readahead main process (425) killed by SEGV signal
init: readahead-col lector main process (421) killed by SEGV signal
init: rcS pre-start process (425) killed by SEGV signal
init: Error while reading from descriptor: Bad file descriptor
init: readahead-col lector post-stop process (424) killed by SEGV signal
init: rcS post-stop process (427) killed by SEGV signal
init: readahead-disable-services main process (428) killed by SEGV signal


When using a rescue CD and chrooting into the root partition, I would get :

Segmentation Fault: Core Dumped


In the end, the fix was to boot into a rescue CD with networking, and SCP
the entire contents of /bin and /sbin from another (working) server to the
broken installation. This finally allowed CentOS 6 to boot correctly.

So I'm left to assume some of the files in /bin *or */sbin were corrupt.

My question is, does anyone have any ideas on how this might have happened?
I did do a quick memory test using the rescue CD (it didn't complete) and
there weren't any errors. The virtual machine is running on VMWare with 3
other VMs, which all seem fine. There wasn't any unexpected power loss
either.

Thanks.