CentOS 7 was working fine, then discovered to be not running. Tried to boot one of the listed kernels, and it appeared to work once - but tried to do software update - broke - and can't get it started any more. Always stops at "Failed to load SELinux policy. Freezing"

How do I recover from this one?

Second question: Once recovered - how do I make a boot disk for future incidents ?

Thanks much !

Update:
Managed to edit /etc/selinux/config to turn off SELINUX
System now boots, but not long after logon, it tells me it ran into an error it couldn't recover from -
logoff and try again.

New slant on the problem. Won't stay on for long. I discovered YUM is probably messed up - because I have
to tell it to complete transactions. Can't stay logged on long enough to download software updates.
Should I try to install a newer version of CentOS (using v7 now)? And if I do that, do I lose my all my web files?
Sorry - too much going on with this one....

is it just the gui that dies or is the whole system, what happens if you boot into runlevel 3? assuming the system stays up long enough you can rebuild the rpm database.

have checked for hardware issues? hard drive dying, bad stick of ram maybe, or power supply going bad.

Thanks so much for the reply and assist !
When I couldn't boot because of the "policy" error, I tried to download and update the software updates. However it broke somewhere in the middle of that process, and got - worse.

I tried to rebuild YUM, but I suspect I'm in a downward spiral.

In response to your questions:
I now have told the system SELINUX=disabled so it now boots.
I get on the GUI, and I do believe the GUI is breaking, because when I click on the "logout" button in the "OH NO..." message, it goes back
to the logon screen (not a reboot). So, it looks like the system is still running, but can't stay on the GUI very long.

I booted into runlevel3, and it's now leaving me logged on - apparently as long as I want !
I did a "ps aux" and discovered something I think may be interesting - unless this is "normal".
I see two lines:

root 942 0.0 0.0 211984 4516 ? Ss May16 0:00 /usr/bin/abrt-watch-log -F BUG: WARNING: CPU: INFO: possible recursive locking detected ernel BUG at list_del corruption list_add corruption do_IRQ: stack overflow

root 951 0.0 0.0 211984 4504 ? Ss May16 0:00 /usr/bin/abrt-watch-log -F Backtrace /var/log/Xorg.0.log -- /usr/bin/abrt-dump-xorg -xD


So, I'm wondering if
- I can rebuild YUM and install the software updates.
- perhaps fix this stack overflow problem?

I have tried to download a memory check software. I was supposed to make an "ISO" image, but I don't know how to do that.

Note that I have THREE CentOS 7 systems. Two are not being used yet. I want to turn one of them into an AMANDA backup system
for our local net. So, if I can benefit by comparing or copying some files - I have that available. I was hoping to make the memory
check DVD disk on one of the other systems, but I need some DVD burner software.

Thanks !

https://wiki.gnome.org/Apps/Brasero

there is also the command line tool cdrecord.

I finally created a boot disk for Memtest86. I ran the checker for about an hour. All I saw was "pass", except I wasn't at the console when I think it finished and rebooted. I'm thinking the memory is fine. Will check the power supply, etc. If all is "well", I will have to figure out how to re-do the Secure Linux (still turned off).