Server with OS - Red Hat. Kernel 2.6.32-279.el6.x86_64

The server is spinning on VmWare. It worked fine, but suddenly, in the morning, I stopped letting the user in. I decided to restart the server. After the reboot, the server fell into the Kernel Panic.

Grub.conf config

root (hd0,0) kernel /vmlinuz-2.6.32-279.el6.x86_64 ro root=UUID=357f520e-d03a-4990-996b-06b24ac965bb rd_NO_LUKS KEYBOARDTYPE=pc KEYTABLE=us LANG-en_US.UTF-8 rd_NOMD SYSFONT=latarcyrheb-sun16 crashkernel=auto rd_NO_LVM biosdevname=0rd_NO_DM rhgb quiet initrd /initramfs-2.6.32-279.el6.x86_64.img

From the boot logs you can see that the problem with SELinux. Loading SELinux policy ,

Although the load_policy file is in place.

 [root@localhost sbin]# ls -la | grep load_ -rwxr-xr-x. 1 root root 11216 Mar 6 06:04 load_policy

If you disable SELinux when loading the kernel, it crashes into a new error: enter image description here

And in fact, I read that it does not make sense to disable SELinux.

I found a similar problem with xen , but I didn’t quite realize how he solved it. If someone translates on fingers, I will be grateful.

Update: chroot

Update2: bash.static

Update3: bash.static.ldd

Update4: fstab

  • one
    @PetrAbdulin, you should not close (and argue about this) (there are words on the link that OS administration issues are resolved). And where else can a man solve his problem? - avp
  • one
    @avp hmm, yes indeed, probably this list has changed recently, or I have carelessly read it (naively believing that it is equivalent to the original SO). Well, if it is allowed, then there are no questions, I apologize. - Petr Abdulin
  • 2
    root=UUID=357f520e-d03a-4990-996b- 06b24ac965bb - is there really a space (even two) or is it a copy error? - aleksandr barakin
  • one
    Here is a person who writes about solving a similar problem, because some lib64 packages that chroot needs to have a perfect behavior was not included inside guest filesystem that I had created When these pachages were included into /lib64 of the guest root filesystem everything worked well. / Maybe you (or auto-update system) did something with / lib64? / To be honest, how specifically to restore in VmWare (remount the disk to another virtual machine and roll there?) I do not know. - avp
  • four
    @ 1d0, do you have administrative access (VmWare vCenter or something like that)? / Ie First you need to see if there are any problems with a real physical disk. - avp

2 answers 2

Most likely the file system is damaged. You need to access the disk in any way without mounting the problem filesystem, then:

 fsck -f /dev/sda2

    As a result, the root partition, one folder from a disk image, was thrown into the crashed system.