I was given a VMDK file with a CentOS 7 image on it for a pre-configured appliance. We do not run a VMWare shop but run Hyper-V, so I converted the VMDK to a VHDX and created a Hyper-V virtual machine attaching this VHDX virtual hard drive to it.

Upon boot the startup bar progressed until it was full and the system just sat there. pressing escape I was able to see that I was getting dracut timeouts, and finally a cannot boot message. I am able to boot into rescue mode. In rescue mode I modified the grub boot menu and disabled rhgb quiet, I basically received the same things after this except that I didn't get the loading bar splash screen. After some research I booted a live distro, mounted the drive and ran xfs_repair against the drive partition (it's xfs file format). I also verified the UUID matches what's being passed to dracut and I tried regenerating an initramfs using dracut in rescue mode.

I finally turned on rd.shell and was dropped into the emergency dracut shell shell after the timeouts. I found that there is no /dev/disk/by-uuid directory in the dracut shell, and when I exit the shell I get a warning that not all disks have been found, and that I might want to regenerate my initramfs. So I did just that, this time also making sure that the UUID's for both partitions were in the fstab file (the swap partition was not originally).

I am assuming the timeouts are dracut trying to find the disks, but I am not 100% positive.

Does anyone have any ideas how I can get this booted?

Found that the boot was detecting Hyper-V disks and ignoring them, and wasn't seeing them in the initramfs. After a little more research I saw someone said this issue was resolved in newer kernels 3.12 and up, I was on 3.10 upgraded to a newer mainline kernel and rebooted and it works wonderfully now.