After upgrading the cluster from 4.12 to 4.13. Nodes getting booted into emergency mode. Due to error 
`blockdev: cannot open /dev/dasda2: No such file or directory` 

From the sosreport collected after successful boot we could see that there were following symlinks setup in /dev/disk/by-label:	


	[sosreport]$ less sos_commands/block/ls_-lanR_.dev 
	[...]
	/dev/disk/by-label:
	total 0
	drwxr-xr-x. 2 0 0 100 Apr 24 10:09 .
	drwxr-xr-x. 7 0 0 140 Apr 24 10:09 ..
	lrwxrwxrwx. 1 0 0  12 Apr 24 10:09 boot -> ../../dasda1
	lrwxrwxrwx. 1 0 0  12 Apr 24 10:09 crypt_rootfs -> ../../dasda2
	lrwxrwxrwx. 1 0 0  10 Apr 24 10:09 root -> ../../dm-0                   <<----------


The command outputs collected from emergency mode, during failed boot process, shows that   "root -> ../../dm-0" link was not setup in by-label directory. However /dev/dm-0 device was setup by the time boot process failed:

Command outputs from emergency mode:


	[Console logs]$ less 0200-worker-3-emergency-mode.txt 
	[...]
	11:56:41 ls -l /dev/disk/by-label/                                              
	11:56:42  ¬?2004l 11:56:42 total 0                                              
	11:56:42 lrwxrwxrwx 1 root root 12 Apr 26 08:11 boot -> ../../dasda1            
	11:56:42 lrwxrwxrwx 1 root root 12 Apr 26 08:11 crypt_rootfs -> ../../dasda2    		<<---------- "root -> ../../dm-0" symlink is missing


After multiple retries it gets booted successfully. 

4.13.36

NA

    1. Upgrade cluster to 4.13
    2. Check the master and worker node boot
    3. Observe the nodes if they booting in emergency mode and collect console logs.
    

Node went into emergency mode

Node should boot successfully without any issue.

Customer is using Zvm to VM provisioning.