I have read a few tutorials on setting up an encrypted file system using cryptsetup. They all start with the following creation of a random file
dd if=/dev/urandom of=/etc/cryptfile bs=1M count=10
This file would be further used to create a loopback device. The reason given for the randomising the file is generally given that the attacker won't be able to find what parts of the file are used to write and what parts are empty. My question is that since the loop back device is any way going to be formatted when we write the file system on it then why would we care about randomizing it in the first place?