I run into this every so often and it's an annoying stumbling block, normally the result is CentOS will throw some error at start of the install. This is normally because the drives either had mdraid or some software controller RAID on them so CentOS crashes when trying to parse the drives... well at least that's what I think.

A simple quick fix to destroy all signatures (before I was dd'ing the whole drive..., not good for onsite visits...)

https://redhatlinux.guru/2016/08/24/how-to-remove-mdadm-raid-devices/ \\
https://www.systutorials.com/how-to-clean-raid-signatures-on-linux/

Get list of drives that you want to wipe clean with lsblk then:
<code>
dd if=/dev/zero of=/dev/sda bs=512 count=1024
</code>
Repeat for each drive.

Sometimes if you had a ZFS partition it will still show, to erase the ZFS label/ID use fdisk to write (w) to the drive and it will clear the ZFS label
<code>
fdisk /dev/sda
w
</code>

Now reboot and lsblk again and all should be blank. Though a mdraid device did survive this and the steps below were needed as well.

List mdadm RAID devices
<code>
cat /proc/mdstat
</code>
Stop them
<code>
mdadm --stop /dev/md1
</code>
Remove them
<code>
mdadm --remove /dev/md1
</code>
If you get an error on removal this:
<code>
mdadm --zero-superblock /dev/sdf1 /dev/sde1
</code>
Check again
<code>
cat /proc/mdstat
</code>