loop-lvm is only appropriate for testing. Loopback devices are slow and resource-intensive, and require you to create a file on disk at specific sizes. They can also introduce race conditions. They are supposed for testing because the set-up is easier. - as noted here
By creating a loopback block device as such:
dd if=/dev/zero of=/opt/docker-dev-pool bs=1M count=25000 mknod /dev/docker-dev0 b 7 200 losetup /dev/docker-dev0 /opt/docker-dev-pool
Does using this loopback device when configuring direct-lvm as specified here result in the same issues as just using loop-lvm?
Essentially, are you required to use a “physical” block device in order to gain the benifits of direct-lvm?