As OpenShift Virtualization adoption continues to accelerate, we are seeing more platform administrators who are NOT familiar with the concept of containers, container images and image registries.  While the process to create a VDDK image is well documented, it still assumes a basic understanding of containerization and also assumes the availability of a container image registry.

The workflow should be made simpler. While the admin will still need to download the VDDK from VMware to stay within licensing constraints, the remainder of the process should be streamlined, not requiring the use of podman, containers or an image registry.

Four ways to handle this occur to me:

1. Upload the vddk as a config map. This seems like a anti-pattern as storing a 36MB file in etcd sounds like a bad idea
2. Upload the vddk to a RWX filesystem so that it can be mounted to the migration pod. This assumes that a PV that support RWX file is available in the cluster. While ODF can supply this, not all OCP-V installs are using ODF
3. Upload to an S3/object bucket and pull from there. This also would work well if ODF is available in the cluster, but not all clusters will have ODF
4. Upload the file to a PV, and then have a "helper pod" that makes the file available via HTTP. As part of the migration pod start-up, it can down load a copy of the VDDK and then extract and use it.