There is a request to enable additional disks to be attached to the OpenShift Nodes at install time.

This applies to all the platforms supported for OpenShift and multiple use cases:

• Dedicated disk for ETCD
• Dedicated disk for SWAP
• Dedicated disk for container images
• Dedicated disk for container logs

The goal for this spike is to evaluate the work required to implement such feature into the product.

The expected outcome is to determine the feasibility of implementing this feature and the effort required for each platform as well as define if this work can be reused for every use case or if it makes more sense to break it down into multiple features, one per use case.

The expected timeline is to get all the features defined by 4.19 planning so the implementation is started during 4.19 development and completed for all the platforms and main use cases (ETCD and SWAP) by 4.20 when we expect CAPI to be GA not only for the installation but also for machine management.

At high-level, this is how priorities look like:

ETCD

1. Azure
2. AWS
3. GCP
4. Others

SWAP

1. AWS
2. Bare Metal
3. GCP
4. Others