Background

Today we support partitioning devices during install time, via Machineconfigs that are processed during bootstrap. However there are limitations to this, one of which is that the partition name must be determined during install time and is static. This means that users cannot add e.g. different instance types as nodes if they have different device naming schema. This is also not configurable post-install, since the MCO does not allow for modification of partitions post-install.

This Epic is not aiming to track a long term solution, but rather to document an existing workaround, test its operation, and ensure we are able to support it.

The workaround in question makes use of custom user-data secrets and modifying machinesets to use these secrets, which contain a snippet of ignition that does the partitioning. For details please see:

Associated bugzillas (private):

https://bugzilla.redhat.com/show_bug.cgi?id=1944035

https://bugzilla.redhat.com/show_bug.cgi?id=1969261

Epic Goal

• Test the operation of different partition device naming
• Document and officially support this

Why is this important?

• Customer request

Scenarios

1. Post-install, a user wishes to add different instance types to the cluster, to run e.g. specific workloads

Acceptance Criteria

• CI
• Official Openshift Documentation
• ...

Dependencies (internal and external)

1. ...

Previous Work (Optional):

1. …

Open questions::

1. …

Done Checklist

• CI - CI is running, tests are automated and merged.
• Release Enablement <link to Feature Enablement Presentation>
• DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
• DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
• DEV - Downstream build attached to advisory: <link to errata>
• QE - Test plans in Polarion: <link or reference to Polarion>
• QE - Automated tests merged: <link or reference to automated tests>
• DOC - Downstream documentation merged: <link to meaningful PR>