With the implementation of MCO-1228, we will be populating MachineConfigPool (MCP) statuses from MachineConfigNode (MCN) statuses. This test should ensure that MCP statuses are properly updated throughout a standard node update in a custom MCP.

At a high-level, the sequence of this test will likely be something like:

1. Apply a node disruption policy.
2. Apply a MachineConfig (MC) to trigger an update.
3. Validate that the MCP statuses are properly reported throughout the update.
4. Cleanup to return cluster to pre-test state, validating the machine counts throughout.

Done when:

• Test exists for the ImageModeStatusReporting feature gate to validate that MCP machine counts are correctly populated from the MCN on a rebootless node update targeting a default MCP in openshift/machine-config-operator/test/extended

References:

• Docs for creating and using a custom MCP can be found here.
• Information for creating and apply MCs can be found here.
• Information on node disruption policies can be found here.
• A bug highlighting issues with the previous implementation of this idea that can be used to test for problem edge cases is OCPBUGS-32812.

Depends on:

• This test will need to be tagged as `Disruptive` since it will take longer than 5 minutes to complete. Thus, completing this story depends on the MCO team having a test suite where running disruptive tests is allowed (see MCO-1679 & MCO-1652).
• This tests functionality implemented as part of MCO-1228, so that story will need to be completed prior to implementing this test.
• These tests should contribute to the StatusReporting feature gate’s component readiness score. Therefore, this test should be created after the StatusReporting feature gate exists (see MCO-1681).