ControlPlaneMachineSet Fails to Evenly Distribute Master Nodes Across Availability Zones on IPI RHOCP4

Solution Verified - Updated -

Issue

  • When using the ControlPlaneMachineSet to modify the resources of master nodes in an RHOCP IPI cluster on AWS, the nodes are expected to be evenly distributed across all defined availability zones.

However, an uneven distribution was observed despite correctly configured failure domains. For example:

Expected distribution:

  • 2 master nodes – ap-southeast-1a
  • 1 master node – ap-southeast-1b
  • 0 master nodes – ap-southeast-1c

Actual distribution:

  • 2 master nodes – ap-southeast-1a
  • 1 master node – ap-southeast-1b
  • 0 master nodes – ap-southeast-1b

This causes failure in achieving high availability across AZs.

Environment

  • Red Hat OpenShift Container Platform (RHOCP) 4.12+
  • ControlPlaneMachineSet(CPMS) enabled

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content