9.3. Replacing Compute Nodes

If a Compute node fails, you can replace the node with a working one. Replacing a Compute node uses the following process:
  1. Migrate workload off the existing Compute node and shutdown the node. See Section 8.9, “Migrating VMs from an Overcloud Compute Node” for this process.
  2. Remove the Compute node from the Overcloud. See Section 9.2, “Removing Compute Nodes” for this process.
  3. Scale out the Overcloud with a new Compute node. See Chapter 9, Scaling and Replacing Nodes for this process.
This process ensures that a node can be replaced without affecting the availability of any instances.