What is a rolling update, and how does OpenShift perform upgrades without downtime?

• A. Rolling updates replace pods all at once.
• B. It creates new pods before deleting old ones, but only using daemon sets.
• C. It restarts old pods without creating new ones.
• D. It replaces pods incrementally, creating new pods before terminating old ones, according to the Deployment's strategy.

Answer: (D)

Rolling updates in OpenShift use the Deployment’s rolling update strategy. When you deploy a new version, OpenShift creates new pods with the updated version first and keeps the old pods running until the new ones are ready. As each batch of new pods passes readiness checks, traffic continues to be served by healthy pods, and old pods are terminated gradually. This is controlled by settings like maxSurge and maxUnavailable, which define how many pods can be added or taken down at once, ensuring the application remains available throughout the upgrade. In short, the process replaces pods incrementally by bringing up new ones before terminating old ones, following the Deployment’s strategy.