Easily expand your on-premises OpenShift cluster
summary
New workflow uses the oc CLI to bake a join artifact: oc adm node-image create introspects the cluster and emits an ISO (or PXE artifacts) — currently requires a MAC address flag but accepts NMState files or a YAML list for many nodes so one ISO can provision hundreds. Booting a node runs prevalidation, writes the disk, reboots and reports events via node-image monitor (point at the node IP). Final step is manual CSR handling with certificate approve. Works on OpenShift/OKD 4.17+, workers only, suitable for disconnected UPI environments.
AI-written summary · about
chapters (11)
- 0:00Expanding OpenShift clusters
- 1:07Manual node addition problems
- 4:23Why adding a node is difficult
- 8:07Design pillars for new approach
- 11:19Three-step workflow
- 14:59Live demo: node addition
- 20:55Customizing nodes and multiples
- 23:24Alternative boot methods and scalability
- 24:49Summary and limitations
- 26:17Q&A: CSR approval and NTP errors
- 30:53Q&A: Certificate requests and security
related talks
19mOpenShift in your own backyard
01h 49mHands-on Intro to Kubernetes (and OpenShift) for JS Developers
32mOptimizing your Hybrid Cloud Operating System Experience
01h 45mHands On Intro to Kubernetes & OpenShift for JS Hackers
33mBootable Containers: An entire OS as a Containerfile
05mBare Metal Provisioning With Tinkerbell
05mORAS: Create and Distribute a Multi-platform Image with Secu...
45mOpenstack-on-OpenStack Overview