| name | rhoai-uninstallation |
| metadata | {"author":"rhoai3-demo","version":"1.0.0","platform-family":"rhoai","platform-baseline":"repo","ocp-baseline":"repo","skill-group":"RHOAI Platform"} |
| description | Use when documenting, reviewing, planning, or executing Red Hat OpenShift AI Self-Managed uninstallation and decommissioning: Operator-managed resource removal, retained user resources, PVC backup requirements, the official CLI uninstall ConfigMap and addon-managed-odh-delete label workflow, deletion of redhat-ods-operator, verification of rhods-operator Subscription removal, namespace cleanup checks, GitOps/ArgoCD decommission sequencing, and safe distinction between shutdown and uninstall. Do NOT use for normal demo shutdown, GPU cost reduction, temporary resource scale-down, reinstall troubleshooting, or unrelated OpenShift namespace cleanup; use env-manage- resources, env-troubleshoot, or rhoai-installation-troubleshooting instead.
|
RHOAI Uninstallation
Use this skill to plan or review OpenShift AI Self-Managed uninstallation for
the active baseline in docs/PLATFORM_BASELINE.md.
Source Grounding
Read references/source-capture.md before using product uninstallation
details. Official Red Hat documentation is product authority. This skill adapts
the official CLI uninstall procedure to this repo's GitOps operating model and
live-cluster safety guard.
Hard Boundary
Uninstalling OpenShift AI is destructive and is not the same as shutting down
the demo to save cost.
Use:
env-manage-resources for stopping models or scaling GPU nodes down.
rhoai-uninstallation only when the user intentionally wants to decommission
or remove RHOAI Self-Managed from a cluster.
Do not run live uninstall commands unless the user explicitly asks for
uninstallation and confirms the target cluster.
Official Behavior Summary
The official Red Hat procedure recommends using the OpenShift CLI (oc) for
uninstallation because web console uninstall behavior can leave the final
cluster state unclear.
Uninstall removes Operator-created resources but retains user-created resources
to avoid accidental data loss.
Before uninstalling, persistent disks or volumes used by PVCs must be backed up.
GitOps Policy
For this repo:
- Disable, remove, or retarget ArgoCD Applications that manage RHOAI resources
before running the official uninstall flow, so ArgoCD does not recreate
resources while the Operator is deleting them.
- Record uninstall as an environment decommission operation, not as a normal
deployment step.
- Preserve enough GitOps state to redeploy into a fresh environment later.
- Do not delete retained user projects or CRDs automatically. Review them after
uninstall and delete only what the user explicitly approves.
- Do not commit raw backup archives, kubeconfigs, or sensitive uninstall logs.
Workflow
- Confirm the active baseline in
docs/PLATFORM_BASELINE.md.
- Confirm this is uninstall/decommissioning, not shutdown or scale-down.
- Read
references/official-doc-extraction.md.
- If a live cluster is involved, follow the OpenShift safety guard in
AGENTS.md.
- Confirm backup of persistent disks or PVC-backed data.
- Suspend or remove relevant ArgoCD Applications for RHOAI-managed resources.
- Run the official CLI uninstall sequence from
examples/official-cli-uninstall-sequence.md.
- Verify removed and retained resources with
references/validation-checklist.md.
- Review retained user resources and decide whether they should remain, be
exported, or be deleted.
References
references/source-capture.md
references/official-doc-extraction.md
references/validation-checklist.md
examples/official-cli-uninstall-sequence.md