Installation Manual - 111 spoke Pipelines operator batch

Batch 1 app-delivery foundation: spoke OpenShift Pipelines operator-only install, validation, and promotion decision.

This chapter records OP-GF-OPERATORS-21, the Batch 1 app-delivery foundation install for spoke-dc-v7.

Governance

Field	Value
Issue	`#437` / `OP-GF-OPERATORS-21`
Milestone	`Workspace Governance`
Governing ADRs	`ADR 0016`, `ADR 0024`
Predecessor	`#436` / `OP-GF-OPERATORS-20`
Platform GitOps commit	`8c41a86` / `Install spoke OpenShift Pipelines operator`
Blog commit	`b40c50c` / `Document spoke Pipelines operator batch`

Scope

Field	Value
Target	`spoke-dc-v7`
Package	`openshift-pipelines-operator-rh`
Channel	`pipelines-1.22`
CSV	`openshift-pipelines-operator-rh.v1.22.0`
CatalogSource	`cs-redhat-operator-index-v4-20`

The batch installed the operator through GitOps and stopped at operator readiness. It did not create app namespaces, CI/CD credentials, build workloads, PipelineRuns, TaskRuns, TriggerBindings, TriggerTemplates, or EventListeners.

Preflight

Preflight used the standard v7 access path:

local workspace -> dl385-2 -> gf-ocp-bootstrap-01 -> spoke-dc-v7 kubeconfig

Readiness checks passed before install:

Check	Result
OpenShift	`4.20.18`
ClusterVersion	Available, not Progressing, not Failing
Nodes	`3` masters and `3` workers Ready
ClusterOperators	no non-steady rows
Pending CSRs	none
Existing Pipelines state	no namespace, Subscription, CSV, or Tekton API resources
OADP DPA	`spoke-dc-v7`, `Reconciled=True`
OADP BSL	`spoke-dc-v7`, `Available`
OADP schedule	`platform-resource-daily`, enabled

The PackageManifest resolved from the mirrored Red Hat catalog. Related image source prefixes were mapped through ImageDigestMirrorSet/idms-openshift-pipelines.

GitOps Install

Platform GitOps commit 8c41a86 added the Pipelines operator overlay and mirror policy:

clusters/spoke-dc-v7/operators/openshift-pipelines-operator/
clusters/spoke-dc-v7/platform/image-mirrors/idms-openshift-pipelines.yaml

The Subscription lives in openshift-operators, uses cs-redhat-operator-index-v4-20, channel pipelines-1.22, and starting CSV openshift-pipelines-operator-rh.v1.22.0.

Validation

Final validation:

Check	Result
Argo	hub-side and spoke-side views `Healthy/Synced` at `8c41a86ebc59390f0be4aae765dfb17276a11602`
Subscription	`AtLatestKnown`
CSV	`openshift-pipelines-operator-rh.v1.22.0`, `Succeeded`
TektonConfig	`config`, version `1.22.0`, `Ready=True`
Operator pods	Running
`openshift-pipelines` pods	Running
ClusterVersion	`4.20.18`, Available, not Progressing, not Failing
Non-running pods	none
Pending CSRs	none

The operator created default operator-owned Tekton add-ons. At closeout there were still no workload runs or trigger entry points:

pipelineruns.tekton.dev: 0
taskruns.tekton.dev: 0
triggerbindings.triggers.tekton.dev: 0
triggertemplates.triggers.tekton.dev: 0
eventlisteners.triggers.tekton.dev: 0

Promotion Decision

Keep the install. GitOps was synced, OLM was AtLatestKnown, the CSV was Succeeded, TektonConfig was Ready, operator pods were Running, and spoke cluster health stayed steady.

Rollback is GitOps-first:

Remove the Pipelines operator overlay from the spoke kustomization.
Remove idms-openshift-pipelines only if no other workload needs those mirror mappings.
Sync and validate OLM/operator pruning.
Do not force-delete Tekton CRDs or operator-created cluster-scoped objects without a separate cleanup issue.

Next completed operator batch: 112 Spoke observability foundation operators.