Cluster controller first iteration #1688

DmytroPI-dev · 2026-02-05T11:33:55Z

Description

Cluster controller for CNPG first iteration

Key Changes

Drafted API and controller for Cluster controller

Testing and Verification

Manual tests only

Related Issues

JIRA: CPI-1883

PR Checklist

Code changes adhere to the project's coding standards.
Relevant unit and integration tests are included.
Documentation has been updated accordingly.
All tests pass locally.
The PR description follows the project's guidelines.

github-actions · 2026-02-05T11:34:06Z

CLA Assistant Lite bot All contributors have signed the CLA ✍️ ✅

api/v4/cluster_types.go

config/samples/enterprise_v4_postgrescluster_default.yaml

internal/controller/postgrescluster_controller_test.go

internal/controller/database_cluster_controller.go

internal/controller/cluster_controller.go

api/v4/cluster_types.go

internal/controller/cluster_controller.go

internal/controller/postgrescluster_controller.go

internal/controller/cluster_controller.go

DmytroPI-dev · 2026-02-10T12:04:52Z

I have read the CLA Document and I hereby sign the CLA

DmytroPI-dev · 2026-02-10T12:09:11Z

recheck

config/rbac/kustomization.yaml

config/rbac/role.yaml

internal/controller/postgrescluster_controller.go

internal/controller/database_cluster_controller.go

internal/controller/postgrescluster_controller.go

PROJECT

internal/controller/postgrescluster_controller.go

mploski · 2026-02-12T07:14:24Z

internal/controller/postgrescluster_controller.go

+// syncStatus is a helper function to map CNPG Cluster state to PostgresCluster object and commit it to API.
+func (r *PostgresClusterReconciler) syncStatus(ctx context.Context, postgresCluster *enterprisev4.PostgresCluster, cnpgCluster *cnpgv1.Cluster, err error) error {
+	// will use Patch as we did for main reconciliation loop.
+	latestPGCluster := client.MergeFrom(postgresCluster.DeepCopy())


i dont think we need patching for setting status, simple update is good enough

I was using Update initially, but it causes error as below:

"Operation cannot be fulfilled on postgresclusters.enterprise.splunk.com \"postgresql-cluster-dev-overridden\": the object has been modified; please apply your changes to the latest version and try again".

As this was the very same error I noticed earlier, when decided to switch from Update to Patch in Reconcile method, I decided to use Patch here as well. If the above error is not critical, we can use Update, of course, but I assume that any error logs will be confusing, though it resolves automatically in next reconciliation loop.

internal/controller/postgrescluster_controller.go

mploski · 2026-02-12T07:16:09Z

internal/controller/postgrescluster_controller.go

+
+	// If CNPG is missing, or in a transitional state (not Ready/Failed), we'll keep checking.
+	if postgresCluster.Status.Phase == "Pending" || postgresCluster.Status.Phase == "Provisioning" {
+		return ctrl.Result{RequeueAfter: retryDelay}, nil


same here, not sure why it is here

internal/controller/postgrescluster_controller.go

DmytroPI-dev requested review from limak9182 and mploski February 5, 2026 11:36

limak9182 reviewed Feb 5, 2026

View reviewed changes

api/v4/cluster_types.go Outdated Show resolved Hide resolved

mploski reviewed Feb 5, 2026

View reviewed changes

api/v4/cluster_types.go Outdated Show resolved Hide resolved

mploski reviewed Feb 5, 2026

View reviewed changes

api/v4/cluster_types.go Outdated Show resolved Hide resolved

mploski reviewed Feb 5, 2026

View reviewed changes

api/v4/cluster_types.go Outdated Show resolved Hide resolved