GoogleContainerTools/skaffold · warning

kind %s is not supported

Error message

kind %s is not supported

What it means

ownerMetaObject resolves a Kubernetes resource's owner reference to its owning object by looking it up with the appropriate typed client. Only a fixed set of kinds (Deployment, ReplicaSet, Job, CronJob, StatefulSet, ReplicationController, Pod) is supported; any other owner Kind hits the default branch and returns this error. The caller (TopLevelOwnerKey) logs a warning and returns an empty owner key.

Source

Thrown at pkg/skaffold/kubernetes/owner.go:69

	}

	switch owner.Kind {
	case "Deployment":
		return client.AppsV1().Deployments(ns).Get(ctx, owner.Name, metav1.GetOptions{})
	case "ReplicaSet":
		return client.AppsV1().ReplicaSets(ns).Get(ctx, owner.Name, metav1.GetOptions{})
	case "Job":
		return client.BatchV1().Jobs(ns).Get(ctx, owner.Name, metav1.GetOptions{})
	case "CronJob":
		return client.BatchV1beta1().CronJobs(ns).Get(ctx, owner.Name, metav1.GetOptions{})
	case "StatefulSet":
		return client.AppsV1().StatefulSets(ns).Get(ctx, owner.Name, metav1.GetOptions{})
	case "ReplicationController":
		return client.CoreV1().ReplicationControllers(ns).Get(ctx, owner.Name, metav1.GetOptions{})
	case "Pod":
		return client.CoreV1().Pods(ns).Get(ctx, owner.Name, metav1.GetOptions{})
	default:
		return nil, fmt.Errorf("kind %s is not supported", owner.Kind)
	}
}

View on GitHub (pinned to a1189de023)

Solutions

1. Identify the unsupported kind from the error message and add a case for it in ownerMetaObject's switch with the appropriate typed client lookup
2. As a workaround, note the top-level owner key will be empty (TopLevelOwnerKey returns "") and the feature degrades gracefully
3. If a CRD controller creates the pods, trace ownership manually with kubectl get <kind> <name> -o jsonpath='{.metadata.ownerReferences}'

Example fix

// before
case "Pod":
	return client.CoreV1().Pods(ns).Get(ctx, owner.Name, metav1.GetOptions{})
default:
	return nil, fmt.Errorf("kind %s is not supported", owner.Kind)
// after
case "Pod":
	return client.CoreV1().Pods(ns).Get(ctx, owner.Name, metav1.GetOptions{})
case "DaemonSet":
	return client.AppsV1().DaemonSets(ns).Get(ctx, owner.Name, metav1.GetOptions{})
default:
	return nil, fmt.Errorf("kind %s is not supported", owner.Kind)

// check the owner kind before relying on TopLevelOwnerKey
supported := map[string]bool{"Deployment": true, "ReplicaSet": true, "Job": true, "CronJob": true, "StatefulSet": true, "ReplicationController": true, "Pod": true}
var ownersKnown = true
for _, or := range obj.GetOwnerReferences() {
	if !supported[or.Kind] { ownersKnown = false }
}

key := TopLevelOwnerKey(ctx, obj, kubeContext, kind)
if key == "" {
	// owner resolution failed (possibly unsupported kind); fall back to resource's own Kind-Name
	key = fmt.Sprintf("%s-%s", kind, obj.GetName())
}

When it happens

Trigger: Walking owner references of a resource whose OwnerReference.Kind is outside the supported list — e.g. DaemonSet, CustomResource (CRD), Service, ConfigMap-owned objects, or controller-generated kinds from operators — during TopLevelOwnerKey.

Common situations: Port-forwarding or debugging pods owned by a DaemonSet; resources owned by a custom operator/CRD; newer workload kinds (e.g. custom workload controllers) not in the hardcoded switch.

Related errors

• c.Message (pod status condition message)
• unable to lookup minikube executable. Please add it to PATH
• invalid local-registry-hosting ConfigMap
• strings.Join(errMsgs, "\n") (joined helm cleanup error messa
• rs.ae.Message (actionable error message from status check)

AI-assisted analysis of GoogleContainerTools/skaffold@a1189de023 (2026-09-05). Data as JSON: /api/errors/842b740f3cf02b90. Report an issue: GitHub.