GoogleContainerTools/skaffold · error

getting Kubernetes client: %w

Error message

getting Kubernetes client: %w

What it means

Task.jobsManager obtains a Kubernetes clientset (kubernetesclient.Client for the task's kubecontext) to get a Job interface. If client construction fails — bad kubeconfig, unreadable context, or client init error — it returns 'getting Kubernetes client'. Callers are Exec and Cleanup, so this can hit during both run and teardown.

Source

Thrown at pkg/skaffold/actions/k8sjob/task.go:149

			Name:  e.Name,
			Value: e.Value,
		})
	}
}

func (t Task) Cleanup(ctx context.Context, out io.Writer) error {
	jm, err := t.jobsManager()
	if err != nil {
		return err
	}

	return k8sjobutil.ForceJobDelete(ctx, t.Name(), jm, t.kubectl)
}

func (t Task) jobsManager() (typesbatchv1.JobInterface, error) {
	clientset, err := kubernetesclient.Client(t.kubectl.KubeContext)
	if err != nil {
		return nil, fmt.Errorf("getting Kubernetes client: %w", err)
	}

	return clientset.BatchV1().Jobs(t.jobManifest.Namespace), nil
}

func (t Task) getContainerToDeploy() corev1.Container {
	return corev1.Container{
		Name:    t.cfg.Name,
		Image:   t.artifact.Tag,
		Command: t.cfg.Command,
		Args:    t.cfg.Args,
		Env:     append(t.getK8SEnvVars(t.cfg.Env), t.envVars...),
	}
}

func (t Task) getK8SEnvVars(envVars []latest.VerifyEnvVar) (k8sEnvVar []corev1.EnvVar) {
	for _, envVar := range envVars {
		k8sEnvVar = append(k8sEnvVar, corev1.EnvVar{Name: envVar.Name, Value: envVar.Value})

View on GitHub (pinned to a1189de023)

Solutions

1. Validate the kubeconfig: kubectl config current-context and kubectl auth can-i get jobs --context <ctx> to ensure the context resolves and is authorized.
2. Fix KUBECONFIG / kubeContext setting in skaffold.yaml to point at a valid config file containing the context.
3. Regenerate credentials for the context (e.g. gcloud container clusters get-credentials) if the kubeconfig is stale.

Example fix

// before
echo $KUBECONFIG  # /nonexistent/config
// after
export KUBECONFIG=$HOME/.kube/config
kubectl config current-context  # sanity check, then rerun skaffold

// Verify kubeconfig resolves the context before running k8sjob actions
[ -f "${KUBECONFIG:-$HOME/.kube/config}" ] && kubectl config get-contexts "<ctx>" > /dev/null

try {
  await skaffold.run(...);
} catch (e) {
  if (/getting Kubernetes client/.test(String(e))) {
    console.error('Fix KUBECONFIG / kubeContext; ensure the context exists');
  }
  throw e;
}

When it happens

Trigger: Executing or cleaning up a k8sjob action where kubernetesclient.Client(kubeContext) fails: missing/invalid kubeconfig file, unknown context name, malformed kubeconfig YAML, or client-go construction error.

Common situations: KUBECONFIG env var pointing to a nonexistent file; context renamed after cluster deletion; kubeconfig with wrong apiVersion after manual editing; HOME not set in CI so default kubeconfig path unresolvable.

Related errors

• error getting Kubernetes dynamic client: %w
• error getting Kubernetes client: %w
• getting Kubernetes client: %w
• getting current cluster context: %w
• getting current namespace: %w

AI-assisted analysis of GoogleContainerTools/skaffold@a1189de023 (2026-09-05). Data as JSON: /api/errors/88543034640b46bd. Report an issue: GitHub.