derailed/k9s · warning

no matching pods for %v

Error message

no matching pods for %v

What it means

podFromSelector lists pods in a namespace with the given label selector and picks the first as the log/inspect target. When the list comes back empty, no pod backs the workload and this error names the selector that matched nothing.

Source

Thrown at internal/dao/svc.go:78

	err = runtime.DefaultUnstructuredConverter.FromUnstructured(o.(*unstructured.Unstructured).Object, &svc)
	if err != nil {
		return nil, errors.New("expecting Service resource")
	}

	return &svc, nil
}

// ----------------------------------------------------------------------------
// Helpers...

func podFromSelector(f Factory, ns string, sel map[string]string) (string, error) {
	oo, err := f.List(client.PodGVR, ns, true, labels.Set(sel).AsSelector())
	if err != nil {
		return "", err
	}

	if len(oo) == 0 {
		return "", fmt.Errorf("no matching pods for %v", sel)
	}

	var pod v1.Pod
	err = runtime.DefaultUnstructuredConverter.FromUnstructured(oo[0].(*unstructured.Unstructured).Object, &pod)
	if err != nil {
		return "", err
	}

	return client.FQN(pod.Namespace, pod.Name), nil
}

View on GitHub (pinned to 2d3ccc6ba2)

Solutions

1. Verify pods exist for the selector: kubectl get pods -n <ns> -l '<selector>'
2. Compare the Service/workload selector with actual pod labels: kubectl get pods --show-labels
3. Fix the label mismatch in the Service or workload template and re-apply
4. If the workload is scaled to 0, scale it up: kubectl scale deploy <name> --replicas=1

Example fix

# selector typo before
spec:
  selector:
    app: Api        # pods are labeled app: api
# after
spec:
  selector:
    app: api

pods, err := clientset.CoreV1().Pods(ns).List(ctx, metav1.ListOptions{LabelSelector: labels.Set(sel).AsSelector().String()})
if err != nil { return err }
if len(pods.Items) == 0 {
    return fmt.Errorf("no pods match %v in %s — scale up the workload or fix labels", sel, ns)
}

pod, err := svcDAO.Pod(fqn)
if err != nil && strings.Contains(err.Error(), "no matching pods") {
    // don't retry: the workload has 0 pods; tell the user which selector failed
    return fmt.Errorf("%w — verify with: kubectl get pods -n %s -l '%s'", err, ns, labels.Set(sel).AsSelector())
}

When it happens

Trigger: Calling Service.Pod or any helper that resolves a workload to a single pod when zero pods carry the selector's labels: the backing Deployment/StatefulSet is scaled to 0, crashed pods were evicted, labels in the Service selector don't match pod template labels (typo like app: Api vs app: api), or the pods live in another namespace.

Common situations: Scaled-to-zero preview environments; label drift after chart upgrades renaming labels; crashed CrashLoopBackOff pods that were garbage-collected; selectors copied from a different namespace.

Related errors

• no valid selector found on deployment: %s
• no valid selector found on daemonset %q
• no valid selector found on statefulset: %s
• no valid selector found on Service %s
• user is not authorized to list pod metrics

AI-assisted analysis of derailed/k9s@2d3ccc6ba2 (2026-08-15). Data as JSON: /api/errors/707c5e2b069573a5. Report an issue: GitHub.