kubernetes/kops · error

error deleting %s configuration %q: %v

Error message

error deleting %s configuration %q: %v

What it means

delete() removes the configuration file for a named object from the VFS state store. NotExist is treated as success; any other Remove error is wrapped with this message including kind and name. It means the state-store could not perform the deletion.

Source

Thrown at pkg/client/simple/vfsclientset/commonvfs.go:212

		objectMeta.SetCreationTimestamp(metav1.NewTime(time.Now().UTC()))
	}

	err = c.writeConfig(ctx, cluster, c.basePath.Join(objectMeta.GetName()), i, vfs.WriteOptionOnlyIfExists)
	if err != nil {
		return fmt.Errorf("error writing %s: %v", c.kind, err)
	}

	return nil
}

func (c *VFSClientBase) delete(ctx context.Context, name string, options metav1.DeleteOptions) error {
	p := c.basePath.Join(name)
	err := p.Remove(ctx)
	if err != nil {
		if os.IsNotExist(err) {
			return nil
		}
		return fmt.Errorf("error deleting %s configuration %q: %v", c.kind, name, err)
	}
	return nil
}

func (c *VFSClientBase) listNames(ctx context.Context) ([]string, error) {
	keys, err := listChildNames(ctx, c.basePath)
	if err != nil {
		return nil, fmt.Errorf("error listing %s in state store: %v", c.kind, err)
	}

	// Seems to be an assumption in k8s APIs that items are always returned sorted
	sort.Strings(keys)

	return keys, nil
}

func (c *VFSClientBase) readAll(ctx context.Context, items interface{}) (interface{}, error) {
	sliceValue := reflect.ValueOf(items)

View on GitHub (pinned to 4c8573c808)

Solutions

1. Inspect the wrapped %v error and fix backend permissions or connectivity.
2. Check IAM policy allows DeleteObject on the state bucket.
3. Retry after transient errors (throttling, network).
4. Verify the name/kind matches the actually stored object.

Example fix

// before
policy with only s3:GetObject,s3:PutObject
// after
policy actions ["s3:GetObject","s3:PutObject","s3:DeleteObject"] on arn:aws:s3:::my-state-bucket/*

names, err := listChildNames(ctx, basePath)
if err != nil {
    return fmt.Errorf("cannot list state store before delete: %w", err)
}
// confirm target exists before deleting

if err := client.Delete(ctx, name, metav1.DeleteOptions{}); err != nil {
    if strings.Contains(err.Error(), "error deleting") && isTransient(err) {
        return retryWithBackoff(func() error { return client.Delete(ctx, name, metav1.DeleteOptions{}) })
    }
    return err
} // NotExist is already treated as success by the library

When it happens

Trigger: DeleteAutoScalingGroup / ReleaseAddress / DeleteDhcpOptions / DeleteEgressOnlyInternetGateway / DeleteInternetGateway / DeleteKeyPair call delete() and vfs Remove fails with permission-denied, network error, or backend failure.

Common situations: Credentials lacking s3:DeleteObject, bucket versioning/retention policies blocking deletes, API throttling, network outage during teardown (`kops delete cluster`).

Related errors

• reading file %q: %v
• error reading full cluster spec for %q: %v
• reading kops-channels manifest %s: %w
• error parsing ConfigStore.Base %q: %v
• error reading state store: %v

AI-assisted analysis of kubernetes/kops@4c8573c808 (2026-09-05). Data as JSON: /api/errors/660a1acb9c4ac53b. Report an issue: GitHub.