kubernetes/kops · error

error trying to locate asset

Error message

error trying to locate asset %q: %v

What it means

KubectlBuilder.Build installs kubectl on API-server nodes by looking up the "kubectl" asset through b.Assets.Find(assetName, assetPath). When the asset resolution mechanism itself fails (network/download error, hash mismatch, malformed assets config), the error is wrapped as "error trying to locate asset %q: %v". Note this is distinct from the asset simply being absent, which produces "unable to locate asset" on the next line.

Solutions

1. Verify network access from the node to the kops asset location (KOPS_BASE_URL / asset mirror) and fix connectivity or proxy settings
2. Set a correct, reachable base URL (e.g. KOPS_BASE_URL or --kops-base-url) matching your kops version
3. Use a custom asset configuration (kops set cluster cluster.spec.assets) pointing at an internally hosted kubectl binary with the correct hash
4. Upgrade/downgrade kops so a published kubectl artifact exists for the node OS/arch

Example fix

# before: node cannot reach the default mirror
export KOPS_BASE_URL=https://unreachable.internal/kops
# after: point at a reachable mirror hosting the matching kubectl asset
export KOPS_BASE_URL=https://s3.amazonaws.com/kops-releases/latest
# or pre-pin the asset
kops set cluster mycluster cluster.spec.assets.nodeUrlAmd64=https://internalMirror/kubectl

// Verify the kubectl asset is reachable before bootstrapping
baseURL := os.Getenv("KOPS_BASE_URL")
url := baseURL + "/linux/amd64/kubectl"
resp, err := http.Head(url)
if err != nil || resp.StatusCode != http.StatusOK {
	return fmt.Errorf("kubectl asset unreachable at %s: %v", url, err)
}

if err := runNodeup(cfg); err != nil {
	if strings.Contains(err.Error(), "error trying to locate asset") {
		log.Errorf("asset resolution failed; check KOPS_BASE_URL/mirror reachability and assets config: %v", err)
		return fmt.Errorf("bootstrap blocked by asset lookup: %w", err)
	}
	return err
}

When it happens

Trigger: Running nodeup on a node with an API server where kops' Assets.Find for name="kubectl" returns a non-nil error — e.g. failure resolving the kubectl download URL, an unreachable/misconfigured KOPS_BASE_URL or assets location, or an error computing the asset hash.

Common situations: Air-gapped or firewalled clusters where the kops asset mirror/base URL is unreachable; broken KOPS_BASE_URL or custom assetBundles config; kops version with no published kubectl artifact for the target GOOS/GOARCH; proxy/TLS problems on the node.

Related errors

• failed to get node config from server
• reading kops-channels manifest
• building nodeConfig for instanceGroup
• cannot parse ConfigBase
• certificate not found

AI-assisted analysis of kubernetes/kops@4c8573c808 (2026-09-05). Data as JSON: /api/errors/2708f32df06ce078. Report an issue: GitHub.

Appendix: source

Thrown at nodeup/pkg/model/kubectl.go:48

type KubectlBuilder struct {
	*NodeupModelContext
}

var _ fi.NodeupModelBuilder = &KubectlBuilder{}

// Build is responsible for managing the kubectl on the nodes
func (b *KubectlBuilder) Build(c *fi.NodeupModelBuilderContext) error {
	if !b.HasAPIServer {
		return nil
	}

	{
		// TODO: Extract to common function?
		assetName := "kubectl"
		assetPath := ""
		asset, err := b.Assets.Find(assetName, assetPath)
		if err != nil {
			return fmt.Errorf("error trying to locate asset %q: %v", assetName, err)
		}
		if asset == nil {
			return fmt.Errorf("unable to locate asset %q", assetName)
		}

		c.AddTask(&nodetasks.File{
			Path:     b.KubectlPath() + "/" + assetName,
			Contents: asset,
			Type:     nodetasks.FileType_File,
			Mode:     s("0755"),
		})
	}

	{
		name := nodetasks.PKIXName{
			CommonName:   "kubecfg",
			Organization: []string{rbac.SystemPrivilegedGroup},
		}

View on GitHub (pinned to 4c8573c808)