abiosoft/colima · error

error starting %s: %w

Error message

error starting %s: %w

What it means

The second phase of `colima kubernetes reset`: teardown and Provision succeeded, but k.Start (booting the k3s cluster) failed. The VM itself is healthy; only cluster startup failed. Common causes surface in the wrapped error: k3s image pull failures, guest CPU/memory pressure, or control-plane port (6443) conflicts.

Source

Thrown at cmd/kubernetes.go:124

	Args: cobra.NoArgs,
	RunE: func(cmd *cobra.Command, args []string) error {
		app := newApp()
		k, err := app.Kubernetes()
		if err != nil {
			return err
		}

		if err := k.Teardown(context.Background()); err != nil {
			return fmt.Errorf("error deleting %s: %w", kubernetes.Name, err)
		}

		ctx := context.Background()
		if err := k.Provision(ctx); err != nil {
			return err
		}

		if err := k.Start(ctx); err != nil {
			return fmt.Errorf("error starting %s: %w", kubernetes.Name, err)
		}

		return nil
	},
}

func init() {
	root.Cmd().AddCommand(kubernetesCmd)
	kubernetesCmd.AddCommand(kubernetesStartCmd)
	kubernetesCmd.AddCommand(kubernetesStopCmd)
	kubernetesCmd.AddCommand(kubernetesDeleteCmd)
	kubernetesCmd.AddCommand(kubernetesResetCmd)
}

View on GitHub (pinned to c3a5f9184d)

Solutions

1. Retry just the start phase — images are now cached, so `colima kubernetes start` is cheap and often succeeds
2. If pulls failed, fix guest networking (proxy env, DNS) and retry; check the wrapped error and colima logs
3. Recreate the VM with more headroom: `colima stop && colima delete && colima start --kubernetes --cpu 4 --memory 6`

# cheap retry: images are cached after reset, so just start again
colima kubernetes start

if err := resetCmd.Execute(); err != nil {
    if strings.Contains(err.Error(), "error starting Kubernetes") {
        // teardown+provision succeeded; only the cluster boot failed -> a plain start retry is safe
        _ = startCmd.Execute()
    }
}

When it happens

Trigger: k.Start failing after successful re-provision: k3s images cannot be pulled (network/proxy/DNS issues inside the guest); insufficient VM cpu/memory for the control plane; k3s in the guest crash-looping after the reset; leftover port binding from the pre-reset cluster.

Common situations: Corporate proxy/VPN blocking registry pulls during reset; VM created with small `colima start --cpu 1 --memory 2` defaults; flaky guest networking after host sleep/reset cycles.

Related errors

• error deleting %s: %w
• %s is not running
• %s is not enabled
• error in config file: %w
• error starting vm: %w

AI-assisted analysis of abiosoft/colima@c3a5f9184d (2026-08-15). Data as JSON: /api/errors/c1c4ecdeb390c8bd. Report an issue: GitHub.