abiosoft/colima · error

error stopping vm: %w

Error message

error stopping vm: %w

What it means

The final and only fatal stop step failed: limactl could not stop the guest VM (c.guest.Stop). Unlike the runtime stop warnings, this error is returned, so `colima stop` exits non-zero and the VM may still be running or wedged.

Source

Thrown at app/app.go:214

			log := log.WithField("context", cont.Name())
			log.Println("stopping ...")

			if err := cont.Stop(ctx, force); err != nil {
				// failure to stop a container runtime is not fatal
				// it is only meant for graceful shutdown.
				// the VM will shut down anyways.
				log.Warnln(fmt.Errorf("error stopping %s: %w", cont.Name(), err))
			}
		}
	} else if c.guest.Running(ctx) {
		log.Warnln("vm not responsive, skipping runtime shutdown")
	}

	// stop vm
	// no need to check running status, it may be in a state that requires stopping.
	if err := c.guest.Stop(ctx, force); err != nil {
		return fmt.Errorf("error stopping vm: %w", err)
	}

	log.Println("done")

	if err := generateSSHConfig(false); err != nil {
		log.Trace("error generating ssh_config: %w", err)
	}
	return nil
}

func (c colimaApp) Delete(data, force bool) error {
	confirmContainerDestruction := func() bool {
		return cli.Prompt("\033[31m\033[1mthis will delete ALL container data. Are you sure you want to continue")
	}

	s, _ := store.Load()
	diskInUse := s.DiskFormatted

View on GitHub (pinned to c3a5f9184d)

Solutions

1. Escalate to `colima stop --force` which kills the VM without graceful shutdown
2. Drive lima directly: `limactl list` then `limactl stop colima`, reading ~/.lima/colima logs for the cause
3. Kill a wedged hypervisor process (qemu-system-*) if it persists, then `colima stop --force`
4. If the instance is unrecoverable: `colima delete --force && colima start`

Example fix

# before
colima stop
# error: error stopping vm: ...

# after
colima stop --force

// confirm the instance is not wedged before stop
out, _ := exec.Command("limactl", "list", "--json").CombinedOutput()
// if the profile shows an invalid/running-though-crashed state, go straight to --force

if err := a.Stop(false); err != nil {
    if strings.Contains(err.Error(), "error stopping vm") {
        return a.Stop(true) // escalate to force stop
    }
}

When it happens

Trigger: limactl stop fails: QEMU/VZ process wedged after host sleep or crash, guest unreachable so lima cannot issue the shutdown, or the lima instance state file is inconsistent.

Common situations: Mac woke from sleep with the VM frozen; colima/lima upgraded while the old VM was running; VM stuck after a forced host reboot.

Related errors

• error starting vm: %w
• dependency check failed for VM: %w
• error stopping %s: %w
• error during teardown of vm: %w
• error stopping :%w

AI-assisted analysis of abiosoft/colima@c3a5f9184d (2026-08-15). Data as JSON: /api/errors/2ffdbc629c736ce3. Report an issue: GitHub.