lima-vm/lima · critical

daemon %#q is not available

Error message

daemon %#q is not available

What it means

Config.StartCmd builds the shell command that launches the network daemon. If IsDaemonInstalled reports the daemon is missing, it panics with 'daemon %#q is not available' because StartCmd returns only a string (no error) and cannot proceed sensibly. Unlike User(), this is a panic, not a returned error.

Source

Thrown at pkg/networks/commands.go:100

	if ok, _ := c.IsDaemonInstalled(daemon); !ok {
		daemonPath, _ := c.DaemonPath(daemon)
		return osutil.User{}, fmt.Errorf("daemon %#q (path=%#q) is not available", daemon, daemonPath)
	}
	//nolint:gocritic // singleCaseSwitch: should rewrite switch statement to if statement
	switch daemon {
	case SocketVMNet:
		return osutil.LookupUser("root")
	}
	return osutil.User{}, fmt.Errorf("daemon %#q not defined", daemon)
}

func (c *Config) MkdirCmd() string {
	return fmt.Sprintf("/bin/mkdir -m 775 -p %s", c.Paths.VarRun)
}

func (c *Config) StartCmd(name, daemon string) string {
	if ok, _ := c.IsDaemonInstalled(daemon); !ok {
		panic(fmt.Errorf("daemon %#q is not available", daemon))
	}
	var cmd string
	switch daemon {
	case SocketVMNet:
		nw := c.Networks[name]
		if c.Paths.SocketVMNet == "" {
			panic("c.Paths.SocketVMNet is empty")
		}
		cmd = fmt.Sprintf("%s --pidfile=%s --socket-group=%s --vmnet-mode=%s",
			c.Paths.SocketVMNet, c.PIDFile(name, SocketVMNet), c.Group, nw.Mode)
		switch nw.Mode {
		case ModeBridged:
			cmd += fmt.Sprintf(" --vmnet-interface=%s", nw.Interface)
		case ModeHost, ModeShared:
			cmd += fmt.Sprintf(" --vmnet-gateway=%s --vmnet-dhcp-end=%s --vmnet-mask=%s",
				nw.Gateway, nw.DHCPEnd, nw.NetMask)
		}
		cmd += " " + c.Sock(name)

View on GitHub (pinned to dd909d0973)

Solutions

1. Install socket_vmnet and point paths.socketVMNet at the real binary
2. Call IsDaemonInstalled(daemon) and abort before invoking StartCmd
3. Prefer using the higher-level startDaemon flow that performs the check via User()

Example fix

// before
cmd := cfg.StartCmd(net, networks.SocketVMNet) // panics if missing
// after
if ok, _ := cfg.IsDaemonInstalled(networks.SocketVMNet); !ok {
    return fmt.Errorf("socket_vmnet not installed")
}
cmd := cfg.StartCmd(net, networks.SocketVMNet)

ok, _ := cfg.IsDaemonInstalled(networks.SocketVMNet)
if !ok {
    return fmt.Errorf("refusing to build start command: socket_vmnet is not installed")
}
cmd := cfg.StartCmd(name, networks.SocketVMNet)

// StartCmd panics rather than returning an error — guard with recover()
func safeStartCmd(cfg *networks.Config, name, daemon string) (cmd string, err error) {
    defer func() { if r := recover(); r != nil { err = fmt.Errorf("%v", r) } }()
    return cfg.StartCmd(name, daemon), nil
}

When it happens

Trigger: startDaemon calls StartCmd(name, daemon) when Paths.SocketVMNet is empty or the configured binary does not exist — the same precondition that makes User() return its error.

Common situations: socket_vmnet not installed; networks.yaml missing the paths.socketVMNet entry; calling StartCmd from custom tooling without first verifying installation via IsDaemonInstalled.

Related errors

• unexpected daemon %#q
• unknown daemon type %#q
• daemon %#q (path=%#q) is not available
• c.Paths.SocketVMNet is empty
• daemon %#q not defined

AI-assisted analysis of lima-vm/lima@dd909d0973 (2026-09-01). Data as JSON: /api/errors/a1b1d89219ff189b. Report an issue: GitHub.