shadowsocks/shadowsocks-rust · error

missing `manager_port`

Error message

missing `manager_port`

What it means

When the config is a manager-server config and no unix-domain-socket address is available (non-unix builds or absent socket path), the code requires an explicit TCP manager_port. If the address matches a host-only form and the port portion is missing on a non-unix platform, it returns ErrorKind::MissingField with 'missing `manager_port`'.

Source

Thrown at crates/shadowsocks-service/src/config.rs:2476

        }

        // Manager Address
        if let Some(ma) = config.manager_address {
            let manager = match config.manager_port {
                Some(port) => {
                    match ma.parse::<IpAddr>() {
                        Ok(ip) => ManagerAddr::from(SocketAddr::new(ip, port)),
                        Err(..) => {
                            // treated as domain
                            ManagerAddr::from((ma, port))
                        }
                    }
                }
                #[cfg(unix)]
                None => ManagerAddr::from(PathBuf::from(ma)),
                #[cfg(not(unix))]
                None => {
                    let e = Error::new(ErrorKind::MissingField, "missing `manager_port`", None);
                    return Err(e);
                }
            };

            let mut manager_config = ManagerConfig::new(manager);
            manager_config.mode = global_mode;

            if let Some(ref m) = config.method {
                match m.parse::<CipherKind>() {
                    Ok(method) => manager_config.method = Some(method),
                    Err(..) => {
                        let err = Error::new(
                            ErrorKind::Invalid,
                            "unsupported method",
                            Some(format!("`{m}` is not a supported method")),
                        );
                        return Err(err);
                    }

View on GitHub (pinned to 8eb0f0a65b)

Solutions

1. Add a `manager_port` field to the config
2. Include the port in the manager address, e.g. `127.0.0.1:8787`
3. Run on unix and use a unix domain socket path for manager_addr instead

Example fix

// before
{"manager_addr": "127.0.0.1"}
// after
{"manager_addr": "127.0.0.1", "manager_port": 8787}

fn validate_manager_cfg(cfg: &serde_json::Value) -> Result<(), String> {
    let addr = cfg["manager_addr"].as_str().unwrap_or("");
    let has_port = addr.rsplit(':').next().map_or(false, |p| p.parse::<u16>().is_ok());
    if !has_port && cfg.get("manager_port").is_none() {
        return Err("manager config requires manager_port on non-unix".into());
    }
    Ok(())
}

fn has_manager_port(cfg: &serde_json::Value) -> bool {
    cfg.get("manager_port").and_then(|v| v.as_u64()).map_or(false, |p| p > 0 && p <= 65535)
}

match load_manager_config() {
    Ok(c) => run(c),
    Err(e) if e.to_string().contains("missing `manager_port`") => {
        eprintln!("add manager_port to config: {e}");
    }
    Err(e) => return Err(e),
}

When it happens

Trigger: Loading a manager config whose manager address parses to a bare host without a port, on a non-unix target (no PathBuf fallback), so no usable ManagerAddr can be constructed.

Common situations: Writing `"manager_addr": "127.0.0.1"` without `"manager_port"` in the JSON config on Windows/macOS-less non-unix builds; copying a unix config that relies on a socket path onto a non-unix host.

Understand the failure class

Background: "is required", "must be set", "missing required field": configuration validation errors across open-source libraries — this error's family across 36 libraries.

Related errors

• missing `manager_addr` and `manager_port` in configuration
• missing `local_dns_addr` or `remote_dns_addr` in configurati
• missing `forward_addr` in configuration
• missing `local_port`
• `protocol` invalid

AI-assisted analysis of shadowsocks/shadowsocks-rust@8eb0f0a65b (2026-09-09). Data as JSON: /api/errors/b74daa304287dac4. Report an issue: GitHub.