1
1
mirror of https://github.com/wez/wezterm.git synced 2024-11-24 07:46:59 +03:00
wezterm/docs/multiplexing.md
Wez Furlong 4238488a9b docs: suggest that people use default_gui_startup_args
instead of connect_automatically
2022-01-01 17:41:52 -07:00

8.0 KiB

Notice: multiplexing is still a young feature and is evolving rapidly. Your feedback is welcomed!

Multiplexing

The out-of-the-box experience with wezterm allows you to multiplex local tabs and windows which will persist until they are closed. With a little extra configuration you can enable local terminal multiplexing with features similar to those in tmux or screen.

Multiplexing in wezterm is based around the concept of multiplexing domains; a domain is a distinct set of windows and tabs. When wezterm starts up it creates a default local domain to manage the windows and tabs in the UI, but it can also be configured to start or connect to additional domains.

Once connected to a domain, wezterm can attach its windows and tabs to the local native UI, providing a more natural experience for interacting with the mouse, clipboard and scrollback features of the terminal.

Key bindings allow you to spawn new tabs in the default local domain, the domain of the current tab, or a specific numbered domain.

SSH Domains

wezterm also supports regular ad-hoc ssh connections. This section of the docs refers to running a wezterm daemon on the remote end of a multiplexing session that uses ssh as a channel

A connection to a remote wezterm multiplexer made via an ssh connection is referred to as an SSH domain. A compatible version of wezterm must be installed on the remote system in order to use SSH domains. SSH domains are supported on all systems via libssh2.

To configure an SSH domain, place something like the following in your .wezterm.lua file:

return {
  ssh_domains = {
    {
      -- This name identifies the domain
      name = "my.server",
      -- The address to connect to
      remote_address = "192.168.1.1",
      -- The username to use on the remote host
      username = "wez",
    }
  }
}

See SshDomain for more information on possible settings to use with SSH domains.

To connect to the system, run:

$ wezterm connect my.server

This will launch an SSH session that connects to the specified address and may pop up authentication dialogs (using SSH keys for auth is strongly recommended!). Once connected, it will attempt to spawn the wezterm multiplexer daemon on the remote host and connect to it via a unix domain socket using a similar mechanism to that described in the Unix Domains section below.

Unix Domains

A connection to a multiplexer made via a unix socket is referred to as a unix domain. Unix domains are supported on all systems, even Windows and are a way to connect the native win32 GUI into the Windows Subsystem for Linux (WSL).

The bare minimum configuration to enable a unix domain is this, which will spawn a server if needed and then connect the gui to it automatically when wezterm is launched:

return {
  unix_domains = {
    {
      name = "unix",
    }
  },

  -- This causes `wezterm` to act as though it was started as
  -- `wezterm connect unix` by default, connecting to the unix
  -- domain on startup.
  -- If you prefer to connect manually, leave out this line.
  default_gui_startup_args = {"connect", "unix"},
}

If you prefer to connect manually, omit the default_gui_startup_args setting and then run:

$ wezterm connect unix

Note that in earlier versions of wezterm, a connect_automatically domain option was shown as the way to connect on startup. Using default_gui_startup_args is recommended instead as it works more reliably.

The possible configuration values are:

return {
  unix_domains = {
    {
      -- The name; must be unique amongst all domains
      name = "unix",

      -- The path to the socket.  If unspecified, a resonable default
      -- value will be computed.

      -- socket_path = "/some/path",

      -- If true, do not attempt to start this server if we try and fail to
      -- connect to it.

      -- no_serve_automatically = false,

      -- If true, bypass checking for secure ownership of the
      -- socket_path.  This is not recommended on a multi-user
      -- system, but is useful for example when running the
      -- server inside a WSL container but with the socket
      -- on the host NTFS volume.

      -- skip_permissions_check = false,

    }
  }
}

Since: 20220101-133340-7edc5b5a

It is now possible to specify a proxy_command that will be used in place of making a direct unix connection. When proxy_command is specified, it will be used instead of the optional socket_path.

This example shows a redundant use of nc (netcat) to connect to the unix socket path on my mac. This isn't useful on its own, but may help with the WSL 2 issue mentioned below when translated to an appropriate invocation of netcat/socat on Windows:

return {
  unix_domains = {
    {
      name = "unix",
      proxy_command = {"nc", "-U", "/Users/wez/.local/share/wezterm/sock"},
    }
  }
}

Connecting into Windows Subsystem for Linux

Note: this only works with WSL 1. WSL 2 doesn't support AF_UNIX interop

Inside your WSL instance, configure .wezterm.lua with this snippet:

return {
  unix_domains = {
    {
      name = "wsl"
      -- Override the default path to match the default on the host win32
      -- filesystem.  This will allow the host to connect into the WSL
      -- container.
      socket_path = "/mnt/c/Users/USERNAME/.local/share/wezterm/sock",
      -- NTFS permissions will always be "wrong", so skip that check
      skip_permissions_check = true,
    }
  }
}

In the host win32 configuration, use this snippet:

return {
  unix_domains = {
    {
      name = "wsl",
      serve_command = ["wsl", "wezterm-mux-server", "--daemonize"],
    }
  },
  default_gui_startup_args = {"connect", "wsl"},
}

Now when you start wezterm you'll be presented with a WSL tab.

You can also omit default_gui_startup_args and use:

$ wezterm connect wsl

to manually connect into your WSL instance.

TLS Domains

A connection to a multiplexer made via a TLS encrypted TCP connection is referred to as a TLS Domain.

Starting with version 20200202-180558-2489abf9, wezterm can bootstrap a TLS session by performing an initial connection via SSH to start the wezterm multiplexer on the remote host and securely obtain a key. Once bootstrapped, the client will use a TLS protected TCP connection to communicate with the server.

Configuring the client

For each server that you wish to connect to, add a client section like this:

return {
  tls_clients = {
    {
      -- A handy alias for this session; you will use `wezterm connect server.name`
      -- to connect to it.
      name = "server.name",
      -- The host:port for the remote host
      remote_address = "server.hostname:8080",
      -- The value can be "user@host:port"; it accepts the same syntax as the
      -- `wezterm ssh` subcommand.
      bootstrap_via_ssh = "server.hostname",
    }
  }
}

See TlsDomainClient for more information on possible settings.

Configuring the server

return {
  tls_servers = {
    {
      -- The host:port combination on which the server will listen
      -- for connections
      bind_address = "server.hostname:8080"
    }
  }
}

See TlsDomainServer for more information on possible settings.

Connecting

On the client, running this will connect to the server, start up the multiplexer and obtain a certificate for the TLS connection. A connection window will show the progress and may prompt you for SSH authentication. Once the connection has been initiated, wezterm will automatically reconnect using the certificate it obtained during bootstrapping if your connection was interrupted and resume your remote terminal session

$ wezterm connect server.name