File

net/server.lua @ 13652:a08065207ef0

net.server_epoll: Call :shutdown() on TLS sockets when supported Comment from Matthew: This fixes a potential issue where the Prosody process gets blocked on sockets waiting for them to close. Unlike non-TLS sockets, closing a TLS socket sends layer 7 data, and this can cause problems for sockets which are in the process of being cleaned up. This depends on LuaSec changes which are not yet upstream. From Martijn's original email: So first my analysis of luasec. in ssl.c the socket is put into blocking mode right before calling SSL_shutdown() inside meth_destroy(). My best guess to why this is is because meth_destroy is linked to the __close and __gc methods, which can't exactly be called multiple times and luasec does want to make sure that a tls session is shutdown as clean as possible. I can't say I disagree with this reasoning and don't want to change this behaviour. My solution to this without changing the current behaviour is to introduce a shutdown() method. I am aware that this overlaps in a conflicting way with tcp's shutdown method, but it stays close to the OpenSSL name. This method calls SSL_shutdown() in the current (non)blocking mode of the underlying socket and returns a boolean whether or not the shutdown is completed (matching SSL_shutdown()'s 0 or 1 return values), and returns the familiar ssl_ioerror() strings on error with a false for completion. This error can then be used to determine if we have wantread/wantwrite to finalize things. Once meth_shutdown() has been called once a shutdown flag will be set, which indicates to meth_destroy() that the SSL_shutdown() has been handled by the application and it shouldn't be needed to set the socket to blocking mode. I've left the SSL_shutdown() call in the LSEC_STATE_CONNECTED to prevent TOCTOU if the application reaches a timeout for the shutdown code, which might allow SSL_shutdown() to clean up anyway at the last possible moment. Another thing I've changed to luasec is the call to socket_setblocking() right before calling close(2) in socket_destroy() in usocket.c. According to the latest POSIX[0]: Note that the requirement for close() on a socket to block for up to the current linger interval is not conditional on the O_NONBLOCK setting. Which I read to mean that removing O_NONBLOCK on the socket before close doesn't impact the behaviour and only causes noise in system call tracers. I didn't touch the windows bits of this, since I don't do windows. For the prosody side of things I've made the TLS shutdown bits resemble interface:onwritable(), and put it under a combined guard of self._tls and self.conn.shutdown. The self._tls bit is there to prevent getting stuck on this condition, and self.conn.shutdown is there to prevent the code being called by instances where the patched luasec isn't deployed. The destroy() method can be called from various places and is read by me as the "we give up" error path. To accommodate for these unexpected entrypoints I've added a single call to self.conn:shutdown() to prevent the socket being put into blocking mode. I have no expectations that there is any other use here. Same as previous, the self.conn.shutdown check is there to make sure it's not called on unpatched luasec deployments and self._tls is there to make sure we don't call shutdown() on tcp sockets. I wouldn't recommend logging of the conn:shutdown() error inside close(), since a lot of clients simply close the connection before SSL_shutdown() is done.
author Martijn van Duren <martijn@openbsd.org>
date Thu, 06 Feb 2025 15:04:38 +0000
parent 13453:0b48cf880e79
line wrap: on
line source

-- Prosody IM
-- Copyright (C) 2008-2010 Matthew Wild
-- Copyright (C) 2008-2010 Waqas Hussain
--
-- This project is MIT/X11 licensed. Please see the
-- COPYING file in the source package for more information.
--

local function log(level, format, ...)
	print("net.server", level, format:format(...));
end

local default_backend = "select";
local server_type = default_backend;

if (prosody and prosody.config_loaded) then
	default_backend = "epoll";
	log = require"prosody.util.logger".init("net.server");
	server_type = require"prosody.core.configmanager".get("*", "network_backend") or default_backend;

	if require"prosody.core.configmanager".get("*", "use_libevent") then
		server_type = "event";
	end
elseif pcall(require, "prosody.util.poll") then
	server_type = "epoll";
end

if server_type == "event" then
	if not pcall(require, "luaevent.core") then
		log("error", "libevent not found, falling back to %s", default_backend);
		server_type = default_backend;
	end
end

local server;
local set_config;
if server_type == "event" then
	server = require "prosody.net.server_event";

	local defaults = {};
	for k,v in pairs(server.cfg) do
		defaults[k] = v;
	end
	function set_config(settings)
		local event_settings = {
			ACCEPT_DELAY = settings.accept_retry_interval;
			ACCEPT_QUEUE = settings.tcp_backlog;
			CLEAR_DELAY = settings.event_clear_interval;
			CONNECT_TIMEOUT = settings.connect_timeout;
			DEBUG = settings.debug;
			HANDSHAKE_TIMEOUT = settings.ssl_handshake_timeout;
			MAX_CONNECTIONS = settings.max_connections;
			MAX_HANDSHAKE_ATTEMPTS = settings.max_ssl_handshake_roundtrips;
			MAX_READ_LENGTH = settings.max_receive_buffer_size;
			MAX_SEND_LENGTH = settings.max_send_buffer_size;
			READ_TIMEOUT = settings.read_timeout;
			WRITE_TIMEOUT = settings.send_timeout;
		};

		for k,default in pairs(defaults) do
			server.cfg[k] = event_settings[k] or default;
		end
	end
elseif server_type == "select" then
	-- TODO Remove completely.
	log("warn", "select is deprecated, the new default is epoll. For more info see https://prosody.im/doc/network_backend");
	server = require "prosody.net.server_select";

	local defaults = {};
	for k,v in pairs(server.getsettings()) do
		defaults[k] = v;
	end
	function set_config(settings)
		local select_settings = {};
		for k,default in pairs(defaults) do
			select_settings[k] = settings[k] or default;
		end
		server.changesettings(select_settings);
	end
else
	server = require("prosody.net.server_"..server_type);
	set_config = server.set_config;
	if not server.get_backend then
		function server.get_backend()
			return server_type;
		end
	end
end

-- If server.hook_signal exists, replace signal.signal()
local has_signal, signal = pcall(require, "prosody.util.signal");
if has_signal then
	if server.hook_signal then
		function signal.signal(signal_id, handler)
			if type(signal_id) == "string" then
				signal_id = signal[signal_id:upper()];
			end
			if type(signal_id) ~= "number" then
				return false, "invalid-signal";
			end
			return server.hook_signal(signal_id, handler);
		end
	else
		server.hook_signal = signal.signal;
	end
else
	if not server.hook_signal then
		server.hook_signal = function()
			return false, "signal hooking not supported"
		end
	end
end

if prosody and set_config then
	local config_get = require "prosody.core.configmanager".get;
	local function load_config()
		local settings = config_get("*", "network_settings") or {};
		return set_config(settings);
	end
	load_config();
	prosody.events.add_handler("config-reloaded", load_config);
end

if prosody and server.tls_builder then
	local tls_builder = server.tls_builder;
	-- resolving the basedir here avoids util.sslconfig depending on
	-- prosody.paths.config
	function server.tls_builder()
		return tls_builder(prosody.paths.config or "")
	end
end

-- require "prosody.net.server" shall now forever return this,
-- ie. server_select or server_event as chosen above.
return server;