File

net/adns.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 12974:ba409c67353b
child 13659:5abdcad8c2e0
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 server = require "prosody.net.server";
local new_resolver = require "prosody.net.dns".resolver;
local promise = require "prosody.util.promise";

local log = require "prosody.util.logger".init("adns");

log("debug", "Using legacy DNS API (missing lua-unbound?)"); -- TODO write docs about luaunbound
-- TODO Raise log level once packages are available

local coroutine, pcall = coroutine, pcall;
local setmetatable = setmetatable;

local function dummy_send(sock, data, i, j) return (j-i)+1; end -- luacheck: ignore 212

local _ENV = nil;
-- luacheck: std none

local async_resolver_methods = {};
local async_resolver_mt = { __index = async_resolver_methods };

local query_methods = {};
local query_mt = { __index = query_methods };

local function new_async_socket(sock, resolver)
	local peername = "<unknown>";
	local listener = {};
	local handler = {};
	function listener.onincoming(conn, data) -- luacheck: ignore 212/conn
		if data then
			resolver:feed(handler, data);
		end
	end
	function listener.ondisconnect(conn, err)
		if err then
			log("warn", "DNS socket for %s disconnected: %s", peername, err);
			local servers = resolver.server;
			if resolver.socketset[conn] == resolver.best_server and resolver.best_server == #servers then
				log("warn", "Exhausted all %d configured DNS servers, next lookup will try %s again", #servers, servers[1]);
			end

			resolver:servfail(conn); -- Let the magic commence
		end
	end
	do
		local err;
		handler, err = server.wrapclient(sock, "dns", 53, listener);
		if not handler then
			return nil, err;
		end
	end
	if handler.set then
		-- server_epoll: only watch for incoming data
		-- avoids sending empty packet on first 'onwritable' event
		handler:set(true, false);
	end

	handler.settimeout = function () end
	handler.setsockname = function (_, ...) return sock:setsockname(...); end
	handler.setpeername = function (_, ...) peername = (...); local ret, err = sock:setpeername(...); _:set_send(dummy_send); return ret, err; end
	handler.connect = function (_, ...) return sock:connect(...) end
	--handler.send = function (_, data) _:write(data);  return _.sendbuffer and _.sendbuffer(); end
	handler.send = function (_, data)
		log("debug", "Sending DNS query to %s", peername);
		return sock:send(data);
	end
	return handler;
end

function async_resolver_methods:lookup(handler, qname, qtype, qclass)
	local resolver = self._resolver;
	return coroutine.wrap(function (peek)
				if peek then
					log("debug", "Records for %s already cached, using those...", qname);
					handler(peek);
					return;
				end
				log("debug", "Records for %s not in cache, sending query (%s)...", qname, coroutine.running());
				local ok, err = resolver:query(qname, qtype, qclass);
				if ok then
					coroutine.yield(setmetatable({ resolver, qclass or "IN", qtype or "A", qname, coroutine.running()}, query_mt)); -- Wait for reply
					log("debug", "Reply for %s (%s)", qname, coroutine.running());
				end
				if ok then
					ok, err = pcall(handler, resolver:peek(qname, qtype, qclass));
				else
					log("error", "Error sending DNS query: %s", err);
					ok, err = pcall(handler, nil, err);
				end
				if not ok then
					log("error", "Error in DNS response handler: %s", err);
				end
			end)(resolver:peek(qname, qtype, qclass));
end

function async_resolver_methods:lookup_promise(qname, qtype, qclass)
	return promise.new(function (resolve, reject)
		local function handler(answer)
			if not answer then
				return reject();
			end
			resolve(answer);
		end
		self:lookup(handler, qname, qtype, qclass);
	end);
end

function query_methods:cancel(call_handler, reason) -- luacheck: ignore 212/reason
	log("warn", "Cancelling DNS lookup for %s", self[4]);
	self[1].cancel(self[2], self[3], self[4], self[5], call_handler);
end

local function new_async_resolver()
	local resolver = new_resolver();
	resolver:socket_wrapper_set(new_async_socket);
	return setmetatable({ _resolver = resolver}, async_resolver_mt);
end

return {
	lookup = function (...)
		return new_async_resolver():lookup(...);
	end;
	resolver = new_async_resolver;
	new_async_socket = new_async_socket;
};