File

util/fsm.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 13165:9c13c11b199d
line wrap: on
line source

local events = require "prosody.util.events";

local fsm_methods = {};
local fsm_mt = { __index = fsm_methods };

local function is_fsm(o)
	local mt = getmetatable(o);
	return mt == fsm_mt;
end

local function notify_transition(fire_event, transition_event)
	local ret;
	ret = fire_event("transition", transition_event);
	if ret ~= nil then return ret; end
	if transition_event.from ~= transition_event.to then
		ret = fire_event("leave/"..transition_event.from, transition_event);
		if ret ~= nil then return ret; end
	end
	ret = fire_event("transition/"..transition_event.name, transition_event);
	if ret ~= nil then return ret; end
end

local function notify_transitioned(fire_event, transition_event)
	if transition_event.to ~= transition_event.from then
		fire_event("enter/"..transition_event.to, transition_event);
	end
	if transition_event.name then
		fire_event("transitioned/"..transition_event.name, transition_event);
	end
	fire_event("transitioned", transition_event);
end

local function do_transition(name)
	return function (self, attr)
		local new_state = self.fsm.states[self.state][name] or self.fsm.states["*"][name];
		if not new_state then
			return error(("Invalid state transition: %s cannot %s"):format(self.state, name));
		end

		local transition_event = {
			instance = self;

			name = name;
			to = new_state;
			to_attr = attr;

			from = self.state;
			from_attr = self.state_attr;
		};

		local fire_event = self.fsm.events.fire_event;
		local ret = notify_transition(fire_event, transition_event);
		if ret ~= nil then return nil, ret; end

		self.state = new_state;
		self.state_attr = attr;

		notify_transitioned(fire_event, transition_event);
		return true;
	end;
end

local function new(desc)
	local self = setmetatable({
		default_state = desc.default_state;
		events = events.new();
	}, fsm_mt);

	-- states[state_name][transition_name] = new_state_name
	local states = { ["*"] = {} };
	if desc.default_state then
		states[desc.default_state] = {};
	end
	self.states = states;

	local instance_methods = {};
	self._instance_mt = { __index = instance_methods };

	for _, transition in ipairs(desc.transitions or {}) do
		local from_states = transition.from;
		if type(from_states) ~= "table" then
			from_states = { from_states };
		end
		for _, from in ipairs(from_states) do
			if not states[from] then
				states[from] = {};
			end
			if not states[transition.to] then
				states[transition.to] = {};
			end
			if states[from][transition.name] then
				return error(("Duplicate transition in FSM specification: %s from %s"):format(transition.name, from));
			end
			states[from][transition.name] = transition.to;
		end

		-- Add public method to trigger this transition
		instance_methods[transition.name] = do_transition(transition.name);
	end

	if desc.state_handlers then
		for state_name, handler in pairs(desc.state_handlers) do
			self.events.add_handler("enter/"..state_name, handler);
		end
	end

	if desc.transition_handlers then
		for transition_name, handler in pairs(desc.transition_handlers) do
			self.events.add_handler("transition/"..transition_name, handler);
		end
	end

	if desc.handlers then
		self.events.add_handlers(desc.handlers);
	end

	return self;
end

function fsm_methods:init(state_name, state_attr)
	local initial_state = assert(state_name or self.default_state, "no initial state specified");
	if not self.states[initial_state] then
		return error("Invalid initial state: "..initial_state);
	end
	local instance = setmetatable({
		fsm = self;
		state = initial_state;
		state_attr = state_attr;
	}, self._instance_mt);

	if initial_state ~= self.default_state then
		local fire_event = self.events.fire_event;
		notify_transitioned(fire_event, {
			instance = instance;

			to = initial_state;
			to_attr = state_attr;

			from = self.default_state;
		});
	end

	return instance;
end

function fsm_methods:is_instance(o)
	local mt = getmetatable(o);
	return mt == self._instance_mt;
end

return {
	new = new;
	is_fsm = is_fsm;
};