File

net/connect.lua @ 11523:5f15ab7c6ae5

Statistics: Rewrite statistics backends to use OpenMetrics The metric subsystem of Prosody has had some shortcomings from the perspective of the current state-of-the-art in metric observability. The OpenMetrics standard [0] is a formalization of the data model (and serialization format) of the well-known and widely-used Prometheus [1] software stack. The previous stats subsystem of Prosody did not map well to that format (see e.g. [2] and [3]); the key reason is that it was trying to do too much math on its own ([2]) while lacking first-class support for "families" of metrics ([3]) and structured metric metadata (despite the `extra` argument to metrics, there was no standard way of representing common things like "tags" or "labels"). Even though OpenMetrics has grown from the Prometheus world of monitoring, it maps well to other popular monitoring stacks such as: - InfluxDB (labels can be mapped to tags and fields as necessary) - Carbon/Graphite (labels can be attached to the metric name with dot-separation) - StatsD (see graphite when assuming that graphite is used as backend, which is the default) The util.statsd module has been ported to use the OpenMetrics model as a proof of concept. An implementation which exposes the util.statistics backend data as Prometheus metrics is ready for publishing in prosody-modules (most likely as mod_openmetrics_prometheus to avoid breaking existing 0.11 deployments). At the same time, the previous measure()-based API had one major advantage: It is really simple and easy to use without requiring lots of knowledge about OpenMetrics or similar concepts. For that reason as well as compatibility with existing code, it is preserved and may even be extended in the future. However, code relying on the `stats-updated` event as well as `get_stats` from `statsmanager` will break because the data model has changed completely; in case of `stats-updated`, the code will simply not run (as the event was renamed in order to avoid conflicts); the `get_stats` function has been removed completely (so it will cause a traceback when it is attempted to be used). Note that the measure_*_event methods have been removed from the module API. I was unable to find any uses or documentation and thus deemed they should not be ported. Re-implementation is possible when necessary. [0]: https://openmetrics.io/ [1]: https://prometheus.io/ [2]: #959 [3]: #960
author Jonas Schäfer <jonas@wielicki.name>
date Sun, 18 Apr 2021 11:47:41 +0200
parent 10945:2edb72ef312a
child 11901:26406ce35e20
line wrap: on
line source

local server = require "net.server";
local log = require "util.logger".init("net.connect");
local new_id = require "util.id".short;

-- TODO #1246 Happy Eyeballs
-- FIXME RFC 6724
-- FIXME Error propagation from resolvers doesn't work
-- FIXME #1428 Reuse DNS resolver object between service and basic resolver
-- FIXME #1429 Close DNS resolver object when done

local pending_connection_methods = {};
local pending_connection_mt = {
	__name = "pending_connection";
	__index = pending_connection_methods;
	__tostring = function (p)
		return "<pending connection "..p.id.." to "..tostring(p.target_resolver.hostname)..">";
	end;
};

function pending_connection_methods:log(level, message, ...)
	log(level, "[pending connection %s] "..message, self.id, ...);
end

-- pending_connections_map[conn] = pending_connection
local pending_connections_map = {};

local pending_connection_listeners = {};

local function attempt_connection(p)
	p:log("debug", "Checking for targets...");
	if p.conn then
		pending_connections_map[p.conn] = nil;
		p.conn = nil;
	end
	p.target_resolver:next(function (conn_type, ip, port, extra)
		if not conn_type then
			-- No more targets to try
			p:log("debug", "No more connection targets to try");
			if p.listeners.onfail then
				p.listeners.onfail(p.data, p.last_error or "unable to resolve service");
			end
			return;
		end
		p:log("debug", "Next target to try is %s:%d", ip, port);
		local conn, err = server.addclient(ip, port, pending_connection_listeners, p.options.pattern or "*a", p.options.sslctx, conn_type, extra);
		if not conn then
			log("debug", "Connection attempt failed immediately: %s", err);
			p.last_error = err or "unknown reason";
			return attempt_connection(p);
		end
		p.conn = conn;
		pending_connections_map[conn] = p;
	end);
end

function pending_connection_listeners.onconnect(conn)
	local p = pending_connections_map[conn];
	if not p then
		log("warn", "Successful connection, but unexpected! Closing.");
		conn:close();
		return;
	end
	pending_connections_map[conn] = nil;
	p:log("debug", "Successfully connected");
	conn:setlistener(p.listeners, p.data);
	return p.listeners.onconnect(conn);
end

function pending_connection_listeners.ondisconnect(conn, reason)
	local p = pending_connections_map[conn];
	if not p then
		log("warn", "Failed connection, but unexpected!");
		return;
	end
	p.last_error = reason or "unknown reason";
	p:log("debug", "Connection attempt failed: %s", p.last_error);
	attempt_connection(p);
end

local function connect(target_resolver, listeners, options, data)
	local p = setmetatable({
		id = new_id();
		target_resolver = target_resolver;
		listeners = assert(listeners);
		options = options or {};
		data = data;
	}, pending_connection_mt);

	p:log("debug", "Starting connection process");
	attempt_connection(p);
end

return {
	connect = connect;
};