In my cucumber support directory I have the following in vcr.rb:
require 'vcr'
VCR.configure do |c|
c.cassette_library_dir = 'fixtures/vcr_cassettes'
c.hook_into :webmock
c.ignore_localhost = true
c.default_cassette_options = { record: :new_episodes }
end
I am geocoding city names which makes calls to the Google Maps API. I'm trying to record and stub these requests, but it keeps recording the same requests to the same yml file:
- request:
method: get
uri: http://maps.googleapis.com/maps/api/geocode/json?address=Miami,%20FL&language=en&sensor=false
body:
encoding: US-ASCII
string: ''
headers:
Accept-Encoding:
- gzip;q=1.0,deflate;q=0.6,identity;q=0.3
Accept:
- ! '*/*'
User-Agent:
- Ruby
# response...
- request:
method: get
uri: http://maps.googleapis.com/maps/api/geocode/json?address=Miami,%20FL&language=en&sensor=false
body:
encoding: US-ASCII
string: ''
headers:
Accept-Encoding:
- gzip;q=1.0,deflate;q=0.6,identity;q=0.3
Accept:
- ! '*/*'
User-Agent:
- Ruby
It's the same URL and very same request, shouldn't VCR stub the request? How can I prevent my specs from hitting the API every time I try to search for the same city?
It's hard to say for sure what's going on with just what you've posted, but I can explain a bit more about how VCR works and make some guesses about possible reasons for this behavior.
VCR uses request matchers to try to find a previously recorded HTTP interaction to play back. During a single cassette session, when an HTTP interaction is played back, it is considered to be "used" and it will not be played back again (unless you use the allow_playback_repeats option).
So...here are a couple possibilities that come to mind:
Maybe VCR is unable to match your HTTP requests. What request matchers are you using? There are some easy ways to troubleshoot this (see below).
If you're not using :allow_playback_repeats (which is the default, and how I recommend you use VCR), then the behavior you're seeing could happen if multiple duplicate requests are being made in your test--e.g., maybe the cassette has only one matching request but you're test makes 2 of them--that would play back one and record one (since you're using :new_episodes).
To troubleshoot this, I recommend you use the debug_logger option to get VCR to print what it's doing and how it's trying to match each request. That should give you insight into what's going on. You can also override any of the built in request matchers and provide your own logic and/or set a breakpoint in the matcher:
VCR.configure do |c|
c.register_request_matcher :uri do |request_1, request_2|
debugger # so you can inspect the requests here
request_1.uri == request_2.uri
end
end
You may also have run into a VCR bug, although comparing the URIs (using String#==) is such a basic operation that I have a hard time imagining a bug there. Feel free to open a github issue (hopefully with the debug logger output and/or a code sample that triggers this) if you can't figure it out.
On a side note, I recommend you use the :once record mode (the default) rather than :new_episodes. :once will never record additional HTTP interactions to an existing cassette--it only allows the cassette to be recorded once. If a request fails to match, it'll raise an error alerting you to the fact it couldn't match. :new_episodes, on the other hand, will record any request that it can't find a match for, which is the behavior you're seeing.
When I had a similar problem, I fixed it by making the match_requests_on setting more specific:
VCR.configure do |c|
c.default_cassette_options = {
match_requests_on: [:uri, :body, :method]
}
end
I experienced similar behaviour, so what I do is basically keeping to set to :none. If any new requests come up, I use :any, run the part of the test suite that did the requests and set it back to :none.
Seems like :new_episodes uses some strange heuristics to detemermine what new requests are and what requests already occurred. In our case it marked two different requests to the payment gateways as the same ones, leading to endless hours of debugging - because we got a RefundOk answer to a CaptureRequest and things like that. Better don't use :new_episodes...
Using Elasticsearch with VCR will always regenerate the cassettes if you not define the match parameters. I've to define the match to only :method because the :uri and :body could change each time that the test is running.
VCR.configure do |c|
c.hook_into :webmock
c.ignore_localhost = true
c.configure_rspec_metadata!
c.cassette_library_dir = 'spec/cassettes'
c.default_cassette_options = { record: :new_episodes,
match_requests_on: [:method] }
end
Related
(This question replaces this one, hopefully with better information.)
I've got three servers that I'm gonna call Alice, Charlie, and Harry (Harry being the server's actual nickname, so as not to confuse myself). They all talk to each other to perform quite a complicated authentication flow:
A client performs a sequence of three requests through Alice to Harry.
On the third one, Harry makes a call to Charlie.
Charlie is prone to timeouts during periods of heavy traffic. If it does, Harry returns a 503 with a Retry-After header to Alice.
Harry is returning a 503, I have confirmed that in its own logs.
Alice is not receiving a 503 but a 500, and without the header.
Alice's other clients (which I'm not in control of) treat a 500 the same as other errors, which is what I'm ultimately trying to fix.
Some extra information, such as I have been able to ascertain:
Alice proxies calls to Harry using RestClient, which uses Net::HTTP under the hood.
Using Net::HTTP directly gives the same result.
It's not environment specific; I have had this problem both in Production and Development.
I have been trying to simulate Alice using Postman, but haven't had any luck yet; Charlie's traffic is quieter at the moment, and I can't force or simulate a timeout, so so far I've only been getting successful 200 responses from Harry.
Fixing Charlie's timeouts would obviously be ideal, but I'm not in control of Charlie's hardware either.
Is there something I can change about Alice so it properly detects Harry's 503?
Or, is it possible that something about Harry is changing its 503 to a 500 after it's returned and logged?
Here's Alice's code for that third call, if it's likely to help, but nothing's jumping out at me; I have been wondering if RestClient or Net::HTTP has some configuration that I don't know about.
http_verb = :post
args = [ # actually constructed differently, but this is a reasonable mock up
'https://api.harry/path?sso_token=token',
'',
{
'content_type' => 'application/json',
'accept' => '*/*',
# some other app-specific headers
}
]
RestClient.send(http_verb, *args) do |response, request, result, &block|
# `result` is present and has a 500 code at the start of this block if Harry returns a 503.
#status_code = result.present? ? result.code : :internal_server_error
cors.merge!( response.headers.slice(:access_control_allow_origin, :access_control_request_method) )
#body = response.body
end
Turns out it's something way simpler and more obvious than anyone thought.
A separate error was raised in the middleware responsible for returning the 503. As with any other exception, this got rendered as a 500.
The thing that was causing it was a line that was supposed to tell the client to wait five seconds and try again:
response.headers['Retry-After'] = 5
... some middleware component was complaining of undefined method 'each' on 5:Fixnum, because it was expecting an Array where it wasn't a String; it started working when we changed 5 to '5'.
Ruby 1.9.3, RSpec 2.13.0, WebMock 1.17.4, Rails 3
I am writing tests for a company app. The controller in question displays a table of a customer's placed calls, and allows for sort/filter options.
EDIT The test fails because with my current setup, the path does not render, because the recorder_server is either not running locally, OR not setup correctly. Please help with this, too.
A Errno::ECONNREFUSED occurred in recordings#index:
Connection refused - connect(2)
/usr/local/lib/ruby/1.9.1/net/http.rb:763:in `initialize'
-------------------------------
Request:
-------------------------------
* URL : http://www.recorder.example.com:8080/recorded_calls
* IP address: 127.0.0.1
* Parameters: {"controller"=>"recordings", "action"=>"index"}
* Rails root: /var/www/rails/<repository>
As a call is placed, its data joins an xml file, created by an external API, called Recorder
The RecordingsController takes the xml file, and parses it into a hash.
When you visit the associated path, you see the results of the hash -- a table of placed calls, their attributes, and parameters for sort/filter.
Here is my spec so far.
require 'spec_helper'
include Helpers
feature 'Exercise recordings controller' do
include_context "shared admin context"
background do
canned_xml = File.open("spec/support/assets/canned_response.xml").read
stub_request(:post, "http://recorder.example.com:8080/recorder/index").
with(body: {"durations"=>["1"], "durations_greater_less"=>["gt"], "filter_from_day"=>"29", "filter_from_hour"=>"0", "filter_from_minute"=>"0", "filter_from_month"=>"12", "filter_from_year"=>"2014", "filter_prefix"=>true, "filter_to_day"=>"29", "filter_to_hour"=>"23", "filter_to_minute"=>"59", "filter_to_month"=>"12", "filter_to_year"=>"2014"}, # "shared_session_id"=>"19f9a08807cc70c1bf41885956695bde"},
headers: {'Accept'=>'*/*', 'Content-Type'=>'application/x-www-form-urlencoded', 'User-Agent'=>'Ruby'}).
to_return(status: 200, body: canned_xml, headers: {})
uri = URI.parse("http://recorder.example.com:8080/recorder/index")
visit recorded_calls_path
end
scenario 'show index page with 1 xml result' do
#page.save_and_open_page
expect(title).to eq("Recorded Calls")
end
end
And here is the RecordingsController
class RecordingsController < ApplicationController
# before_filter options
def index
test_session_id = request.session_options[:id]
#Make request to recording app for xml of files
uri = URI.parse("http://#{Rails.application.config.recorder_server}:#{Rails.application.config.recorder_server_port}/recorder/index")
http = Net::HTTP.new(uri.host, uri.port)
xml_request = Net::HTTP::Post.new(uri.request_uri)
xml_request_data = Hash.new
# sorting params
xml_request_data[:shared_session_id] = request.session_options[:id]
xml_request.set_form_data(xml_request_data)
response = http.request(xml_request)
if response.class == Net::HTTPOK
#recordings_xml = XmlSimple.xml_in(response.body)
#recordings_sorted = #recordings_xml["Recording"].sort { |a,b| Time.parse("#{a["date"]} #{a["time"]}") <=> Time.parse("#{b["date"]} #{b["time"]}") } unless #recordings_xml["Recording"].nil?
else #recordings_xml = Hash.new
end
end
# other defs
end
Any and all advice is much appreciated. Thank you.
How I configured WebMock
I am answering my own question, with the help of B-Seven and a string of comments. File by file, I will list the changes made in order to properly use WebMock.
Add WebMock to Gemfile under group :test, :development.
bundle install to resolve dependencies
my current setup included Ruby 1.9.3, Rails 2.13.0, WebMock 1.17.4
Setup spec_helper.rb to disable "Real HTTP connections". (This was a backtrace error received later on in this puzzling process.) This allows, to my understanding, all "real connections" to translate into localhost connections and work offline... Which is great since, ideally, I do not want the external app's server to run simultaneously.
require 'webmock/rspec'
WebMock.disable_net_connect!(allow_localhost: true)
In my test.rb environment file, the configurations for recorder_server and port were commented out... If left uncommented, the controller would raise an exception stating uninitialized constants. I used the test server/port (substituting the company name for example) as my layout for the spec stubbing.
In recordings_controller_spec.rb, I had already figured out how to make a canned XML response. With these changes above, my spec was able to correctly stub a response on an external, secondary app, and use such response to correctly render the view associated with the controller being tested.
require 'spec_helper'
include Helpers
feature "Exercise recordings_controller" do
include_context "shared admin context"
# A background is currently not used, because I have 3 scenario types... No xml
# results, 1 result, and 2 results. I will later DRY this out with a background,
# but the heavy lifting is over, for now.
scenario "show index page with 1 xml result" do
canned_xml_1 = File.open("spec/support/assets/canned_response_1.xml").read
stub_request(:post, "http://recorder.example.com:8080/recorder/index").
with(headers: {'Accept'=>'*/*', 'User-Agent'=>'Ruby'}).
to_return(status: 200, body: canned_xml_1, headers: {})
uri = URI.parse("http://recorder.example.com:8080/recorder/index")
visit recorded_calls_path
title.should == "Recorded Calls"
page.should have_content("Search Results")
page.should have_content("Inbound", "5551230000", "175", "December 24 2014", "12:36:24", "134")
end
end
Advice/Resources that helped
With B-Seven's suggestion to my original question (see revisions), I was initially stubbing localhost:3000. He said this was incorrect. After further research, I agree since stubbing with WebMock is typically reserved for outside http connections.
In comments after his answer, B-Seven listed articles to refer to. I will list the ones that helped me the most.
http://robots.thoughtbot.com/how-to-stub-external-services-in-tests
http://railscasts.com/episodes/275-how-i-test
https://github.com/bblimke/webmock
http://www.agileventures.org/articles/testing-with-rspec-stubs-mocks-factories-what-to-choose
It is very important to read the backtrace generated from an errors. What took me so long to figure out how to mock was mainly reading them incorrectly. As you can see from my question, I was making a :get stub request. A coworker pointed out that the backtrace suggested to use :post. That was the final piece to make my spec pass.
I decided not to input the configuration variables as my stub request, for it would result in long lines of code. Instead, this is why I needed to uncomment out those configurations in test.rb.
Why are you stubbing localhost? I think you want to
stub_request(:get, "http://#{Rails.application.config.recorder_server}:#{Rails.application.config.recorder_server_port}/recorder/index").
I have VCR set up and it runs on several tests I have written without issue. The most recent test I tried to write will pass the first time it is run, but will remain failing after that unless I delete the cassette. The code for the test is:
it "doesn't blow up if a user doesn't have billing info" do
VCR.use_cassette('tax_reconciler/no_method_error')do
user_guid = rand(10000000)
CreateRecurlyTestData.create_account(user_guid, nil, nil)
tax_reconciler = TaxReconciler.new
new_tax_amount = rand(100000)
user = create_test_user(:guid => user_guid)
expect(tax_reconciler.update_tax_amount(user, new_tax_amount)).to_not raise_error
end
end
The error message is as follows:
Failure/Error: CreateRecurlyTestData.create_account(user_guid, nil, nil)
VCR::Errors::UnhandledHTTPRequestError:
================================================================================
An HTTP request has been made that VCR does not know how to handle:
GET https://2b64d08ef45c446dbba75720a37b7d41:#api.recurly.com/v2/accounts/3276643
VCR is currently using the following cassette:
- /Users/Evan/dish/stallone/fixtures/vcr_cassettes/tax_reconciler/no_method_error.yml
- :record => :once
- :match_requests_on => [:method, :uri]
Under the current configuration VCR can not find a suitable HTTP interaction
to replay and is prevented from recording new requests. There are a few ways
you can deal with this:
* If you're surprised VCR is raising this error
and want insight about how VCR attempted to handle the request,
you can use the debug_logger configuration option to log more details [1].
* You can use the :new_episodes record mode to allow VCR to
record this new request to the existing cassette [2].
* If you want VCR to ignore this request (and others like it), you can
set an `ignore_request` callback [3].
* The current record mode (:once) does not allow new requests to be recorded
to a previously recorded cassette. You can delete the cassette file and re-run
your tests to allow the cassette to be recorded with this request [4].
* The cassette contains 4 HTTP interactions that have not been
played back. If your request is non-deterministic, you may need to
change your :match_requests_on cassette option to be more lenient
or use a custom request matcher to allow it to match [5].
[1] https://www.relishapp.com/vcr/vcr/v/2-9-3/docs/configuration/debug-logging
[2] https://www.relishapp.com/vcr/vcr/v/2-9-3/docs/record-modes/new-episodes
[3] https://www.relishapp.com/vcr/vcr/v/2-9-3/docs/configuration/ignore-request
[4] https://www.relishapp.com/vcr/vcr/v/2-9-3/docs/record-modes/once
[5] https://www.relishapp.com/vcr/vcr/v/2-9-3/docs/request-matching
The only config options in spec helper are:
VCR.configure do |c|
c.cassette_library_dir = 'fixtures/vcr_cassettes'
c.hook_into :webmock # or :fakeweb
end
Your VCR setup is configured in recording mode :once: https://www.relishapp.com/vcr/vcr/v/1-8-0/docs/record-modes/once
In short, it only records requests when there is no cassette. For subsequent runs, it will only read the recorded requests, so a request that doesn't match any of the recorded requests will raise this error.
One common way for things to go wrong with this mode is if you delete your cassette and the only run one spec, instead of the whole suite. Bingo - your cassette is created and all requests except for the one in that particular spec will be unrecognized.
An upside to this mode is that you can catch spurious requests - if you know that your cassette contains all valid requests, you can catch bugs that are due to incorrect requests by using this mode.
If you want to be more flexible, you can use the recording mode new_episodes. In this mode, recognized requests will be replayed from the cassette, and unrecoqnized ones will be performed and recorded.
I'm trying to run cucumber using VCR gem, chrome driver, and webmock but it's not working. One problem is that none of the requests are getting recorded. I do have VCR set to ignore_localhost which I believe is needed due to javascript as explained in this SO answer.
But why doesn't my step request to visit a url (say www.test.com) get recorded? I see the chrome browser loading up and the address in the URL so I know the cucumber step is going there.
If i turn off ignore_localhost then I see some recordings but the uri's are all 127.0.0.1. The uri I'd expect is showing up in body: string instead of the uri. Why is this happening?
request:
method: post
uri: http://127.0.0.1:9592/session/87bb12aa8b5b1230399b98bcc7d0ba11/url
body:
encoding: US-ASCII
string: ! '{"url":"http://www.test.com"}'
headers:
Accept:
- application/json
Content-Type:
- application/json; charset=utf-8
Content-Length:
- '46'
User-Agent:
- Ruby
Incase it helps, here's my VCR config
VCR.configure do |c|
c.cassette_library_dir = 'features/support/vcr_cassettes'
c.hook_into :webmock
c.ignore_localhost = true
end
First of all, set your configuration like this:
VCR.configure do |c|
c.cassette_library_dir = 'features/support/vcr_cassettes'
c.hook_into :webmock
c.ignore_localhost = true
c.allow_http_connections_when_no_cassette = false
end
With the last option you tell VCR that no real http requests are allowed, without a cassette. Now run your tests. I bet that it'll throw an error! Now you have to tell VCR when to use which cassette. Have a look at the cucumber helpers for VCR here https://www.relishapp.com/vcr/vcr/v/2-4-0/docs/test-frameworks/usage-with-cucumber
VCR includes some helpers for other testing frameworks too https://www.relishapp.com/vcr/vcr/v/2-4-0/docs/test-frameworks
For reasons similar to the ones in this discussion, I'm experimenting with messaging in lieu of REST for a synchronous RPC call from one Rails 3 application to another. Both apps are running on thin.
The "server" application has a config/initializers/amqp.rb file based on the Request / Reply pattern in the rubyamqp.info documentation:
require "amqp"
EventMachine.next_tick do
connection = AMQP.connect ENV['CLOUDAMQP_URL'] || 'amqp://guest:guest#localhost'
channel = AMQP::Channel.new(connection)
requests_queue = channel.queue("amqpgem.examples.services.time", :exclusive => true, :auto_delete => true)
requests_queue.subscribe(:ack => true) do |metadata, payload|
puts "[requests] Got a request #{metadata.message_id}. Sending a reply..."
channel.default_exchange.publish(Time.now.to_s,
:routing_key => metadata.reply_to,
:correlation_id => metadata.message_id,
:mandatory => true)
metadata.ack
end
Signal.trap("INT") { connection.close { EventMachine.stop } }
end
In the 'client' application, I'd like to render the results of a synchronous call to the 'server' in a view. I realize this is a bit outside the comfort zone of an inherently asynchronous library like the amqp gem, but I'm wondering if there's a way to make it work. Here is my client config/initializers/amqp.rb:
require 'amqp'
EventMachine.next_tick do
AMQP.connection = AMQP.connect 'amqp://guest:guest#localhost'
Signal.trap("INT") { AMQP.connection.close { EventMachine.stop } }
end
Here is the controller:
require "amqp"
class WelcomeController < ApplicationController
def index
puts "[request] Sending a request..."
WelcomeController.channel.default_exchange.publish("get.time",
:routing_key => "amqpgem.examples.services.time",
:message_id => Kernel.rand(10101010).to_s,
:reply_to => WelcomeController.replies_queue.name)
WelcomeController.replies_queue.subscribe do |metadata, payload|
puts "[response] Response for #{metadata.correlation_id}: #{payload.inspect}"
#message = payload.inspect
end
end
def self.channel
#channel ||= AMQP::Channel.new(AMQP.connection)
end
def self.replies_queue
#replies_queue ||= channel.queue("reply", :exclusive => true, :auto_delete => true)
end
end
When I start both applications on different ports and visit the welcome#index view.
#message is nil in the view, since the result has not yet returned. The result arrives a few milliseconds after the view is rendered and is displayed on the console:
$ thin start
>> Using rack adapter
>> Thin web server (v1.5.0 codename Knife)
>> Maximum connections set to 1024
>> Listening on 0.0.0.0:3000, CTRL+C to stop
[request] Sending a request...
[response] Response for 3877031: "2012-11-27 22:04:28 -0600"
No surprise here: subscribe is clearly not meant for synchronous calls. What is surprising is that I can't find a synchronous alternative in the AMQP gem source code or in any documentation online. Is there an alternative to subscribe that will give me the RPC behavior I want? Given that there are other parts of the system in which I'd want to use legitimately asynchronous calls, the bunny gem didn't seem like the right tool for the job. Should I give it another look?
edit in response to Sam Stokes
Thanks to Sam for the pointer to throw :async / async.callback. I hadn't seen this technique before and this is exactly the kind of thing I was trying to learn with this experiment in the first place. send_response.finish is gone in Rails 3, but I was able to get his example to work for at least one request with a minor change:
render :text => #message
rendered_response = response.prepare!
Subsequent requests fail with !! Unexpected error while processing request: deadlock; recursive locking. This may have been what Sam was getting at with the comment about getting ActionController to allow concurrent requests, but the cited gist only works for Rails 2. Adding config.allow_concurrency = true in development.rb gets rid of this error in Rails 3, but leads to This queue already has default consumer. from AMQP.
I think this yak is sufficiently shaven. ;-)
While interesting, this is clearly overkill for simple RPC. Something like this Sinatra streaming example seems a more appropriate use case for client interaction with replies. Tenderlove also has a blog post about an upcoming way to stream events in Rails 4 that could work with AMQP.
As Sam points out in his discussion of the HTTP alternative, REST / HTTP makes perfect sense for the RPC portion of my system that involves two Rails apps. There are other parts of the system involving more classic asynchronous event publishing to Clojure apps. For these, the Rails app need only publish events in fire-and-forget fashion, so AMQP will work fine there using my original code without the reply queue.
You can get the behaviour you want - have the client make a simple HTTP request, to which your web app responds asynchronously - but you need more tricks. You need to use Thin's support for asynchronous responses:
require "amqp"
class WelcomeController < ApplicationController
def index
puts "[request] Sending a request..."
WelcomeController.channel.default_exchange.publish("get.time",
:routing_key => "amqpgem.examples.services.time",
:message_id => Kernel.rand(10101010).to_s,
:reply_to => WelcomeController.replies_queue.name)
WelcomeController.replies_queue.subscribe do |metadata, payload|
puts "[response] Response for #{metadata.correlation_id}: #{payload.inspect}"
#message = payload.inspect
# Trigger Rails response rendering now we have the message.
# Tested in Rails 2.3; may or may not work in Rails 3.x.
rendered_response = send_response.finish
# Pass the response to Thin and make it complete the request.
# env['async.callback'] expects a Rack-style response triple:
# [status, headers, body]
request.env['async.callback'].call(rendered_response)
end
# This unwinds the call stack, skipping the normal Rails response
# rendering, all the way back up to Thin, which catches it and
# interprets as "I'll give you the response later by calling
# env['async.callback']".
throw :async
end
def self.channel
#channel ||= AMQP::Channel.new(AMQP.connection)
end
def self.replies_queue
#replies_queue ||= channel.queue("reply", :exclusive => true, :auto_delete => true)
end
end
As far as the client is concerned, the result is indistinguishable from your web app blocking on a synchronous call before returning the response; but now your web app can process many such requests concurrently.
CAUTION!
Async Rails is an advanced technique; you need to know what you're doing. Some parts of Rails do not take kindly to having their call stack abruptly dismantled. The throw will bypass any Rack middlewares that don't know to catch and rethrow it (here is a rather old partial solution). ActiveSupport's development-mode class reloading will reload your app's classes after the throw, without waiting for the response, which can cause very confusing breakage if your callback refers to a class that has since been reloaded. You'll also need to ask ActionController nicely to allow concurrent requests.
Request/response
You're also going to need to match up requests and responses. As it stands, if Request 1 arrives, and then Request 2 arrives before Request 1 gets a response, then it's undefined which request would receive Response 1 (messages on a queue are distributed round-robin between the consumers subscribed to the queue).
You could do this by inspecting the correlation_id (which you'll have to explicitly set, by the way - RabbitMQ won't do it for you!) and re-enqueuing the message if it's not the response you were waiting for. My approach was to create a persistent Publisher object which would keep track of open requests, listen for all responses, and lookup the appropriate callback to invoke based on the correlation_id.
Alternative: just use HTTP
You're really solving two different (and tricky!) problems here: persuading Rails/thin to process requests asynchronously, and implementing request-response semantics on top of AMQP's publish-subscribe model. Given you said this is for calling between two Rails apps, why not just use HTTP, which already has the request-response semantics you need? That way you only have to solve the first problem. You can still get concurrent request processing if you use a non-blocking HTTP client library, such as em-http-request.