jnormington / mimiq Goto Github PK

##Mimiq

As a QA Tester I am always testing with third party integration services such as;

• postcode services
• analytic services
• sms sending/receiving services

The problem is not all web applications rescue the common scenarios adequately or another layer in the application/webserver expires before hand - this app allows you to mimic these scenarios.

This application trys to build certain path scenarios to test on applications that have third party integrations - for which you can point your configuration url to this one for the AUT (Application under Test) - as we can't ask third party services to provide these for testing

It tries to provide the following;

• timeout (configurable by param)
• 404'ing (although easily replicatable with external service)
• 500'ing
• GET request - with different configurable responses
• POST request - with different configurable responses

##Deployment

• Clone this repository
• Bundle install bundle
• Run the webbrick rails server locally rails s

• Within this repository there is an ansible directory - this contains a playbook for
• creating/uploading ssh public key to digital ocean
• creating a ubuntu 15.04 droplet
• configuring the box (firewall/ruby/user/webserver)
• deploying the application

####Steps

• Install ansible (>1.9) on your machine (osx use homebrew/ubuntu use apt etc) brew install ansible

• Install ansible requirements ansible-galaxy install -f -r requirements.yml

• Generate tokens on digital ocean api v1

• Take note of the token and your client_api key

• Change directory to the ansible within the cloned application directory

• Replace both <YOUR_API_TOKEN> and <YOUR_CLIENT_KEY> with the respective values below.

• Create ssh public key and create droplet

 ansible-playbook site.yml --extra-vars="DO_API_KEY=<YOUR_API_TOKEN> DO_CLIENT_ID=<YOUR_CLIENT_KEY>" -i hosts --tags prereq

• Configure the droplet and deploy application
• You will be asked two questions for the droplet size and droplet location please input the id of both (slugs don't work)

 ssh-add ~/.ssh/mimiq_rsa && ansible-playbook site.yml --extra-vars="DO_API_KEY=<YOUR_API_TOKEN> DO_CLIENT_ID=<YOUR_CLIENT_KEY>" -i hosts --tags python_setup,configure,deploy

• Once completed both visit the ip address in the hosts file in the ansible directory

You can configure a response with the following attributes

• request_by - this makes the response unique by requesting this specific url
• response_type (this allows the following responses)
  • generic 404/500 (doesn't use content - just put hypen - in the content)
  • 422 Unprocessable (possible to replicate with API - but still good to test)
  • 200 (XML or JSON response - the content isn't validated in any way)
• wait time (this is the time it waits before responding - good for testing timeouts with analytics)
• request_type (this is whether you are posting or doing a GET request for the response)
• content (this is the content returned)

###Example

• When you create a response at /responses/new you are able to access that url in the following way.

  • GET response /scenario/get/<request_by>
  • POST response /scenario/post/<request_by>

• When hitting either url with the correct request (GET/POST) then you will receive the response setup. Below shows such example

####GET request/response

• When a response for GET request with the request_by set to test - then you can hit the application with the below url http://localhost:3000/scenario/get/test you get the content in the response, as below is shown.

$ curl -D- http://localhost/scenario/get/test; echo

HTTP/1.1 200 OK
X-Frame-Options: SAMEORIGIN
X-Xss-Protection: 1; mode=block
X-Content-Type-Options: nosniff
Content-Type: application/json; charset=utf-8
Etag: W/"195bc37dd38929786ad8ba5d0aea71d3"
Cache-Control: max-age=0, private, must-revalidate
X-Request-Id: 4e080c83-e112-44c1-99f2-19ee86c2cd26
X-Runtime: 0.004928
Server: WEBrick/1.3.1 (Ruby/2.2.2/2015-04-13)
Date: Sun, 14 Jun 2015 18:20:16 GMT
Content-Length: 14
Connection: Keep-Alive

{blank: 'yes'}  <--- RESPONSE HERE

####POST request/response

• When a response for POST request with the request_by set to a postcode AA11CU - then you can hit the application with the below url with a post request http://localhost:3000/scenario/post/test and you get the content in the response, as below is shown.

As shown below the application doesn't care what is posted to the server be nothing to something it just responds based on your custom response you have built.

$ wget --post-data='' http://localhost:3000/scenario/post/AA11CU -O- -v

--2015-06-14 19:51:12--  http://localhost:3000/scenario/post/AA11CU
Resolving localhost... ::1, 127.0.0.1
Connecting to localhost|::1|:3000... connected.
HTTP request sent, awaiting response... 302 Found
Location: http://localhost:3000/scenario/get/AA11CU?response_id=4 [following]
--2015-06-14 19:51:12--  http://localhost:3000/scenario/get/AA11CU?response_id=4
Reusing existing connection to [localhost]:3000.
HTTP request sent, awaiting response... 200 OK
Length: 144 [application/xml]
Saving to: 'STDOUT'

<-- RESPONSE BELOW THIS LINE -->

<xml>
  <postcodes>
    <postcode>
      <borough>nowhere</borough>
      <address>1 sadroad</road>
    <postcode>
  </postcodes>
</xml>