PerimeterX Ruby SDK

Latest stable version: v2.3.2

Usage

Dependencies
Installation
Basic Usage Example

Configuration

Configuring Required Parameters
Blocking Score
Custom Verification Action
Custom Block Page
Extracting Real IP Address
Custom URI
Filter Sensitive Headers
API Timeouts
Send Page Activities
Additional Page Activity Handler
Monitor Only
Debug Mode
Whitelist Routes
Update Configuration on Runtime
First Party

Additional Information Contributing

Dependencies

Ruby version 2.3+
Rails version 4.2
concurrent-ruby
typhoeus
mustache

Installation

Install it through command line gem install perimeter_x

Basic Usage Example

Configuration & Initialization

Create a configuration file at <rails_app>/config/initializers/perimeterx.rb and initialize PerimeterX instance on the rails application startup

params = {
  :app_id => "APP_ID",
  :cookie_key => "COOKIE_KEY",
  :auth_token => "AUTH_TOKEN"
}

PxModule.configure(params)

On the Rails controller include the PerimeterX SDK via the before_action and call PerimterX middleware function.

class HomeController < ApplicationController
  include PxModule

  before_action :px_verify_request
  ...
  ...
end

Configuration options

Configuring Required Parameters Configuration options are set on the params variable on the initializer file.

app_id
cookie_key
auth_token

All parameters are obtainable via the PerimeterX Portal. (Applications and Policies pages)

Changing the Minimum Score for Blocking

Note: Default blocking value: 100

params = {
  ...
  :blocking_score => 100
  ...
}

Custom Verification Handler

Note: This handler replaces the now deprecated custom_block_handler.

A custom verification handler is being executed inside px_verify_request instead of the the default behavior and allows a user to use a custom action based on the risk score returned by PerimeterX.

When implemented, this method receives a hash variable as input which represents data from the PerimeterX context of the request (px_ctx).

px_ctx.context[:score] - contains the risk score
px_ctx.context[:uuid] - contains the request UUID
px_ctx.context[:verified] - contains indication whether the request passed verification or was blocked (inspect px_ctx.context[:block_reason] for block reason)

Note: to determine whether to return a captcha/block page (HTML) or block JSON payload a reference key on the context will be available: px_ctx.context[:format]

To replace the default verification behavior, add the configuration a lambda member as shown in the example below.

params = {
  :app_id => <APP_ID>,
  :auth_token => <AUTH_TOKEN>,
  :custom_verification_handler => -> (px_ctx) {
    if px_ctx.context[:score] >= 60
      # take your action and render an html page or JSON with applicable status code.    
      render json: { :score => px_ctx.context[:score] }
    end
  }
}

Note: Unlike previous versions, the method no longer needs to return a boolean value.

Example

Serving a Custom HTML Page

params = {
  :app_id => <APP_ID>,
  :auth_token => <AUTH_TOKEN>,
  ...
  :custom_verification_handler => -> (px_ctx) {
    block_score = px_ctx.context[:score];
    client_uuid = px_ctx.context[:uuid];
    full_url = px_ctx.context[:full_url];

    html = "<html>
            <body>
            <div>Access to #{full_url} has been blocked.</div>    
            <div>Block reference - #{client_uuid} </div>
            <div>Block score - #{block_score} </div>
            </body>
            </html>".html_safe
    response.headers["Content-Type"] = "text/html"
    response.status = 403
    render :html => html
  }
}

Custom User IP

Note: IP extraction, according to your network setup, is very important. It is common to have a load balancer/proxy on top of your applications, in which case the PerimeterX module will send the system's internal IP as the user's. In order to properly perform processing and detection on server-to-server calls, PerimeterX module needs the real user's IP.

By default the clients IP is taken from the REMOTE_ADDR header, in case the user decides to use different header or custom function that extract the header the following key should be added to the configuration.

Custom header

Note: If there are multiple headers configured, the module will evaluate the headers in order and use the first value it finds.

configuration = {
  "app_id" => <APP_ID>,
  "auth_token" => <AUTH_TOKEN>,
  "ip_headers" => [<HTTP_HEADER_NAME>, <BACKUP_HTTP_HEADER_NAME>],

Custom Function

Note: the function receive as a first parameter the controller request and must return the ip at the end as string

configuration = {
  "app_id" => <APP_ID>,
  "auth_token" => <AUTH_TOKEN>,
  "ip_header_function" => -> (req) {
    # Method body
    return "1.2.3.4"
  }
}

Customizing Default Block Pages

Adding a custom logo to the blocking page is by providing the params a key custom_logo , the logo will be displayed at the top div of the the block page The logo's max-heigh property would be 150px and width would be set to auto

The key custom_logo expects a valid URL address such as https://s.perimeterx.net/logo.png

params = [
    :app_id => 'APP_ID',
    :cookie_key => 'COOKIE_SECRET',
    :auth_token => 'AUTH_TOKEN',
    :custom_logo => 'LOGO_URL'
];

Custom JS/CSS The block page can be modified with a custom CSS by adding to the params the key css_ref and providing a valid URL to the css In addition there is also the option to add a custom JS file by adding js_ref key to the pxConfig and providing the JS file that will be loaded with the block page, this key also expects a valid URL

params = [
    :app_id => 'APP_ID',
    :cookie_key => 'COOKIE_SECRET',
    :auth_token => 'AUTH_TOKEN',
    :css_ref => 'CSS',
    :js_ref => 'JS'
];

Note: Custom logo/js/css can be added together

No Blocking, Monitor Only Default mode: PxModule::MONITOR_MODE

PxModule::ACTIVE_MODE - Module blocks users crossing the predefined block threshold. Server-to-server requests are sent synchronously.
PxModule::MONITOR_MODE - Module does not block users crossing the predefined block threshold. The custom_block_handler function will be eval'd in case one is supplied, upon crossing the defined block threshold.

params[:module_mode] = PxModule::ACTIVE_MODE

Custom URI

Default: 'REQUEST_URI'

The URI can be returned to the PerimeterX module, using a custom user function, defined on the params variable

params[:custom_uri] = -> (request) {
  return request.headers['HTTP_X_CUSTOM_URI']
}

Filter sensitive headers A list of sensitive headers can be configured to prevent specific headers from being sent to PerimeterX servers (lower case header names). Filtering cookie headers for privacy is set by default, and can be overridden on the params variable.

Default: cookie, cookies

params[:sensitive_headers] = ['cookie', 'cookies', 'secret-header']

API Timeouts

Note: Controls the timeouts for PerimeterX requests. The API is called when a Risk Cookie does not exist, or is expired or invalid

The API Timeout, in seconds (int), to wait for the PerimeterX server API response.

Default: 1

params[:api_timeout] = 4

Send Page Activities Default: true A boolean flag to enable or disable sending of activities and metrics to PerimeterX on each page request. Enabling this feature will provide data that populates the PerimeterX portal with valuable information, such as the amount of requests blocked and additional API usage statistics.

params[:send_page_activities] = false

Additional Page Activity Handler

Adding an additional activity handler is done by setting additional_activity_handler with a user defined function on the params variable. The additional_activity_handler function will be executed before sending the data to the PerimeterX portal.

Default: Only send activity to PerimeterX as controlled by params.

params[:additional_activity_handler] = -> (activity_type, px_ctx, details){
    // user defined logic comes here
};

Debug Mode Default: false

Enables debug logging mode to STDOUT

  params[:debug] = true

Whitelist Routes Default: []

An array of route prefixes and/or regular expressions that are always whitelisted and not validated by PerimeterX. A string value of a path will be treated as a prefix. A regexp value of a path will be treated as is.

  params[:whitelist_routes] = ["/example", /\A\/example\z/]

Update Configuration on Runtime

As mentioned before, PerimeterX Module should be configured in <rails_app>/config/initializers/perimeterx.rb. However, it is possible to override configuration options on each request. To do so, send the configuration options as an argument when calling to px_verify_request as described in the following example. Notice that in case of an invalid argument, the module will raise an error. Therefore, when using this feature, make sure to wrap the call to px_verify_request with begin and rescue. It is highly recommended to log the error message to follow such errors.

Usage example:

class HomeController < ApplicationController
  include PxModule

  before_action do call_perimeterx_verify_request end

  def call_perimeterx_verify_request
    params = {
    :blocking_score => 70,
    :module_mode => 2
    }
    begin
      px_verify_request(params) 
    rescue StandardError => e
      # $stdout.write(e.message)
    end
  end

end

First Party

To enable first party on your enforcer, add the following routes to your config/routes.rb file:

  get '/:appid_postfix/init.js', to: 'home#index', constraints: { appid_postfix: /XXXXXXXX/ }
  get '/:appid_postfix/captcha/:all', to: 'home#index', constraints: { appid_postfix: /XXXXXXXX/, all:/.*/ }
  post '/:appid_postfix/xhr/:all', to: 'home#index', constraints: { appid_postfix: /XXXXXXXX/, all:/.*/  }

Notice that all occurences of XXXXXXXX should be replaced with your px_app_id without the "PX" prefix. For example, if your px_app_id is PX2H4seK9L, replace XXXXXXXX with 2H4seK9L.

In case you are using more than one px_app_id, provide all of them with a | sign between them. For example: 2H4seK9L|9bMs6K94|Lc5kPMNx

First Party configuration:

Default: true

  params[:first_party_enabled] = false

Additional Information

URI Delimiters

PerimeterX processes URI paths with general- and sub-delimiters according to RFC 3986. General delimiters (e.g., ?, #) are used to separate parts of the URI. Sub-delimiters (e.g., $, &) are not used to split the URI as they are considered valid characters in the URI path.

Contributing

The following steps are welcome when contributing to our project. ###Fork/Clone First and foremost, Create a fork of the repository, and clone it locally. Create a branch on your fork, preferably using a self descriptive branch name.

###Code/Run Help improve our project by implementing missing features, adding capabilities or fixing bugs.

To run the code, simply follow the steps in the installation guide. Grab the keys from the PerimeterX Portal, and try refreshing your page several times continously. If no default behaviours have been overriden, you should see the PerimeterX block page. Solve the CAPTCHA to clean yourself and start fresh again.

PerimeterX / perimeterx-ruby-sdk