Tired of using the JMeter GUI or looking at hairy XML files?

This gem lets you write test plans for JMeter in your favourite text editor, and optionally run them on flood.io.

Code Quality Rank: L5
Monthly Downloads: 26,299
Programming language: Ruby
License: MIT License
Tags: Testing     Extra     Projects    
Latest version: v3.1.08

Ruby-JMeter alternatives and similar gems

Based on the "Extra" category.
Alternatively, view Ruby-JMeter alternatives based on common mentions on social networks and blogs.

Do you think we are missing an alternative of Ruby-JMeter or a related project?

Add another 'Extra' Gem


Ruby-JMeter Build Status Code Climate Gem Version

Ruby-JMeter is built and maintained by Flood IO, an easy to use load testing platform for any scale of testing.

Tired of using the JMeter GUI or looking at hairy XML files?

This gem lets you write test plans for JMeter in your favourite text editor, and optionally run them on flood.io.


Install it yourself as:

$ gem install ruby-jmeter

Basic Usage

RubyJmeter exposes easy-to-use domain specific language for fluent communication with JMeter.It also includes API integration with flood.io, a cloud based load testing service.

To use the DSL, first let's require the gem:

require 'ruby-jmeter'

Basic Example

Let's create a test and save the related jmx testplan to file, so we can edit/view it in JMeter.

test do
  threads count: 10 do
    visit name: 'Google Search', url: 'http://google.com'

So in this example, we just created a test plan, with 10 threads, each of which visited the search page at Google.

Generating a JMeter Test Plan (JMX)

Note also how we called the jmx method of the test. Calling this method will write the contents of the JMeter test plan to file like this.

$ ruby testplan.rb
[2013-04-23T10:29:03.275743 #42060]  INFO -- : Test plan saved to: jmeter.jmx
<?xml version="1.0" encoding="UTF-8"?>
<jmeterTestPlan version="1.2" properties="2.1">
    <TestPlan guiclass="TestPlanGui" testclass="TestPlan" testname="Test Plan" enabled="true">
JMX saved to: jmeter.jmx

The file that is created can then be executed in the JMeter GUI. If you want to create the file with a different filename and/or path, just add the file parameter to the jmx method call like this.

test do
  threads count: 10 do
    visit name: 'Google Search', url: 'http://google.com'
end.jmx(file: "/tmp/my_testplan.jmx")

Windows users should specify a path like this.

.jmx(file: "C:\\TEMP\\MyTestplan.jmx")

Running a JMeter Test Plan locally

You can execute the JMeter test plan by calling the run method of the test like this.

test do
  threads count: 10 do
    visit name: 'Google Search', url: 'http://google.com'

This will launch JMeter in headless (non-GUI mode) and execute the test plan. This is useful for shaking out the script before you push it to the Grid. There are a few parameters that you can set such as the path to the JMeter binary, the file path/name for the JMX file, the log path/name to output JMeter logs, the jtl path/name for JMeter results like this, and the properties path/name for the additional JMeter property file.

test do
  threads count: 10 do
    visit name: 'Google Search', url: 'http://google.com'
  path: '/usr/share/jmeter/bin/',
  file: 'jmeter.jmx',
  log: 'jmeter.log',
  jtl: 'results.jtl',
  properties: 'jmeter.properties')

Running a JMeter Test Plan on Flood IO

You can also execute JMeter test plans on Flood IO using our API. To do so, you require an account and API token. If you don't know your token, sign in to flood.io and check your account settings.

To execute the test on Flood IO, call the flood method on the test and pass it the API token like this.

test do
  threads count: 10 do
    visit name: 'Google Search', url: 'http://google.com'
  name: 'Demo',
  privacy: 'public',
  ## Select a grid or region to distribute this flood to
  # grid: 'pmmi24XaSMnKjGVEtutroQ',
  # region: 'ap-southeast-2'

This will then provide you with a link to the live test results on Flood IO like this.

I, [2015-02-24T11:15:25.669029 #14010]  INFO -- : Flood results at: https://flood.io/AbRWkFl7VODYCkQuy3ffvA

Note you will need to provide a grid or region parameter to the .flood method to describe which grid to distribute the flood test to. Otherwise it will default to the shared grid. You can find the Grid ID from the URL of the target grid in your grids dashboard e.g.:

Flood IO provides a shared grid for free, suitable for 5 minute tests, check your dashboard for the latest grid:

Alternatively upgrade to a paid subscription on Flood IO and start your own grids on demand.

Advanced Usage


Each of the methods take an optional block delimited by do and end or braces {}

Blocks let you nest methods within methods, so you can scope the execution of methods as you would in a normal JMeter test plan. For example.

test do
  threads count: 100 do
    visit name: 'Home', url: 'http://altentee.com' do
      extract regex: "content='(.+?)' name='csrf-token'", name: 'csrf-token'

This would create a new test plan, with a 100 user thread group, each user visiting the "Home" page and extracting the CSRF token from the response of each visit.

All methods are nestable, but you should only have one test method, and typically only one threads method. For example, it wouldn't make sense to have a test plan within a test plan, or a thread group within a thread group. You can have multiple thread groups per test plan though. This implies some knowlege of how JMeter works.

All methods take a parameter hash to configure related options.


You can use the threads method to define a group of users:

threads count: 100
threads count: 100, continue_forever: true
threads count: 100, loops: 10
threads count: 100, rampup: 30, duration: 60
threads count: 100, scheduler: true,
  start_time: Time.now.to_i * 1000,
  end_time:   (Time.now.to_i * 1000) + (3600 * 1000)


You can use the cookies method to define a Cookie Manager:

test do

This methods takes an optional parameters hash. This is based on the HTTP Cookie Manager.

test do
  cookies clear_each_iteration: false

test do
  cookies policy: 'rfc2109', clear_each_iteration: true
User-Defined Cookies

The cookies method parameters hash supports user_defined_cookies:

test do
  cookie1 = { value: 'foo', name: 'bar', domain: 'google.co.uk', path: '/' }
  cookie2 = { value: 'hello', name: 'world', domain: 'google.co.uk', secure: true }

  cookies user_defined_cookies: [ cookie1, cookie2 ]

name and value are required. domain and path are optional and default to blank. secure is optional and defaults to false.


You can use the cache method to define a Cache Manager:

test do

This methods takes an optional parameters hash. This is based on the HTTP Cache Manager.

test do
  cache clear_each_iteration: false

test do
  cache use_expires: true, clear_each_iteration: true


You can use the auth method to define an Authorization Manager:

test do

This methods takes an optional parameters hash. This is based on the HTTP Authorization Manager.

test do
  auth url: '/', username: 'tim', password: 'secret', domain: 'altentee.com'


You can use the visit method to navigate to pages:

visit name: 'Google Search', url: 'http://google.com'
visit name: 'Google Search', url: 'http://google.com'
visit name: 'Google Search', url: 'http://google.com',
  method: 'POST',
visit name: 'Google Search', url: 'http://google.com',
  use_keepalive: 'false'
visit name: 'Google Search', url: 'http://google.com',
  connect_timeout: '1000',
  response_timeout: '60000'
visit name: 'View Login', url: '/login',
  protocol: "https",
  port: 443

Submitting a Form

You can use the submit method to POST a HTTP form:

submit name: 'Submit Form', url: 'http://altentee.com/',
  fill_in: {
    username: 'tim',
    password: 'password',
    'csrf-token' => '${csrf-token}'

This method makes a single request. The fill_in parameter lets you specify key/value pairs for form field parameters. You can also use the built in JMeter ${expression} language to access run time variables extracted from previous responses.


  header [
    { name: 'Content-Type', value: 'application/json' }

  person = { name: "Tom" }

  post name: 'Create Person',
        url: "https://example.com/people.json",
        raw_body: person.to_json do

Think Time

You can use the think_time method to insert pauses into the simulation. This method is aliased as random_timer.

think_time 3000

This method takes 2 parameters: the constant delay, and an optional variable delay. Both are specified in milliseconds. This is based on the Gaussian Random Timer. This timer pauses each thread request for a random amount of time, with most of the time intervals ocurring near a particular value. The total delay is the sum of the Gaussian distributed value (with mean 0.0 and standard deviation 1.0) times the deviation value you specify, and the offset value.

# constant delay of 3 seconds
think_time 3000
# constant delay of 1 seconds with variance up to 6 seconds.
random_timer 1000, 6000

Response Extractor

You can use the extract method to extract values from a server response using a regular expression. This is aliased as the web_reg_save_param method. This method is typically used inside a visit or submit block.

extract regex: "content='(.+?)' name='csrf-token'", name: 'csrf-token'

visit name: 'Google', url: "http://google.com/" do
  extract regex: 'aria-label="(.+?)"', name: 'button_text'
  extract xpath: '//button', name: 'button'

This is based on the Regular Expression Extractor and XPath Extractor

visit name: "Altentee", url: "http://altentee.com" do
  extract regex: "content='(.+?)' name='csrf-token'", name: 'csrf-token'
  extract regex: 'value="(.+?)" name="JESSIONSID"', name: 'JSESSIONID'
  web_reg_save_param regex: 'value="(.+?)" name="VIEWSTATE"', name: 'VIEWSTATE'
  extract name: 'username', regex: 'value="(.+?)", name="username"',
    default: 'Tim Koopmans',
    match_number: 1
  extract name: 'shopping_item', regex: 'id="(.+?)" name="book"',
    match_number: 0 # random

You can later use the extracted values with subsequent requests:

post name: 'Authenticate', url: 'http://example.com/api/authentication/facebook', raw_body: '{"auth_token": "FB_TOKEN"}' do
  extract name: 'auth_token', regex: %q{.*"token":"([^"]+)".*}
  extract name: 'user_id', regex: %q{.*"user_id":([^,]+),.*}

header({name: 'X-Auth-Token', value: '${auth_token}'})
visit name: 'User profile', url: 'http://example.com/api/users/${user_id}'

Response Assertion

You can use the assert method to extract values from a server response using a regular expression. This is aliased as the web_reg_find method. This method is typically used inside a visit or submit block.

visit "Altentee", "http://altentee.com" do
  assert contains: "We test, tune and secure your site"

This method takes 3 parameters: the matching rule, the test string, and an optional parameters hash. This is based on the Response Assertion.

visit "Altentee", "http://altentee.com" do
  assert "contains": "We test, tune and secure your site"
  assert "not-contains": "We price gouge on cloud services"
  assert "matches": "genius"
  assert "not-matches": "fakers"
  assert "contains": "magic"
  assert "not-contains": "unicorns", scope: 'all'


This work is being sponsored by Flood IO. Get in touch with us if you'd like to be involved.


  1. Fork it
  2. Create your feature branch (git checkout -b my-new-feature)
  3. Create some specs, make them pass
  4. Commit your changes (git commit -am 'Add some feature')
  5. Push to the branch (git push origin my-new-feature)
  6. Create new Pull Request


We use an interface description language (IDL) as a bridge between JMeter's components expressed as XML in a .jmx test plan to Ruby's DSL objects in this repository.

To automate this lib/ruby-jmeter/idl.rb can be executed from the command line which will read from lib/ruby-jmeter/idl.xml and output to lib/ruby-jmeter/dsl

For example:

flood/ruby-jmeter - [master●] » ruby lib/ruby-jmeter/idl.rb
flood/ruby-jmeter - [master●] » git status
On branch master
Your branch is up-to-date with 'origin/master'.
Changes not staged for commit:
  (use "git add <file>..." to update what will be committed)
  (use "git checkout -- <file>..." to discard changes in working directory)

  modified:   lib/ruby-jmeter/DSL.md
  modified:   lib/ruby-jmeter/dsl/foreach_controller.rb
  modified:   lib/ruby-jmeter/dsl/http_request.rb
  modified:   lib/ruby-jmeter/dsl/http_request_defaults.rb
  modified:   lib/ruby-jmeter/dsl/regular_expression_extractor.rb
  modified:   lib/ruby-jmeter/dsl/response_assertion.rb
  modified:   lib/ruby-jmeter/dsl/test_fragment.rb
  modified:   lib/ruby-jmeter/dsl/user_parameters.rb

You should never manually update code in lib/ruby-jmeter/dsl as this is automatically overwritten whenever we run the IDL script. As new components / plugins are added, or major versions of JMeter are updated, we open lib/ruby-jmeter/idl.xml in the JMeter UI with those updates prior to running the IDL script. This makes updating between versions more easy.


Much of the behaviour of the gem is defined in lib/ruby-jmeter/dsl.rb which is where you should be updating code. You can extend individual DSL component behaviour in live/ruby-jmeter/extend/**/*.rb


Some custom code has been contributed particularly for support of JMeter plugins. These are not included in the IDL and as such should be added to lib/ruby-jmeter/plugins. Please follow some of the other examples.


We recommend using the Ruby gem bundle to manage your dependencies. Typical setup would be:

gem install bundler
cd <your local clone>
bundle install

Then you can run any rake / test tasks with the prefix bundle exec


If contributing please add an appropriate test. See spec/*_spec.rb for examples. Tests can be run from the command line as follows:

$ bundle exec rspec


It is often useful to add an appropriate example for other users and for testing your changes locally with the JMeter UI. See examples for different types of examples. To let your examples work locally from your own changes / commits simply prefix the examples with:

$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
$ flood/ruby-jmeter - [master●] » ruby examples/basic_assertion.rb
  W, [2015-10-17T19:31:12.021004 #33216]  WARN -- : Test executing locally ...

Note: most of the examples assume the JMeter binary is installed in /usr/share/jmeter/bin/ however you can modify this in your example to something that suits your installation e.g.:

end.run(path: 'C/Program Files/JMeter/bin/', gui: true)