Ddosify - High-performance load testing tool

         

Features

📌 Protocol Agnostic - Currently supporting HTTP, HTTPS, HTTP/2. Other protocols are on the way.

📌 Scenario-Based - Create your flow in a JSON file. Without a line of code!

📌 Different Load Types - Test your system's limits across different load types.

📌 Parameterization - Use dynamic variables just like on Postman.

📌 Correlation - Extract variables from earlier phases and pass them on to the following ones.

📌 Test Data - Import test data from CSV and use it in the scenario.

Installation

ddosify is available via Docker, Docker Extension, Homebrew Tap, and downloadable pre-compiled binaries from the releases page for macOS, Linux and Windows.

Docker

docker run -it --rm ddosify/ddosify

Docker Extension

Run Ddosify open-source on Docker Desktop with Ddosify Docker extension. More: https://hub.docker.com/extensions/ddosify/ddosify-docker-extension

Homebrew Tap (macOS and Linux)

brew install ddosify/tap/ddosify

apk, deb, rpm, Arch Linux, FreeBSD packages

• For arm architectures change ddosify_amd64 to ddosify_arm64 or ddosify_armv6.
• Superuser privilege is required.

# For Redhat based (Fedora, CentOS, RHEL, etc.)
rpm -i https://github.com/ddosify/ddosify/releases/latest/download/ddosify_amd64.rpm

# For Debian based (Ubuntu, Linux Mint, etc.)
wget https://github.com/ddosify/ddosify/releases/latest/download/ddosify_amd64.deb
dpkg -i ddosify_amd64.deb

# For Alpine
wget https://github.com/ddosify/ddosify/releases/latest/download/ddosify_amd64.apk
apk add --allow-untrusted ddosify_amd64.apk

# For Arch Linux
git clone https://aur.archlinux.org/ddosify.git
cd ddosify
makepkg -sri

# For FreeBSD
pkg install ddosify

Windows exe from the releases page

• Download *.zip file for your architecture. For example download ddosify version vx.x.x with amd64 architecture: ddosify_x.x.x.zip_windows_amd64
• Unzip ddosify_x.x.x_windows_amd64.zip
• Open Powershell or CMD (Command Prompt) and change directory to unzipped folder: ddosify_x.x.x_windows_amd64
• Run ddosify:

.\ddosify.exe -t http://target_site.com

Using the convenience script (macOS and Linux)

• The script requires root or sudo privileges to move ddosify binary to /usr/local/bin.
• The script attempts to detect your operating system (macOS or Linux) and architecture (arm64, x86, amd64) to download the appropriate binary from the releases page.
• By default, the script installs the latest version of ddosify.
• If you have problems, check common issues
• Required packages: curl and sudo

curl -sSfL https://raw.githubusercontent.com/ddosify/ddosify/master/scripts/install.sh | sh

Go install from source (macOS, FreeBSD, Linux, Windows)

Minimum supported Go version is 1.18

go install -v go.ddosify.com/ddosify@latest

Easy Start

This section aims to show you how to use Ddosify without deep dive into its details easily.

1. Simple load test

    ddosify -t http://target_site.com
   

   The above command runs a load test with the default value that is 100 requests in 10 seconds.

2. Using some of the features

    ddosify -t http://target_site.com -n 1000 -d 20 -m PUT -T 7 -P http://proxy_server.com:80
   

   Ddosify sends a total of 1000 PUT requests to https://target_site.com over proxy http://proxy_server.com:80 in 20 seconds with a timeout of 7 seconds per request.

3. Usage for CI/CD pipelines (JSON output)

    ddosify -t http://target_site.com -o stdout-json | jq .avg_duration
   

   Ddosify outputs the result in JSON format. Then jq (or any other command-line JSON processor) fetches the avg_duration. The rest depends on your CI/CD flow logic.

4. Scenario based load test

    ddosify -config config_examples/config.json
   

   Ddosify first sends HTTP/2 POST request to https://test_site1.com/endpoint_1 using basic auth credentials test_user:12345 over proxy http://proxy_host.com:proxy_port and with a timeout of 3 seconds. Once the response is received, HTTPS GET request will be sent to https://test_site1.com/endpoint_2 along with the payload included in config_examples/payload.txt file with a timeout of 2 seconds. This flow will be repeated 20 times in 5 seconds and response will be written to stdout.

5. Load test with Dynamic Variables (Parameterization)

    ddosify -t http://target_site.com/{{_randomInt}} -d 10 -n 100 -h 'User-Agent: {{_randomUserAgent}}' -b '{"city": "{{_randomCity}}"}'
   

   Ddosify sends a total of 100 GET requests to https://target_site.com/{{_randomInt}} in 10 seconds. {{_randomInt}} path generates random integers between 1 and 1000 in every request. Dynamic variables can be used in URL, headers, payload (body) and basic authentication. In this example, Ddosify generates a random user agent in the header and a random city in the body. The full list of the dynamic variables can be found in the docs.

6. Correlation (Captured Variables)

    ddosify -config ddosify_config_correlation.json
   

   Ddosify allows you to specify variables at the global level and use them throughout the scenario, as well as extract variables from previous steps and inject them to the next steps in each iteration individually. You can inject those variables in requests url, headers and payload(body). The example config can be found in correlation-config-example.

7. Test Data

    ddosify -config ddosify_data_csv.json
   

   Ddosify allows you to load test data from a file, tag specific columns for later use. You can inject those variables in requests url, headers and payload(body). The example config can be found in test-data-example.

Details

You can configure your load test by the CLI options or a config file. Config file supports more features than the CLI. For example, you can't create a scenario-based load test with CLI options.

CLI Flags

ddosify [FLAG]

Flag	Description	Type	Default	Required?
-t	Target website URL. Example: https://ddosify.com	string	-	Yes
-n	Total iteration count	int	100	No
-d	Test duration in seconds.	int	10	No
-m	Request method. Available methods for HTTP(s) are GET, POST, PUT, DELETE, HEAD, PATCH, OPTIONS	string	GET	No
-b	The payload of the network packet. AKA body for the HTTP.	string	-	No
-a	Basic authentication. Usage: -a username:password	string	-	No
-h	Headers of the request. You can provide multiple headers with multiple -h flag. Usage: -h 'Accept: text/html'	string	-	No
-T	Timeout of the request in seconds.	int	5	No
-P	Proxy address as host:port. -P 'http://user:pass@proxy_host.com:port'	string	-	No
-o	Test result output destination. Supported outputs are [stdout, stdout-json] Other output types will be added.	string	stdout	No
-l	Type of the load test. Ddosify supports 3 load types.	string	linear	No
--config	Config File of the load test.	string	-	No
--version	Prints version, git commit, built date (utc), go information and quit	-	-	No
--cert_path	A path to a certificate file (usually called 'cert.pem')	-	-	No
--cert_key_path	A path to a certificate key file (usually called 'key.pem')	-	-	No
--debug	Iterates the scenario once and prints curl-like verbose result. Note that this flag overrides json config.	bool	false	No

Load Types

Linear

ddosify -t http://target_site.com -l linear

Result:

Note: If the iteration count is too low for the given duration, the test might be finished earlier than you expect.

Incremental

ddosify -t http://target_site.com -l incremental

Result:

Waved

ddosify -t http://target_site.com -l waved

Result:

Config File

Config file lets you use all capabilities of Ddosify.

The features you can use by config file;

• Scenario creation
• Custom load type creation
• Payload from a file
• Multipart/form-data payload
• Extra connection configuration, like keep-alive enable/disable logic
• HTTP2 support

Usage;

ddosify -config <json_config_path>

There is an example config file at config_examples/config.json. This file contains all of the parameters you can use. Details of each parameter;

• iteration_count optional

  This is the equivalent of the -n flag. The difference is that if you have multiple steps in your scenario, this value represents the iteration count of the steps.

• load_type optional

  This is the equivalent of the -l flag.

• duration optional

  This is the equivalent of the -d flag.

• manual_load optional

  If you are looking for creating your own custom load type, you can use this feature. The example below says that Ddosify will run the scenario 5 times, 10 times, and 20 times, respectively along with the provided durations. iteration_count and duration will be auto-filled by Ddosify according to manual_load configuration. In this example, iteration_count will be 35 and the duration will be 18 seconds. Also manual_load overrides load_type if you provide both of them. As a result, you don't need to provide these 3 parameters when using manual_load.

  "manual_load": [
      {"duration": 5, "count": 5},
      {"duration": 6, "count": 10},
      {"duration": 7, "count": 20}
  ]

• proxy optional

  This is the equivalent of the -P flag.

• output optional

  This is the equivalent of the -o flag.

• env optional Scenario-scoped global variables. Note that dynamic variables changes every iteration.

  "env": {
          "COMPANY_NAME" :"Ddosify",
          "randomCountry" : "{{_randomCountry}}"
  }

• data optional Config for loading test data from a csv file. Csv data used in below config.

  "data":{
      "info": {
          "path" : "config/config_testdata/test.csv",
          "delimiter": ";",
          "vars": {
                  "0":{"tag":"name"},
                  "1":{"tag":"city"},
                  "2":{"tag":"team"},
                  "3":{"tag":"payload", "type":"json"},
                  "4":{"tag":"age", "type":"int"}
                  },
          "allow_quota" : true,
          "order": "sequential",
          "skip_first_line" : true,
          "skip_empty_line" : true
      }
  }

  Field	Description	Type	Default	Required?
  path	Local path or remote url for your csv file	string	-	Yes
  delimiter	Delimiter for reading csv	string	,	No
  vars	Tag columns using column index as key, use type field if you want to cast a column to a specific type, default is string, can be one of the following: json, int, float,bool.	map	-	Yes
  allow_quota	If set to true, a quote may appear in an unquoted field and a non-doubled quote may appear in a quoted field	bool	false	No
  order	Order of reading records from csv. Can be random or sequential	string	random	No
  skip_first_line	Skips first line while reading records from csv.	bool	false	No
  skip_empty_line	Skips empty lines while reading records from csv.	bool	true	No

• steps mandatory

  This parameter lets you create your scenario. Ddosify runs the provided steps, respectively. For the given example file step id: 2 will be executed immediately after the response of step id: 1 is received. The order of the execution is the same as the order of the steps in the config file.

  Details of each parameter for a step;

  • id mandatory

    Each step must have a unique integer id.

  • url mandatory

    This is the equivalent of the -t flag.

  • name optional

    Name of the step.

  • method optional

    This is the equivalent of the -m flag.

  • headers optional

    List of headers with key:value format.

  • payload optional

    This is the equivalent of the -b flag.

  • payload_file optional

    If you need a long payload, we suggest using this parameter instead of payload.

  • payload_multipart optional

    Use this for multipart/form-data Content-Type.

    Accepts list of form-field objects, structured as below;

    {
        "name": [field-name],
        "value": [field-value|file-path|url],
        "type": <text|file>,    // Default "text"
        "src": <local|remote>   // Default "local"
    }

    Example: Sending form name-value pairs;

    "payload_multipart": [
        {
            "name": "[field-name]",
            "value": "[field-value]"
        }
    ]

    Example: Sending form name-value pairs and a local file;

    "payload_multipart": [
        {
            "name": "[field-name]",
            "value": "[field-value]",
        },
        {
            "name": "[field-name]",
            "value": "./test.png",
            "type": "file"
        }
    ]

    Example: Sending form name-value pairs and a local file and a remote file;

    "payload_multipart": [
        {
            "name": "[field-name]",
            "value": "[field-value]",
        },
        {
            "name": "[field-name]",
            "value": "./test.png",
            "type": "file"
        },
        {
            "name": "[field-name]",
            "value": "http://test.com/test.png",
            "type": "file",
            "src": "remote"
        }
    ]

    Note: Ddosify adds Content-Type: multipart/form-data; boundary=[generated-boundary-value] header to the request when using payload_multipart.

  • timeout optional

    This is the equivalent of the -T flag.

  • capture_env optional

    Config for extraction of variables to use them in next steps. Example: Capture NUM variable from steps response body;

    "steps": [
        {
            "id": 1,
            "url": "http://target.com/endpoint1",
            "capture_env": {
                 "NUM" :{"from":"body","jsonPath":"num"},
            }
        },
    ]

  • sleep optional

    Sleep duration(ms) before executing the next step. Can be an exact duration or a range.

    Example: Sleep 1000ms after step-1;

    "steps": [
        {
            "id": 1,
            "url": "http://target.com/endpoint1",
            "sleep": "1000"
        },
        {
            "id": 2,
            "url": "http://target.com/endpoint2",
        }
    ]

    Example: Sleep between 300ms-500ms after step-1;

    "steps": [
        {
            "id": 1,
            "url": "http://target.com/endpoint1",
            "sleep": "300-500"
        },
        {
            "id": 2,
            "url": "http://target.com/endpoint2",
        }
    ]

  • auth optional

    Basic authentication.

    "auth": {
        "username": "test_user",
        "password": "12345"
    }

  • others optional

    This parameter accepts dynamic key: value pairs to configure connection details of the protocol in use.

    "others": {
        "keep-alive": true,              // Default true
        "disable-compression": false,    // Default true
        "h2": true,                      // Enables HTTP/2. Default false.
        "disable-redirect": true         // Default false
    }

Parameterization (Dynamic Variables)

Just like the Postman, Ddosify supports parameterization (dynamic variables) on URL, headers, payload (body) and basic authentication. Actually, we support all the random methods Postman supports. If you use {{$randomVariable}} on Postman you can use it as {{_randomVariable}} on Ddosify. Just change $ to _ and you will be fine. To simulate a realistic load test on your system, Ddosify can send every request with dynamic variables.

The full list of dynamic variables can be found in the Ddosify Docs.

Parameterization on URL

Ddosify sends 100 GET requests in 10 seconds with random string key parameter. This approach can be also used in cache bypass.

ddosify -t http://target_site.com/?key={{_randomString}} -d 10 -n 100

Parameterization on Headers

Ddosify sends 100 GET requests in 10 seconds with random Transaction-Type and Country headers.

ddosify -t http://target_site.com -d 10 -n 100 -h 'Transaction-Type: {{_randomTransactionType}}' -h 'Country: {{_randomCountry}}'

Parameterization on Payload (Body)

Ddosify sends 100 GET requests in 10 seconds with random latitude and longitude values in body.

ddosify -t http://target_site.com -d 10 -n 100 -b '{"latitude": "{{_randomLatitude}}", "longitude": "{{_randomLongitude}}"}'

Parameterization on Basic Authentication

Ddosify sends 100 GET requests in 10 seconds with random username and password with basic authentication.

ddosify -t http://target_site.com -d 10 -n 100 -a '{{_randomUserName}}:{{_randomPassword}}'

Parameterization on Config File

Dynamic variables can be used on config file as well. Ddosify sends 100 GET requests in 10 seconds with random string key parameter in URL and random User-Key header.

ddosify -config ddosify_config_dynamic.json

{
    "iteration_count": 100,
    "load_type": "linear",
    "duration": 10,
    "steps": [
        {
            "id": 1,
            "url": "https://test_site1.com/?key={{_randomString}}",
            "method": "POST",
            "headers": {
                "User-Key": "{{_randomInt}}"
            }
        }
    ]
}

Correlation

Ddosify enables you to capture variables from steps using jsonPath, xpath, or regular expressions. Later, in the subsequent steps, you can inject both the captured variables and the scenario-scoped global variables.

⚠️ Points to keep in mind

• You must specify 'header_key' when capturing from header.
• For jsonPath syntax, please take a look at gjson syntax doc.
• Regular expression are expected in 'Golang' style regex. For converting your existing regular expressions, you can use regex101.

You can use debug parameter to validate your config.

ddosify -config ddosify_config_correlation.json -debug

Capture With JsonPath

{
    "steps": [
        {
            "capture_env": {
                "NUM" :{"from":"body","jsonPath":"num"},
                "NAME" :{"from":"body","jsonPath":"name"},
                "SQUAD" :{"from":"body","jsonPath":"squad"},
                "PLAYERS" :{"from":"body","jsonPath":"squad.players"},
                "MESSI" : {"from":"body","jsonPath":"squad.players.0"},             
            }         
        }
    ]
}

Capture With XPath

{
    "steps": [
        {
            "capture_env": {
                "TITLE" :{"from":"body","xpath":"//item/title"},             
            }         
        }
    ]
}

Capture With Regular Expressions

{
    "steps": [
        {
            "capture_env": {
               "CONTENT_TYPE" :{"from":"header", "header_key":"Content-Type" ,"regexp":{"exp":"application\/(\\w)+","matchNo":0}} ,
               "REGEX_MATCH_ENV" :{"from":"body","regexp":{"exp" : "[a-z]+_[0-9]+", "matchNo": 1}}          
            }         
        }
    ]
}

Capture Header Value

{
    "steps": [
        {
            "capture_env": {
                "TOKEN" :{"from":"header", "header_key":"Authorization"},
            }         
        }
    ]
}

Scenario-Scoped Variables

{
   "env":{
        "TARGET_URL" : "http://localhost:8084/hello",
        "USER_KEY" : "ABC",
        "COMPANY_NAME" : "Ddosify",
        "RANDOM_COUNTRY" : "{{_randomCountry}}",
        "NUMBERS" : [22,33,10,52] 
    },
}

🔨 Overall Config and Injection

On array-like captured variables or environment vars, the rand( ) function can be utilized.

// ddosify_config_correlation.json
{
    "iteration_count": 100,
    "load_type": "linear",
    "duration": 10,
    "steps": [
        {
            "id": 1,
            "url": "{{TARGET_URL}}",
            "method": "POST",
            "headers": {
                "User-Key": "{{USER_KEY}}",
                "Rand-Selected-Num" : "{{rand(NUMBERS)}}"
            },
            "payload" : "{{COMPANY_NAME}}",
            "capture_env": {
                "NUM" :{"from":"body","jsonPath":"num"},
                "NAME" :{"from":"body","jsonPath":"name"},
                "SQUAD" :{"from":"body","jsonPath":"squad"},
                "PLAYERS" :{"from":"body","jsonPath":"squad.players"},
                "MESSI" : {"from":"body","jsonPath":"squad.players.0"},
                "TOKEN" :{"from":"header", "header_key":"Authorization"},
                "CONTENT_TYPE" :{"from":"header", "header_key":"Content-Type" ,"regexp":{"exp":"application\/(\\w)+","matchNo":0}}             
            }         
        },
        {
            "id": 2,
            "url": "{{TARGET_URL}}",
            "method": "POST",
            "headers": {
                "User-Key": "{{USER_KEY}}",
                "Authorization": "{{TOKEN}}",
                "Content-Type" : "{{CONTENT_TYPE}}"
            },
            "payload_file" : "payload.json",
            "capture_env": {
                "TITLE" :{"from":"body","xpath":"//item/title"},
                "REGEX_MATCH_ENV" :{"from":"body","regexp":{"exp" : "[a-z]+_[0-9]+", "matchNo": 1}}
            }
        }
    ],
    "env":{
        "TARGET_URL" : "http://localhost:8084/hello",
        "USER_KEY" : "ABC",
        "COMPANY_NAME" : "Ddosify",
        "RANDOM_COUNTRY" : "{{_randomCountry}}",
        "NUMBERS" : [22,33,10,52] 
    },
}

// payload.json
{
    "boolField" : "{{_randomBoolean}}",
    "numField" : "{{NUM}}",
    "strField" : "{{NAME}}",
    "numArrayField" : ["{{NUM}}",34],
    "strArrayField" : ["{{NAME}}","hello"],
    "mixedArrayField" : ["{{NUM}}",34,"{{NAME}}","{{SQUAD}}"],
    "{{NAME}}" : "messi",
    "obj" :{
        "numField" : "{{NUM}}",
        "objectField" : "{{SQUAD}}",
        "arrayField" : "{{PLAYERS}}"
    }
}

Test Data Set

Ddosify enables you to load test data from csv files. Later, in your scenario, you can inject variables that you tagged.

We are using this csv data in below config.

// config_data_csv.json
"data":{
      "csv_test": {
          "path" : "config/config_testdata/test.csv",
          "delimiter": ";",
          "vars": {
                  "0":{"tag":"name"},
                  "1":{"tag":"city"},
                  "2":{"tag":"team"},
                  "3":{"tag":"payload", "type":"json"},
                  "4":{"tag":"age", "type":"int"}
                },
          "allow_quota" : true,
          "order": "random",
          "skip_first_line" : true
      }
    }

You can refer to tagged variables in your request like below.

// payload.json
{
    "name" : "{{data.csv_test.name}}",
    "team" : "{{data.csv_test.team}}",
    "city" : "{{data.csv_test.city}}",
    "payload" : "{{data.csv_test.payload}}",
    "age" : "{{data.csv_test.age}}"
}

Tutorials / Blog Posts

• Testing the Performance of User Authentication Flow

Common Issues

macOS Security Issue

"ddosify" can’t be opened because Apple cannot check it for malicious software.

• Open /usr/local/bin
• Right click ddosify and select Open
• Select Open
• Close the opened terminal

Communication

You can join our Discord Server for issues, feature requests, feedbacks or anything else.

More

This repository includes the single-node version of the Ddosify Loader. For distributed and Geo-targeted Load Testing you can use Ddosify Cloud

Disclaimer

Ddosify is created for testing the performance of web applications. Users must be the owner of the target system. Using it for harmful purposes is extremely forbidden. Ddosify team & company is not responsible for its’ usages and consequences.

License

Licensed under the AGPLv3: https://www.gnu.org/licenses/agpl-3.0.html