Control Plane Setup
Step-by-step guide on setting up your self-hosted control plane using the Control Plane Lite utility.
RudderStack's control plane manages the configuration of your sources and destinations.
For more information on the control plane, refer to RudderStack's Architecture.
The easiest way to manage these configurations is through RudderStack's self-hosted control plane. It is free, requires no setup, and has some useful features like Live Events and Transformations.
If you don't wish to sign up for RudderStack and want to self-host these configurations instead, you can use the open-source Control Plane Lite utility to set up your own control plane. You can then manage the source and destination configurations locally by exporting or importing from a JSON file.
The control plane set up using Control Plane Lite does not support Live Events and Transformations.
Setting up the control plane
To set up your self-hosted control plane using the Control Plane Lite utility, follow these steps:
Make sure you have Node.js installed before setting up your self-hosted control plane.
Clone the RudderStack Control Plane Lite repository.
Open your terminal and navigate to the Config Generator folder. Then, run the following commands:
Upon successful setup, you can access the dashboard at
http://localhost:3000
by default.
Exporting workspace configuration
After adding the required sources and destinations in the dashboard, you can export your workspace configuration by clicking on the EXPORT button. The configuration is then exported and saved as a JSON file. This file is required to start the RudderStack server.
Starting RudderStack server with the workspace configuration file
For RudderStack to pick up the exported workspace configuration file, follow the steps for your preferred RudderStack setup.
Docker
Download and open
rudder-docker.yml
.In the
environment
section underbackend
service, uncomment the following lines:
Then, replace
<workspace_config_filepath_in_container>
in the above line with your container file path. By default, you can set it to/etc/rudderstack/workspaceConfig.json
.In the
volumes
section under thebackend
service, uncomment the following line:
Next, replace
<absolute_path_to_workspace_config>
with the local path of yourworkspaceConfig.json
(where your workspace configuration file is saved locally). Also, replace<workspace_config_filepath_in_container>
with the container file path that you set in Step 3.
The value for <workspace_config_filepath_in_container>
should be the same as the value provided for the RSERVER_BACKEND_CONFIG_CONFIG_JSONPATH
variable. Otherwise, your workspace configuration will not be loaded and you will get an error.
At this stage, your
rudder-docker.yml
should look like the following:
Finally, navigate to the directory where you want to install RudderStack and run the following command:
Once you have completed these steps above successfully, send test events to verify your installation.
Kubernetes
Clone the repository containing the RudderStack Helm chart by running the following command:
Navigate to the folder containing the Helm chart, as shown:
Open the
values.yaml
file.Set the parameter
controlPlaneJSON
totrue
.Export the workspace configuration from the dashboard by following the steps in the Exporting workspace configuration section above. Place the exported file in the
rudderstack-helm
folder.Finally, run the following command:
Refer to the Configurable parameters section for information on the parameters that can be configured during deployment.
Once you have completed these steps above successfully, send test events to verify your installation.
Developer machine setup
First, set up the database in your preferred directory using the following commands:
Next, clone the RudderStack server repository.
Then, run
git submodule init
andgit submodule update
to fetch therudder-transformer
repository.Next, navigate to the rudder-transformer directory:
Install the dependencies using the command
npm i
. Then, start the destination transformer using the following command:
Navigate back to the main directory using the command
cd rudder-server
.Next, copy the
sample.env
to the main directory, as shown:
Then, go to the
config
folder and openconfig.yaml
.Under
[BackendConfig]
, look forconfigFromFile
and set it totrue
.Also, change the value of
configJSONPath
to the local path of yourworkspaceConfig.json
(where your workspace configuration file is saved locally), as shown:
Finally, run the RudderStack server using the following command:
Once you have completed these steps above successfully, send test events to verify the installation.
Using hosted control plane for device mode destinations
RudderStack's web and mobile SDKs expect the source configurations and the associated device mode destination configurations for initialization. If you are self-hosting the control plane, you will need to host these configurations on your own server such that the source configuration is available at <CONTROL_PLANE_URL>/sourceConfig
.
To know more about the difference between cloud mode and device mode in RudderStack, read the RudderStack Connection Modes guide.
For RudderStack Cloud, the SDKs fetch this source configuration from https://api.rudderlabs.com/sourceConfig
.
Control plane URL
Since the RudderStack SDKs need the source configuration and the backend/SDKs are not aware of it by default, you must serve the configuration on your web servers and point the SDKs to the same. This configuration must be available at <CONTROL_PLANE_URL>/sourceConfig
. To do so, follow these steps:
Set up your control plane using the Control Plane Lite utility.
Go to the dashboard, set up your source, and export the source configuration by clicking the EXPORT SOURCE CONFIG button, as shown:
Host the exported file on your own server at
/sourceConfig
. Provide the base URL of your server that is serving this file in the SDK initialization code snippet. For more information, refer to the following sections in the respective SDK guides:
The RudderStack SDKs fetch the configuration by appending the /sourceConfig
path to the base URL provided above.
The SDKs will then fetch the config from <CONTROL_PLANE_URL>/sourceConfig
. A sample exported source configuration is shown below:
FAQs
For a self-hosted environment, how do I obtain the control plane URL?
To use the control plane URL to initialize your SDKs, follow these steps:
Set up the control plane using the Control Plane Lite utility.
Go to dashboard, configure the source, and export the source configuration by clicking the EXPORT SOURCE CONFIG button as shown:
Host the exported file on your own server such that the configuration is available at
<CONTROL_PLANE_URL>/sourceConfig
.
This solution assumes that you have already set up the RudderStack data plane (backend) locally.
Contact us
If you across any issues while setting up your self-hosted control plane using the Control Plane Lite utility, contact us or start a conversation in our Slack community.
Last updated