Skip to content

Deploy using Docker

To use MindsDB's Docker Container you first need to have Docker installed on your machine. To make sure Docker is successfully installed on your machine, run:

docker run hello-world

You should see the Hello from Docker! message displayed. If not, check the Get Started documentation.

MindsDB container

MindsDB images are uploaded to the MindsDB repo on docker hub after each release.

Pull image

First, run the below command to pull our latest production image:

docker pull mindsdb/mindsdb

Or, to try out the latest beta version, pull the beta image:

docker pull mindsdb/mindsdb_beta

Publish ports

By default, when you run the MindsDB container, it does not publish any of its ports. To make the ports available you must run the container by providing -p flag as:

  • -p 47334:47334 - Map 47734 port which is used by the MindsDB GUI and the HTTP API.
  • -p 47335:47335 - Map 47335 to use MindsDB MySQL API.
  • -p 47336:47336 - Map 47336 port to use MindsDB MongoDB API.

Start container

Next, run the below command to start the container:

docker run -p 47334:47334 mindsdb/mindsdb

Docker run

That's it. MindsDB should automatically start the Studio on your default browser.

MKL Issues

Note. If you experience any issue related to MKL or if your training process does not complete, please add env var or starter Docker with this command:

docker run --env MKL_SERVICE_FORCE_INTEL=1 -it -p 47334:47334 mindsdb/mindsdb

Extend config.json

If you want to extend the default configuration, you will be able to send the config.json value as JSON string argument to the MDB_CONFIG_CONTENT as:

docker run -e MDB_CONFIG_CONTENT='{"api":{"http": {"host": "0.0.0.0","port": "47334"}}}' mindsdb/mindsdb

Or, you can pipe < the content of the file to the MDB_CONFIG_CONTENT.

Comments