Force Deletion of Connectors

If you must delete a connector and related information before the process of waiting for the background jobs to run, you can run a deletion script using the instructions that follow.

This is not default behavior. Use sparingly

How to find a connector ID?

  1. Navigate to the connectors page.

Connectors page

  1. Click on the connector you want the ID of .

Connector ID

  1. On the connector’s page, the URL’s final number will be the ID of the connector. For example, in this example, the connector’s ID is 2. Here is also a more zoomed in photo of the URL bar.

Zoomed in Connector IDm

Deleting Locally

  1. Gather the ID of the connector you wish to delete.

  2. Navigate to the base folder of your Danswer repository.

  3. Run python3 backend/scripts/ <CONNECTOR_ID>

Deleting in Docker Compose

Follow these steps to manually and immediately delete a connector and its contents within a container:

  1. Ensure your Docker containers are running- you can refer to the quickstart guide:

    docker compose -f -p danswer-stack up -d --build --force-recreate
  2. Identify the <CONTAINER_ID> of your API server container. It should contain the string danswer-api. You can list all running containers with:

    docker ps -a
  3. Gather the ID <CONNECTOR_ID> of the connector you wish to delete.

  4. Run the deletion script inside the Docker container:

    docker exec -it <CONTAINER_ID> python scripts/ <CONNECTOR_ID>

    Replace <CONTAINER_ID> with your actual API container ID, and <CONNECTOR_ID> with the ID of the connector you want to delete.

    For example:

    docker exec -it 92492089ada1 python scripts/ 10
  5. The script will execute inside the container, accessing the necessary resources and performing the deletion.


  • If you encounter permission issues, you may need to run the docker-compose command with sudo:

    sudo docker-compose exec <CONTAINER_ID> python /app/scripts/ `CONNECTOR_ID`
  • Ensure that your user has the necessary permissions to execute docker-compose commands.

  • If the script is not found, verify the path to the script within your Docker container.