How to Stop the Memgraph Daemon for Code-Graph-RAG

Use docker compose -f codebase_rag/docker-compose.yaml stop memgraph to gracefully halt only the Memgraph container while keeping other services running, or run docker compose -f codebase_rag/docker-compose.yaml down to tear down the entire Code-Graph-RAG stack.

The Code-Graph-RAG project by vitali87 orchestrates its graph database layer through Docker Compose, defining the Memgraph daemon as a containerized service in codebase_rag/docker-compose.yaml. Properly stopping this service ensures that your graph data persists correctly and system resources are released cleanly.

Stop Only the Memgraph Service

To halt just the Memgraph daemon without affecting other components like Qdrant or the Lab interface, target the specific service name.

Run the following command from the repository root:

docker compose -f codebase_rag/docker-compose.yaml stop memgraph

For legacy Docker Compose versions, use docker-compose (with a hyphen) instead of docker compose.

This command sends a SIGTERM signal to the Memgraph container, allowing the database to shut down gracefully while preserving the container configuration and persistent volumes. Other services defined in the compose file remain active.

Shut Down the Entire Code-Graph-RAG Stack

When you need to stop all services and free network resources, use the down command:

docker compose -f codebase_rag/docker-compose.yaml down

This stops the Memgraph container, removes the associated Docker network, and cleans up ephemeral resources. To also delete the persistent graph data stored in Docker volumes—effectively wiping the database—add the --volumes flag:

docker compose -f codebase_rag/docker-compose.yaml down --volumes

How the Shutdown Works Under the Hood

The Docker Compose configuration in codebase_rag/docker-compose.yaml declares a service named memgraph that pulls the image memgraph/memgraph-platform and exposes ports 7687 (Bolt protocol) and 7444 (web interface). When you execute docker compose up (or the repository's cgr up helper), Docker creates a container from this image and starts the Memgraph daemon inside it.

According to the source code in codebase_rag/stack/constants.py, the compose filename is stored as a constant used by CLI utilities, ensuring that commands reference the correct configuration file path. When you invoke stop, Docker initiates a graceful shutdown sequence. The docker compose down command extends this by removing containers and networks; adding --volumes also destroys the on-disk graph data mapped to the container.

The codebase_rag/tools/health_checker.py module provides health check functionality that queries port 7687 to verify daemon availability, which is useful for confirming that the service has fully terminated after running stop commands.

Programmatically Stop Memgraph from Python

For automated testing or scripted workflows, you can invoke the Docker Compose command via Python's subprocess module. The repository's test suite uses this approach in fixtures to manage container lifecycles:

import subprocess

def stop_memgraph():
    """Stop only the Memgraph service defined in the project's Docker Compose."""
    subprocess.run(
        ["docker", "compose", "-f", "codebase_rag/docker-compose.yaml", "stop", "memgraph"],
        check=True,
    )

This method executes the same CLI command with the same arguments, ensuring consistency between manual and automated operations.

Summary

• Targeted shutdown: Use docker compose -f codebase_rag/docker-compose.yaml stop memgraph to halt only the graph database while preserving other Code-Graph-RAG services.
• Full stack teardown: Use docker compose -f codebase_rag/docker-compose.yaml down to stop all containers and remove networks; add --volumes to delete persistent graph data.
• Configuration source: The service is defined in codebase_rag/docker-compose.yaml using the memgraph/memgraph-platform image with ports 7687 and 7444 exposed.
• Automation support: Python scripts can trigger shutdowns via subprocess.run() using the same command structure found in the repository's test fixtures.

Frequently Asked Questions

Does stopping Memgraph delete my graph data?

No. Using docker compose stop or docker compose down without the --volumes flag preserves your data in Docker volumes. However, running docker compose -f codebase_rag/docker-compose.yaml down --volumes will permanently delete the persistent storage associated with the Memgraph container, wiping all graph data.

What is the difference between docker compose stop and down?

The stop command halts running containers but leaves the container definitions, networks, and volumes intact, allowing for quick restarts with docker compose start. The down command stops containers and additionally removes the containers, networks, and (optionally) volumes, requiring a fresh up command to recreate the entire stack.

How do I verify that Memgraph has stopped successfully?

You can check running containers with docker ps to confirm the Memgraph container is no longer listed. Alternatively, the repository's codebase_rag/tools/health_checker.py module implements health checks against port 7687; if these checks fail to connect, the daemon has stopped. The cgr status command also utilizes this health checker to report service availability.

Can I restart Memgraph without rebuilding the container?

Yes. If you used docker compose stop, simply run docker compose -f codebase_rag/docker-compose.yaml start memgraph to resume the existing container with data intact. If you used down without the --volumes flag, run docker compose up again; this recreates the container but mounts the existing persistent volume, preserving your graph data.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}