Skip to content

DOCKER.md

Latest commit

 

History

History
149 lines (99 loc) · 4.26 KB

DOCKER.md

File metadata and controls

149 lines (99 loc) · 4.26 KB

Running the MCP Server with Docker

This document provides instructions on how to build and run the Model Context Protocol (MCP) server using Docker.

Building the Image

You can build the Docker image in two ways: from a local clone of the repository or directly from GitHub.

Building from a Local Clone

First, build the Docker image from the root of the project:

docker build -t cloudinary-asset-management-mcp .

Building from GitHub

You can also build the image directly from the GitHub repository without cloning it first.

docker build -t cloudinary-asset-management-mcp https://github.com/cloudinary/asset-management-js.git

Running the Container

The MCP server requires Cloudinary credentials to run. You can provide these credentials in one of three ways. When running the docker run command, you will also map a local port (e.g., 2718) to the container's port 2718.

Note: Replace <your_cloud_name>, <your_api_key>, and <your_api_secret> with your actual Cloudinary credentials.

Option 1: Using Individual Environment Variables

This is the recommended method.

docker run -d -p 2718:2718 \
  -e CLOUDINARY_CLOUD_NAME="<your_cloud_name>" \
  -e CLOUDINARY_API_KEY="<your_api_key>" \
  -e CLOUDINARY_API_SECRET="<your_api_secret>" \
  cloudinary-asset-management-mcp start --transport sse

Note: If you have these variables already set in your shell environment, you can pass them directly to the container without specifying the values:

docker run -d -p 2718:2718 \
  -e CLOUDINARY_CLOUD_NAME \
  -e CLOUDINARY_API_KEY \
  -e CLOUDINARY_API_SECRET \
  cloudinary-asset-management-mcp start --transport sse

Option 2: Using Command-Line Arguments

You can also provide the credentials as arguments to the start command.

docker run -d -p 2718:2718 \
  cloudinary-asset-management-mcp start --transport sse \
  --cloud-name "<your_cloud_name>" \
  --api-key "<your_api_key>" \
  --api-secret "<your_api_secret>"

Option 3: Using CLOUDINARY_URL Environment Variable

This method combines all credentials into a single URL.

docker run -d -p 2718:2718 \
  -e CLOUDINARY_URL="cloudinary://<your_api_key>:<your_api_secret>@<your_cloud_name>" \
  cloudinary-asset-management-mcp start --transport sse

Note: If you have the CLOUDINARY_URL variable already set in your shell environment, you can pass it directly:

docker run -d -p 2718:2718 -e CLOUDINARY_URL cloudinary-asset-management-mcp start --transport sse

Connecting to the Server

Once the container is running with the SSE transport enabled (as shown in the commands above), the MCP server is available at the following endpoint:

http://localhost:2718/sse

If you are running Docker on a different host, replace localhost with the appropriate hostname or IP address.

Stopping the Container

You can find the container ID by running docker ps and then stop it using docker stop.

To stop the container started from the cloudinary-asset-management-mcp image:

docker stop $(docker ps -a -q --filter "ancestor=cloudinary-asset-management-mcp")

Viewing Logs

You can view the logs from your running container to monitor its output or troubleshoot issues.

First, find the ID of your container:

docker ps

This will list all running containers, including their IDs.

Static Logs

To see all logs that have been generated so far, use the docker logs command with the container ID.

docker logs <your_container_id>

Live Logs

To see logs in real time, add the --follow (or -f) flag.

docker logs --follow <your_container_id>

Press Ctrl+C to stop following the logs.

Debugging

You can enable more detailed logging for troubleshooting in two ways.

Using the --log-level Flag

Set the --log-level flag to debug when starting the container.

docker run -d -p 2718:2718 \
  -e CLOUDINARY_URL \
  cloudinary-asset-management-mcp start --transport sse --log-level debug

Using the CLOUDINARY_DEBUG Environment Variable

You can also enable a debug logger by setting the CLOUDINARY_DEBUG environment variable to true.

docker run -d -p 2718:2718 \
  -e CLOUDINARY_URL \
  -e CLOUDINARY_DEBUG=true \
  cloudinary-asset-management-mcp start --transport sse