MCP-OpenStack-Ops

MCP-OpenStack-Ops is a Python-based MCP server that provides project-scoped OpenStack management and monitoring tools. It enforces a single OpenStack project context and includes safety gates so that mutating actions only run when ALLOW_MODIFY_OPERATIONS is true. The server exposes a broad suite of tools for compute, networking, storage, images, identity, and orchestration operations, with capabilities for bulk actions, result feedback, and usage monitoring. Use the get_service_status and get_usage_statistics tools to gauge health and capacity, and rely on the post-action feedback to verify outcomes after any mutating operation. Tools are designed to respect a single tenant by validating OS_PROJECT_NAME and supporting pagination, filtering, and batch operations across resources like instances, volumes, networks, images, snapshots, and keypairs. If you’re deploying via Docker, you can pull the OpenStack-ops image and run it with a simple docker run invocation, or containerize it as part of a larger stack with docker-compose.

Prerequisites:

• Docker and Docker Compose installed on the host
• Access to an OpenStack environment (Epoxy 2025.1 recommended) and credentials for a single project
• Python-based MCP server (this OpenStack Ops server is provided as a container image)

Step 1: Install Docker

• Follow your platform’s instructions to install Docker and Docker Compose.

Step 2: Pull and run the MCP server (Docker)

• You can run the MCP server directly from the image:

  docker run -d --name mcp-openstack-ops -e OS_PROJECT_NAME=your-project
  -e ALLOW_MODIFY_OPERATIONS=true
  -p 8000:8000 call518/mcp-server-openstack-ops

Step 3: Verify the server is up

• Check logs:

  docker logs -f mcp-openstack-ops

Step 4: Optional – use with docker-compose

• If you prefer docker-compose, create a docker-compose.yml referencing the image and environment variables, then run:

  docker-compose up -d

Step 5: Configure client to connect

• Point your MCP client to the server’s endpoint (as exposed by your deployment) and authenticate using the configured OpenStack credentials within the server’s environment.

Notes:

• The server supports OpenStack SDK customization for older releases via Dockerfile.MCP-Server and pyproject.toml; keep both versions in sync to avoid dependency conflicts.
• If mutating commands are needed, ensure ALLOW_MODIFY_OPERATIONS is set to true; otherwise actions will be read-only and auditable.
• For bulk operations, you can target resources using names, filters, or JSON arrays as demonstrated in the tool examples.

Tips and common considerations:

• Compatibility: Optimized for OpenStack Epoxy (2025.1) but broadly compatible with newer and older releases; some endpoints may require adaptation for older releases.
• SDK version alignment: When customizing for older OpenStack releases, ensure the OpenStackSDK version in Dockerfile.MCP-Server and pyproject.toml match to prevent runtime conflicts.
• Safety gates: By default, mutating actions require ALLOW_MODIFY_OPERATIONS to be true to prevent unintended changes.
• Bulk operations: Use filtering options like name_contains, status, and other criteria to perform bulk actions across multiple resources in a single call.
• Monitoring: Use get_service_status, get_resource_monitoring, and get_usage_statistics to observe health and capacity for the active project.
• Debugging: Review server events, quotas, and ownership tooling to diagnose issues quickly and maintain an auditable trail.