MCP Server
About
MCP Server is a tool for injecting faults and performing stress tests on Kubernetes pods and hosts using Chaos Mesh.
Details
- Author
- Hades-gsl
- Downloads
- 88
- Categories
- Other
Jump to
- Pod fault injection (pod failure, pod kill, container kill)
- Pod CPU and memory stress testing
- Host CPU and memory stress testing
- Host disk faults (fill, read payload, write payload)
- Network faults (partition, bandwidth limitation)
- Log retrieval for services and containers
- Load test monitoring and result aggregation
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Clone the repository, set up a virtual environment with uv, install dependencies, and run uv run python server.py. The transport method defaults to sse but can be changed with the --transport argument. You must have Chaos Mesh installed in your Kubernetes cluster and define your services in a services.json file.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"mcp server": {
"mcp-server-hades-gsl": {
"command": "uv",
"args": [
"venv"
]
}
}
}
}McpServers
{
"mcp-server-hades-gsl": {
"command": "uv",
"args": [
"venv"
]
}
}MCP Server
MCP Server is a tool for injecting faults and performing stress tests on Kubernetes pods and hosts using Chaos Mesh. It also provides utilities for retrieving logs and monitoring load test results.
Features
- Pod Fault Injection: Simulate pod failures, pod kills, and container kills.
- Pod Stress Testing: Simulate CPU and memory stress tests on pods.
- Host Stress Testing: Simulate CPU and memory stress tests on hosts.
- Host Disk Faults: Simulate disk fill, read payload, and write payload faults on hosts.
- Network Faults: Simulate network partition and bandwidth limitations.
- Log Retrieval: Retrieve logs for specific services and containers.
- Load Test Monitoring: Parse and retrieve aggregated load test results.
Project Structure
- fault_inject.py: Contains functions for injecting faults and performing stress tests.
- kube.py: Provides utilities for interacting with Kubernetes, such as retrieving pod logs.
- server.py: Implements the MCP server and exposes tools for fault injection and log retrieval.
- test.py: Contains test cases and examples for using the fault injection and logging utilities.
- services.json: Defines the services in the cluster, including their namespaces, labels, and containers.
Installation
Prerequisites:
Before starting, make sure you have uv installed. You can install uv here .
1. Clone the repository:
git clone https://github.com/Hades-gsl/mcp_server.git
cd mcp_server
2. Build and set up the virtual environment using uv:
uv venv
source .venv/bin/activate
3. Install dependencies:
uv sync
4. Install Chaos Mesh in your Kubernetes cluster.
You can intall it here.
Usage
Run the MCP server using the following command:
uv run python server.py
You can specify the transport method by adding the --transport argument. By default, the transport parameter is set to sse.
Service Infomation
Write the services.json file to show the services in your cluster. You can write anything useful, the more detailed the better.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
