OpenShift

Provides OpenShift diagnostics with storage, monitoring, and pod health tools for real-time cluster observability.

other

0

GitHub Stars

other

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "junzzhu-openshift-mcp-server": {
      "command": "uv",
      "args": [
        "run",
        "openshift-mcp-server"
      ]
    }
  }
}

This MCP Server provides a focused set of OpenShift diagnostics and troubleshooting tools you can run locally or via an MCP client to quickly analyze storage, resources, GPU usage, and pod health across your cluster.

How to use

You connect your MCP client to the OpenShift MCP Server and invoke the available tools to gather runtime metrics and health data from your cluster. Typical workflows include analyzing storage usage on all nodes, monitoring cluster resource balance, checking GPU utilization, and inspecting pod health or specific pod logs. Use the tools to ask targeted questions such as overall storage consumption per node, GPU idle resources, or pod restart anomalies. The server surfaces real-time Prometheus metrics via the cluster route for observability.

How to install

Prerequisites: you need the MCP client runtime installed on your workstation. The server can be run in two supported ways.

Option 1: Use the uv runtime (recommended) and run the server directly.

Option 2: Install with Python’s pip package manager and run the server from Python.

Where to configure the MCP client

{
  "mcpServers": {
    "openshift_tools": {
      "command": "uv",
      "args": ["run", "openshift-mcp-server"]
    }
  }
}

Available tools

get_cluster_storage_report

Generates a comprehensive report of ephemeral storage usage across nodes, including top pod storage consumers.

inspect_node_storage_forensics

Performs deep forensics on a node’s disk usage to identify unused images and writable container layers.

check_persistent_volume_capacity

Monitors persistent volume usage across namespaces with configurable thresholds.

get_cluster_resource_balance

Analyzes CPU and memory resource distribution across nodes to reveal imbalances.

detect_pod_restarts_anomalies

Identifies pods with excessive restarts within a specified time window.

get_gpu_utilization

Tracks GPU usage and identifies idle GPU resources across the cluster.

inspect_gpu_pod

Runs nvidia-smi inside a GPU-enabled pod to show real-time processes and memory details.

check_gpu_health

Checks GPU hardware errors (XID) and throttling events across the cluster.

get_vllm_metrics

Monitors vLLM inference server performance metrics such as throughput, queue size, and cache usage.

get_pod_logs

Retrieves and analyzes logs from a specific pod with support for previous container logs, tail limits, and time-based filtering.

get_pod_diagnostics

Performs a comprehensive health check of a pod including status, conditions, container states, restart counts, and issue detection.