PTP Server
FreeNot checkedMonitors and analyzes Precision Time Protocol (PTP) systems in OpenShift clusters, enabling configuration analysis, real-time log monitoring, and health checks.
About
Monitors and analyzes Precision Time Protocol (PTP) systems in OpenShift clusters, enabling configuration analysis, real-time log monitoring, and health checks.
README
A Model Context Protocol (MCP) server for monitoring and analyzing Precision Time Protocol (PTP) systems in OpenShift clusters.
🚀 Features
- PTP Configuration Analysis: Parse and validate PTP configurations from OpenShift
- Real-time Log Monitoring: Access linuxptp daemon logs with intelligent parsing
- Natural Language Queries: Ask questions about PTP status in plain English
- Health Monitoring: Comprehensive PTP system health checks
- Synchronization Analysis: Monitor sync status, offsets, and BMCA state
- Clock Hierarchy: Track grandmaster and clock hierarchy information
- ITU-T Compliance: Validate configurations against ITU-T G.8275.1 standards
📋 Prerequisites
- Python 3.8 or higher
- OpenShift CLI (
oc) installed and configured - Access to OpenShift cluster with PTP operator installed
- PTP namespace (
openshift-ptp) exists
🛠️ Source Installation
- Clone the repository:
git clone https://github.com/redhat-cne/ptp-mcp-server.git cd ptp-mcp-server
To Deploy For Quick Tests
Install dependencies:
pip install -r requirements.txtVerify OpenShift access:
export KUBECONFIG=/path/to/kubeconfig oc whoami oc get namespace openshift-ptpQuick Testing
Run the comprehensive test suite:
export KUBECONFIG=/path/to/kubeconfig
python quick_test.py
Expected output:
🔍 PTP MCP Server API Quick Test
==================================================
Tests Passed: 8/8
Success Rate: 100.0%
🎉 ALL TESTS PASSED! Your API is ready for agent integration.
🔧 MCP Server
The MCP server supports two transport modes:
- stdio: For local MCP clients (Claude Code, Claude Desktop)
- HTTP/SSE: For OpenShift Lightspeed integration
Local Usage (stdio mode)
python ptp_mcp_server.py
Remote Usage (http mode)
# Default port 8080
python ptp_mcp_server.py --http
# Custom port
python ptp_mcp_server.py --http --port 9000
# Or use environment variable
PTP_MCP_PORT=9000 python ptp_mcp_server.py --http
Deploy to OpenShift
podman build -t quay.io/$USER/ptp-mcp-server:latest .
podman push quay.io/$USER/ptp-mcp-server:latest
cd k8s && kustomize edit set image quay.io/redhat-cne/ptp-mcp-server=quay.io/$USER/ptp-mcp-server:latest && cd ..
oc apply -k k8s/
Configure in OpenShift Lightspeed
Add the MCP server to your OLSConfig:
apiVersion: ols.openshift.io/v1alpha1
kind: OLSConfig
metadata:
name: cluster
spec:
featureGates:
- MCPServer
mcpServers:
- name: ptp-monitoring
url: 'http://ptp-mcp-server.openshift-ptp.svc.cluster.local:8080/mcp'
timeout: 30
Usage in OpenShift Lightspeed
By default, the MCP server assumes that it is running on the cluster that is to be monitored/queried. Any tool use will target the local cluster. However, if the MCP server is running on a hub cluster and the user intends for its prompt to target spoke cluster then OLS context should include a base64 copy of the kubeconfig for the spoke cluster.
note: It is important to use a minimal kubeconfig that is token based rather than a client certificate based kubeconfig as the OLS context size will not allow for a larger kubeconfig value.
The following example generates such a kubeconfig for lab testing purposes. In a production environment care should be taken to limit the service account permissions to only the strict minimum RBAC policies required.
oc create sa ols-ptp-user -n default 2>/dev/null
oc adm policy add-cluster-role-to-user cluster-admin -z ols-ptp-user -n default
TOKEN=$(oc create token ols-ptp-user -n default --duration=24h)
API_SERVER=$(oc whoami --show-server)
cat << EOF > minimal-kubeconfig.yaml
apiVersion: v1
kind: Config
clusters:
- cluster:
server: ${API_SERVER}
insecure-skip-tls-verify: true
name: cluster
contexts:
- context:
cluster: cluster
user: user
name: ctx
current-context: ctx
users:
- name: user
user:
token: ${TOKEN}
EOF
📚 API Endpoints
1. Configuration API
from ptp_tools import PTPTools
tools = PTPTools()
result = await tools.get_ptp_config({"namespace": "openshift-ptp"})
2. Logs API
result = await tools.get_ptp_logs({"lines": 1000})
3. Search API
result = await tools.search_logs({"query": "dpll", "time_range": "last_hour"})
4. Health API
result = await tools.check_ptp_health({"check_config": True, "check_sync": True})
5. Natural Language API
result = await tools.query_ptp({"question": "What is the current grandmaster?"})
6. Grandmaster Status API
result = await tools.get_grandmaster_status({"detailed": True})
7. Sync Status API
result = await tools.analyze_sync_status({"include_offsets": True})
8. Clock Hierarchy API
result = await tools.get_clock_hierarchy({"include_ports": True})
🚀 Usage Examples
Basic Health Check
import asyncio
from ptp_tools import PTPTools
async def check_health():
tools = PTPTools()
health = await tools.check_ptp_health({})
if health["success"]:
print(f"Status: {health['overall_status']}")
for check_name, result in health["checks"].items():
print(f"{check_name}: {result}")
else:
print(f"Error: {health.get('error')}")
asyncio.run(check_health())
Natural Language Query
async def ask_question():
tools = PTPTools()
response = await tools.query_ptp({
"question": "What is the current grandmaster?"
})
if response["success"]:
print(f"Answer: {response['response']}")
else:
print(f"Error: {response.get('error')}")
asyncio.run(ask_question())
Log Analysis
async def analyze_logs():
tools = PTPTools()
# Get recent logs
logs = await tools.get_ptp_logs({"lines": 500})
# Search for specific events
sync_loss = await tools.search_logs({"query": "sync loss"})
clock_changes = await tools.search_logs({"query": "clockClass change"})
print(f"Total logs: {logs['logs_count']}")
print(f"Sync loss events: {sync_loss['matching_logs']}")
print(f"Clock changes: {clock_changes['matching_logs']}")
asyncio.run(analyze_logs())
MCP Tools Available
| Tool | Description |
|---|---|
get_ptp_config |
Get PTP configuration |
get_ptp_logs |
Get linuxptp daemon logs |
search_logs |
Search logs for patterns |
get_grandmaster_status |
Get grandmaster info |
analyze_sync_status |
Analyze sync status |
get_clock_hierarchy |
Get clock hierarchy |
check_ptp_health |
Comprehensive health check |
query_ptp |
Natural language interface |
run_pmc_query |
Execute PMC commands |
Deployment Files
| File | Purpose |
|---|---|
Dockerfile |
Container image with Python + oc CLI |
k8s/rbac.yaml |
ServiceAccount, ClusterRole, ClusterRoleBinding |
k8s/deployment.yaml |
Deployment with health checks |
k8s/service.yaml |
ClusterIP Service |
k8s/olsconfig-example.yaml |
Example OLS configuration |
k8s/kustomization.yaml |
Deploy all with oc apply -k k8s/ |
RBAC Permissions
The ServiceAccount is granted these permissions:
| Resource | Verbs | Purpose |
|---|---|---|
ptpconfigs, ptpoperatorconfigs |
get, list, watch | Read PTP configurations |
pods |
get, list, watch | Find linuxptp-daemon pods |
pods/log |
get, list | Read daemon logs |
pods/exec |
create | Execute PMC queries |
namespaces |
get, list | Namespace access |
nodes |
get, list | Node topology (optional) |
📊 Performance
- Average Response Time: 0.78s
- Fastest API: Configuration API (0.22s)
- Concurrent Operations: 4/4 successful in 2.45s
- Success Rate: 100% (8/8 endpoints)
🏗️ Architecture
ptp-mcp-server/
├── ptp_mcp_server.py # Main MCP server (stdio + HTTP modes)
├── ptp_config_parser.py # PTP configuration parser
├── ptp_log_parser.py # Linuxptp log parser
├── ptp_model.py # PTP data models
├── ptp_query_engine.py # Natural language query engine
├── ptp_tools.py # API endpoint implementations
├── quick_test.py # Quick test suite
├── performance_test.py # Performance benchmarking
├── requirements.txt # Python dependencies
├── Dockerfile # Container image definition
└── k8s/ # Kubernetes/OpenShift manifests
├── kustomization.yaml # Kustomize configuration
├── rbac.yaml # ServiceAccount & RBAC
├── deployment.yaml # Deployment specification
├── service.yaml # Service definition
└── olsconfig-example.yaml # OLS integration example
🔍 PTP Concepts Supported
- BMCA (Best Master Clock Algorithm): Clock selection and hierarchy
- Clock Types: OC (Ordinary Clock), BC (Boundary Clock), TC (Transparent Clock)
- ITU-T G.8275.1: Profile compliance and validation
- Synchronization: Offset tracking, frequency adjustment, sync status
- Grandmaster: Primary time source identification and status
- Clock Class: Quality and traceability indicators
- Domain Numbers: PTP domain configuration (24-43 for ITU-T)
🧪 Testing
Run All Tests
python quick_test.py
Performance Testing
python performance_test.py
Individual Component Testing
# Test configuration parser
python -c "from ptp_config_parser import PTPConfigParser; import asyncio; asyncio.run(PTPConfigParser().get_ptp_configs())"
# Test log parser
python -c "from ptp_log_parser import PTPLogParser; import asyncio; asyncio.run(PTPLogParser().get_ptp_logs())"
📖 Documentation
- Testing Guide - Comprehensive testing instructions
- Agent Integration Guide - Integration examples for agents
- Testing Steps - Step-by-step testing process
- Testing Results - Complete test results
🤝 Contributing
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
📄 License
This project is licensed under the MIT License - see the LICENSE file for details.
🙏 Acknowledgments
- OpenShift PTP Operator team
- Linuxptp project
- Model Context Protocol (MCP) community
📞 Support
For issues and questions:
- Create an issue on GitHub
- Check the testing documentation
- Review the agent integration guide
Status: ✅ Production Ready
Last Updated: January 2025
Version: 1.0.0
Installing PTP Server
This server has no published package — it is built from source. Open the repository and follow its README.
▸ github.com/redhat-cne/ptp-mcp-serverFAQ
Is PTP Server MCP free?
Yes, PTP Server MCP is free — one-click install via Unyly at no cost.
Does PTP Server need an API key?
No, PTP Server runs without API keys or environment variables.
Is PTP Server hosted or self-hosted?
A hosted option is available: Unyly runs the server in the cloud, no local setup required.
How do I install PTP Server in Claude Desktop, Claude Code or Cursor?
Open PTP Server on unyly.org, pick your client tab (Claude Desktop, Claude Code, Cursor) and press Install — the config is generated automatically, no JSON editing.
Related MCPs
GitHub
PRs, issues, code search, CI status
by GitHubFilesystem
Secure file operations with configurable access controls.
Memory
Knowledge graph-based persistent memory system.
Template MCP Server
A CLI tool to create a new Model Context Protocol server project with TypeScript support, dual transport options, and an extensible structure
by mcpdotdirectCompare PTP Server with
Not sure what to pick?
Find your stack in 60 seconds
Author?
Embed badge for your README
Browse similar
All development MCPs
