HELASEID - Headless Automatic Service Discovery © by G. Ginanjar

Overview

HELASEID (Headless Automatic Service Discovery) is a lightweight, containerized Python-based solution for dynamically updating Kubernetes endpoints for headless services. It continuously monitors the health of service instances and updates the corresponding Kubernetes Endpoints object accordingly.

The name HELASEID is inspired by two powerful supervillains:

• Hela (Marvel) : Beautiful yet resourceful.
• Darkseid (DC) : Resilient and strategic.

This tool embodies their strengths�efficiency, adaptability, and robustness.

Features

• Automated Health Checks: Monitors service health via TCP probes.
• Kubernetes Endpoint Management: Dynamically updates the Endpoints object.
• Microsoft Teams Integration: Sends alerts on failures.
• Lightweight & Efficient: Runs with minimal CPU and memory footprint.
• Configurable via Kubernetes ConfigMap & Secrets: Environment-based configuration.

Architecture

HELASEID follows a simple yet effective architecture:

1. Loads configuration from Kubernetes ConfigMap.
2. Performs health checks on configured service IPs.
3. Updates Kubernetes Endpoints to reflect live instances.
4. Sends notifications to Microsoft Teams when failures occur.

Installation

Prerequisites

• Kubernetes cluster (tested on v1.24+)
• kubectl access with permissions to modify Endpoints
• Microsoft Teams Webhook (optional for alerts)

Deployment Steps

1. Deploy Kubernetes Manifests:

git clone git@bitbucket.org:mandiri-sekuritas/helaseid.git
cd helaseid
kubectl apply -f manifests/

2. Verify Deployment:

kubectl get pods -n datastream

3. Check Logs:

kubectl logs -f deployment/datastream-helaseid-dev -n datastream

Configuration

HELASEID uses a ConfigMap (datastream-helaseid-dev) to define environment variables:

apiVersion: v1
kind: ConfigMap
metadata:
  name: datastream-helaseid-dev
  namespace: datastream
data:
  config.env: |
    IPS=("10.188.2.3" "10.188.2.4" "10.188.2.5")
    PORT=9092
    MAX_RETRY=10
    INTERVAL=0.01
    LOG_LEVEL=INFO
    NAMESPACE=datastream
    SERVICE_NAME=datastream-redpanda-api-dev

• IPS - List of service instance IPs to monitor.
• PORT - Service port to check health.
• MAX_RETRY - Maximum retries before marking unhealthy.
• INTERVAL - Interval between health checks.
• LOG_LEVEL - Logging verbosity (DEBUG, INFO, ERROR).
• NAMESPACE - Kubernetes namespace.
• SERVICE_NAME - Kubernetes service name.

Microsoft Teams Integration

To enable Teams notifications, add a Kubernetes Secret with a webhook URL:

apiVersion: v1
kind: Secret
metadata:
  name: datastream-hslc-dev
type: Opaque
data:
  TEAMS_WEBHOOK_URL: <Base64-encoded-webhook-URL>

Running Locally

To test outside Kubernetes, build and run the container:

docker build -t helaseid:latest .
docker run --rm -e IPS="10.188.2.3 10.188.2.4" -e PORT=9092 helaseid

Future Enhancements

• Support for multi-cluster service discovery.
• Dynamic IP retrieval via Kubernetes API.
• Metrics export for Prometheus integration.
• Support multi-zone/multi-region discovery with dynamic preferences and master pinning.

Author

• G. Ginanjar.