Deployment Guide

This guide covers deploying the evasys EAI application to production environments.

Building for Production

Create the JAR

cd evasys-eai
mvn clean package

The executable JAR will be created at target/evasys-eai-<version>.jar.

Create a Docker Image

A Dockerfile is provided in the evasys-eai/ directory:

cd evasys-eai
docker build -t evasys-eai:latest .

Deployment Options

Standalone JAR

Run the application directly:

java -jar evasys-eai-<version>.jar \
  --evasys.uri=https://evasys.example.com/soap.php \
  --evasys.username=user \
  --evasys.password=secret

Or with an external configuration file:

java -jar evasys-eai-<version>.jar \
  --spring.config.location=file:/etc/evasys-eai/application.yml

Docker

docker run -d \
  --name evasys-eai \
  -p 8080:8080 \
  -e EVASYS_URI=https://evasys.example.com/soap.php \
  -e EVASYS_USERNAME=user \
  -e EVASYS_PASSWORD=secret \
  -e SAPPO_URI=/ws/training \
  -e SAPPO_USERNAME=sappo \
  -e SAPPO_PASSWORD=secret \
  evasys-eai:latest

Kubernetes

Example Kubernetes deployment:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: evasys-eai
spec:
  replicas: 1
  selector:
    matchLabels:
      app: evasys-eai
  template:
    metadata:
      labels:
        app: evasys-eai
    spec:
      containers:
        - name: evasys-eai
          image: ghcr.io/it-at-m/evasys-eai:latest
          ports:
            - containerPort: 8080
          env:
            - name: EVASYS_URI
              valueFrom:
                secretKeyRef:
                  name: evasys-secrets
                  key: uri
            - name: EVASYS_USERNAME
              valueFrom:
                secretKeyRef:
                  name: evasys-secrets
                  key: username
            - name: EVASYS_PASSWORD
              valueFrom:
                secretKeyRef:
                  name: evasys-secrets
                  key: password
          livenessProbe:
            httpGet:
              path: /actuator/health/liveness
              port: 8080
            initialDelaySeconds: 30
            periodSeconds: 10
          readinessProbe:
            httpGet:
              path: /actuator/health/readiness
              port: 8080
            initialDelaySeconds: 10
            periodSeconds: 5
          resources:
            requests:
              memory: "256Mi"
              cpu: "100m"
            limits:
              memory: "512Mi"
              cpu: "500m"
---
apiVersion: v1
kind: Service
metadata:
  name: evasys-eai
spec:
  selector:
    app: evasys-eai
  ports:
    - port: 80
      targetPort: 8080

Health Checks

The application exposes health endpoints for container orchestration:

Endpoint	Purpose
/actuator/health	Overall health status
/actuator/health/liveness	Liveness probe (is the app running?)
/actuator/health/readiness	Readiness probe (is the app ready to serve?)

Monitoring

Prometheus Metrics

Metrics are exposed at /actuator/metrics in Prometheus format.

Key metrics to monitor:

• http_server_requests_seconds - HTTP request latency
• jvm_memory_used_bytes - JVM memory usage
• jvm_gc_pause_seconds - Garbage collection pauses

Logging

The application outputs structured JSON logs (Logstash format) to stdout:

{
  "@timestamp": "2024-01-15T10:30:00.000Z",
  "level": "INFO",
  "message": "Processing training requests...",
  "application_name": "evasys-eai",
  "logger_name": "de.muenchen.evasys.service.TrainingProcessorService"
}

Configure your log aggregation system (ELK, Loki, etc.) to collect logs from stdout.