From Containers to Observability: Docker Compose Roadmap¶

Monitoring (Prometheus & Grafana)¶

Metrics monitoring allows you to see numbers (CPU, RAM, Request counts) over time.

Prometheus: Collects and stores metrics (The Database).
Grafana: Visualizes the metrics (The Dashboard).

Project Structure¶

project06/
├── docker-compose.yaml
└── prometheus.yml

Step 1: The Prometheus Config¶

prometheus.yml¶

global:
  scrape_interval: 5s # Wie oft Prometheus Daten abfragt (Standard ist 15s)

scrape_configs:
  - job_name: 'prometheus'
    static_configs:
      - targets: ['prometheus:9090'] # Wir nutzen den Servicenamen statt localhost

Step 2: The Compose File¶

docker-compose.yaml¶

services:
  prometheus:
    image: prom/prometheus:latest
    container_name: prometheus
    ports:
      - "9090:9090"
    volumes:
      - ./prometheus.yml:/etc/prometheus/prometheus.yml
      - prometheus_data:/prometheus # Speichert Metriken dauerhaft

  grafana:
    image: grafana/grafana:latest
    container_name: grafana
    ports:
      - "3000:3000"
    depends_on:
      - prometheus
    volumes:
      - grafana_data:/var/lib/grafana # Verhindert Datenverlust der Dashboards

volumes:
  prometheus_data:
  grafana_data:

Practical Exercise: View the Dashboard¶

Start the stack:
```
docker-compose up -d
```
Check if containers are running:
```
docker ps
```
Open Prometheus (Targets check):
```
http://localhost:9090/targets
```
Open Grafana (Initial Setup):
```
http://localhost:3000
```
(Default Login: admin / admin)
Clean up (Important before next project):
```
docker-compose down
```

Alerting (Mattermost & Webhooks)¶

Project Structure¶

project07/
└── docker-compose.yaml

Step 1: The Compose File¶

docker-compose.yaml¶

services:
  mattermost:
    image: mattermost/mattermost-preview
    container_name: mattermost
    ports:
      - "8065:8065"
    # Preview Image nutzt eine interne DB, ideal für Tests

Step 2: Integration Logic¶

Open Mattermost:
```
http://localhost:8065
```
Setup: Erstelle einen Account und ein Team “DevOps”.
Webhook: Gehe zu Product Settings -> Integrations -> Incoming Webhooks.
Grafana Link: In Grafana unter Alerting -> Contact Points den Typ “Webhook” wählen.

Test the Alert:

# Simuliere Last oder stoppe einen Service, um Alarm zu triggern
docker-compose stop mattermost

Logging (The ELK Stack)¶

ElasticSearch: Search engine for logs.
Logstash: (Optional in basic setups) Processes logs.
Kibana: The interface.

Project Structure¶

project08/
└── docker-compose.yaml

Step 1: The Compose File¶

docker-compose.yaml¶

services:
  elasticsearch:
    image: docker.elastic.co/elasticsearch/elasticsearch:8.10.2
    container_name: elasticsearch
    environment:
      - discovery.type=single-node # Verhindert Cluster-Suche (spart Ressourcen)
      - xpack.security.enabled=false # Deaktiviert Auth für lokale Tests
      - "ES_JAVA_OPTS=-Xms512m -Xmx512m" # Begrenzt RAM-Verbrauch der JVM
    ports:
      - "9200:9200"
    ulimits:
      memlock:
        soft: -1
        hard: -1

  kibana:
    image: docker.elastic.co/kibana/kibana:8.10.2
    container_name: kibana
    ports:
      - "5601:5601"
    environment:
      - ELASTICSEARCH_HOSTS=http://elasticsearch:9200
    depends_on:
      - elasticsearch

Practical Exercise: Searching Logs¶

Start (Wait for healthy status):
```
docker-compose up -d
```

Verify Elasticsearch is “Green”:

http://localhost:9200/_cluster/health?pretty

Access Kibana:
```
http://localhost:5601
```
Data View: In Kibana unter Stack Management -> Data Views einen Index (z.B. *) erstellen, um Logs zu sehen.

Summary of Learning Path¶

Tool	Best For	Docker Skill Learned
Prometheus	Numbers/Metrics	Target Scraping & Config Mounting
Mattermost	Alerts/Chat	Webhooks & External API links
ELK Stack	Text/Error Logs	Resource Limits (RAM/Java Opts)