fetch_ml/docker-compose.yml
Jeremie Fraeys cd5640ebd2 Slim and secure: move scripts, clean configs, remove secrets
- Move ci-test.sh and setup.sh to scripts/
- Trim docs/src/zig-cli.md to current structure
- Replace hardcoded secrets with placeholders in configs
- Update .gitignore to block .env*, secrets/, keys, build artifacts
- Slim README.md to reflect current CLI/TUI split
- Add cleanup trap to ci-test.sh
- Ensure no secrets are committed
2025-12-07 13:57:51 -05:00

123 lines
3.3 KiB
YAML

# Homelab Docker Compose with Centralized Monitoring
# Includes: API, Redis, Prometheus, Grafana, Loki
services:
redis:
image: redis:7-alpine
container_name: ml-experiments-redis
ports:
- "6379:6379"
volumes:
- redis_data:/data
restart: unless-stopped
command: redis-server --appendonly yes
healthcheck:
test: [ "CMD", "redis-cli", "ping" ]
interval: 30s
timeout: 10s
retries: 3
api-server:
build:
context: .
dockerfile: build/docker/simple.Dockerfile
container_name: ml-experiments-api
ports:
- "9101:9101"
- "9100:9100" # Prometheus metrics endpoint
volumes:
- ./data:/data/experiments
- ./logs:/logs
- ./configs/environments/config-local.yaml:/app/configs/config.yaml
- ./ssl:/app/ssl
depends_on:
redis:
condition: service_healthy
restart: unless-stopped
command: ["/usr/local/bin/api-server", "-config", "/app/configs/config.yaml"]
environment:
- REDIS_URL=redis://redis:6379
- LOG_LEVEL=info
healthcheck:
test: [ "CMD", "curl", "-k", "https://localhost:9101/health" ]
interval: 30s
timeout: 10s
retries: 3
start_period: 40s
labels:
logging: "promtail"
job: "api-server"
# Prometheus - Metrics collection
prometheus:
image: prom/prometheus:latest
container_name: ml-experiments-prometheus
ports:
- "9090:9090"
volumes:
- ./monitoring/prometheus.yml:/etc/prometheus/prometheus.yml
- prometheus_data:/prometheus
command:
- '--config.file=/etc/prometheus/prometheus.yml'
- '--storage.tsdb.path=/prometheus'
- '--web.console.libraries=/etc/prometheus/console_libraries'
- '--web.console.templates=/etc/prometheus/consoles'
- '--web.enable-lifecycle'
restart: unless-stopped
# Grafana - Visualization
grafana:
image: grafana/grafana:latest
container_name: ml-experiments-grafana
ports:
- "3000:3000"
volumes:
- grafana_data:/var/lib/grafana
- ./monitoring/grafana/provisioning:/etc/grafana/provisioning
- ./monitoring/dashboards:/var/lib/grafana/dashboards
environment:
- GF_SECURITY_ADMIN_PASSWORD=${GRAFANA_ADMIN_PASSWORD:-admin}
- GF_USERS_ALLOW_SIGN_UP=false
- GF_SERVER_ROOT_URL=http://localhost:3000
- GF_AUTH_ANONYMOUS_ENABLED=false
restart: unless-stopped
depends_on:
- prometheus
- loki
# Loki - Log aggregation
loki:
image: grafana/loki:latest
container_name: ml-experiments-loki
ports:
- "3100:3100"
volumes:
- ./monitoring/loki-config.yml:/etc/loki/local-config.yaml
- loki_data:/loki
command: -config.file=/etc/loki/local-config.yaml
restart: unless-stopped
# Promtail - Log collector
promtail:
image: grafana/promtail:latest
container_name: ml-experiments-promtail
volumes:
- ./monitoring/promtail-config.yml:/etc/promtail/config.yml
- ./logs:/var/log/app
- /var/lib/docker/containers:/var/lib/docker/containers:ro
- /var/run/docker.sock:/var/run/docker.sock
command: -config.file=/etc/promtail/config.yml
restart: unless-stopped
depends_on:
- loki
volumes:
redis_data:
driver: local
prometheus_data:
driver: local
grafana_data:
driver: local
loki_data:
driver: local