Files
0pt/services/infra/docker-compose.data-stack.yml

138 lines
4.9 KiB
YAML

# SIAX Cloud Data Stack - Complete BigQuery Replacement
# All OPEN SOURCE, self-hosted on SIAX Cloud Server 2
#
# Stack:
# ClickHouse - Columnar DB (BigQuery replacement)
# Trino - Distributed SQL query engine
# FlareSolverr - Cloudflare bypass
# Tor - Anonymous proxy rotation
# Apache Superset - BI dashboards
# MinIO - S3 object storage (already exists)
#
version: "3.8"
services:
# ═══════════════════════════════════════════════
# CLICKHOUSE - BigQuery ersättning
# ═══════════════════════════════════════════════
clickhouse:
image: clickhouse/clickhouse-server:24.12-alpine
container_name: clickhouse
restart: unless-stopped
ports:
- "8123:8123" # HTTP API (REST)
- "9000:9000" # Native TCP
- "9090:9090" # Prometheus metrics
volumes:
- clickhouse-data:/var/lib/clickhouse
- clickhouse-logs:/var/log/clickhouse-server
- ./clickhouse/config.xml:/etc/clickhouse-server/config.xml:ro
- ./clickhouse/users.xml:/etc/clickhouse-server/users.xml:ro
environment:
- CLICKHOUSE_DB=opt_scraped
- CLICKHOUSE_USER=default
- CLICKHOUSE_PASSWORD=${CLICKHOUSE_PASSWORD:-changeme}
ulimits:
nofile:
soft: 262144
hard: 262144
deploy:
resources:
limits:
memory: 8G
reservations:
memory: 4G
# ═══════════════════════════════════════════════
# TRINO - Distributed SQL (optional, for complex queries)
# ═══════════════════════════════════════════════
trino:
image: trinodb/trino:470
container_name: trino
restart: unless-stopped
ports:
- "8080:8080"
volumes:
- ./trino/etc:/etc/trino:ro
- trino-data:/data/trino
deploy:
resources:
limits:
memory: 4G
# ═══════════════════════════════════════════════
# FLARESOLVERR - Cloudflare bypass
# ═══════════════════════════════════════════════
flaresolverr:
image: ghcr.io/flaresolverr/flaresolverr:latest
container_name: flaresolverr
restart: unless-stopped
ports:
- "8191:8191"
environment:
- LOG_LEVEL=${LOG_LEVEL:-info}
- TZ=Europe/Stockholm
- CAPTCHA_SOLVER=none
deploy:
resources:
limits:
memory: 1G
# ═══════════════════════════════════════════════
# TOR - Anonymous proxy rotation
# ═══════════════════════════════════════════════
tor-proxy:
image: dperson/torproxy:latest
container_name: tor-proxy
restart: unless-stopped
ports:
- "9050:9050" # SOCKS5
- "9051:9051" # Control port
deploy:
resources:
limits:
memory: 256M
# ═══════════════════════════════════════════════
# APACHE SUPERSET - BI Dashboard
# ═══════════════════════════════════════════════
superset:
image: apache/superset:latest
container_name: superset
restart: unless-stopped
ports:
- "8088:8088"
environment:
- SUPERSET_SECRET_KEY=${SUPERSET_SECRET_KEY:-changeme}
- SUPERSET_DATABASE_URI=postgresql://superset:${POSTGRES_PASSWORD:-changeme}@postgres:5432/superset
volumes:
- superset-data:/app/superset_home
deploy:
resources:
limits:
memory: 2G
depends_on:
- clickhouse
- postgres
# ═══════════════════════════════════════════════
# POSTGRES for Superset metadata
# ═══════════════════════════════════════════════
postgres:
image: postgres:16-alpine
container_name: postgres-superset
restart: unless-stopped
environment:
- POSTGRES_DB=superset
- POSTGRES_USER=superset
- POSTGRES_PASSWORD=${POSTGRES_PASSWORD:-changeme}
volumes:
- postgres-data:/var/lib/postgresql/data
volumes:
clickhouse-data:
clickhouse-logs:
trino-data:
superset-data:
postgres-data: