Files
crawle-snapp/docker-compose.yml
T
nfel a436b72bec Use Playwright base image for worker services
Replace chromium/chromedriver apt installs with
mcr.microsoft.com/playwright:v1.62.0-noble for worker containers.
Admin service stays on python:3.11-slim since it doesn't need Chrome.

This removes the build-time patch_driver.py step since Playwright ships
with a pre-patched Chromium binary. Also split docker-compose into
separate images (seed-admin / seed-worker) with explicit build targets.

Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
2026-08-04 21:19:12 +03:30

107 lines
2.7 KiB
YAML

services:
# ── Redis (state + job queue) ─────────────────────────────────────────────────
redis:
image: redis:7-alpine
container_name: seed-redis
restart: unless-stopped
volumes:
- redis-data:/data
healthcheck:
test: ["CMD", "redis-cli", "ping"]
interval: 10s
timeout: 3s
retries: 5
# ── Admin panel (FastAPI) ─────────────────────────────────────────────────────
admin:
build:
context: .
dockerfile: Dockerfile
target: admin
image: seed-admin:latest
hostname: crawler-admin
container_name: seed-admin
restart: unless-stopped
command: python main.py admin
ports:
- "${COMPOSE_ADMIN_PORT:-127.0.0.0:8800}:8000"
env_file:
- .env
environment:
ADMIN_HOST: "0.0.0.0"
DB_PATH: /app/data/tracker.db
LOG_FILE: /app/logs/seed.log
REDIS_URL: "redis://redis:6379/0"
SKIP_INTERNAL_WORKER: "true"
HEADLESS: "true"
RECORDINGS_DIR: /app/recordings
volumes:
- seed-data:/app/data
- seed-logs:/app/logs
- seed-recordings:/app/recordings
depends_on:
redis:
condition: service_healthy
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost:8000/"]
interval: 30s
timeout: 5s
retries: 3
start_period: 10s
networks:
- proxy-network
- default
# ── RQ worker (Chrome crawler, runs batch jobs) ───────────────────────────────
worker:
build:
context: .
dockerfile: Dockerfile
target: worker
image: seed-worker:latest
restart: unless-stopped
deploy:
replicas: 3
command: ["rq", "worker", "--url", "redis://redis:6379/0", "batch"]
env_file:
- .env
environment:
DB_PATH: /app/data/tracker.db
LOG_FILE: /app/logs/seed.log
REDIS_URL: "redis://redis:6379/0"
HEADLESS: "true"
RECORD_SESSIONS: "true"
RECORDING_FPS: "12"
RECORDING_MAX_FILES: "3"
RECORDINGS_DIR: /app/recordings
volumes:
- seed-data:/app/data
- seed-logs:/app/logs
- seed-recordings:/app/recordings
depends_on:
redis:
condition: service_healthy
shm_size: "2gb"
cap_add:
- SYS_ADMIN
security_opt:
- seccomp:unconfined
networks:
proxy-network:
external: true
default:
name: crawle-snapp_default
volumes:
seed-data:
driver: local
seed-logs:
driver: local
seed-recordings:
driver: local
redis-data:
driver: local