Docker Compose
Docker Compose自动为每个项目创建网络,服务间可以通过服务名互相访问。理解网络配置是编排多服务应用的关键。
# Compose自动创建网络 ✅
# 项目名_default 网络,所有服务默认加入
services:
web:
image: nginx # 可通过"db"访问数据库
db:
image: postgres # 可通过"web"访问Web
# 自定义网络 ✅
services:
web:
networks:
- frontend
- backend # web同时在两个网络
api:
networks:
- backend
db:
networks:
- backend
networks:
frontend:
driver: bridge
backend:
driver: bridge
internal: true # 内部网络,无法访问外网
# 自定义子网 ✅
networks:
frontend:
ipam:
config:
- subnet: 172.20.0.0/16
gateway: 172.20.0.1
# 使用已有网络
networks:
external_net:
external: true
name: my-existing-network
# 使用host网络模式
services:
web:
network_mode: host
# 与另一个容器共享网络
services:
app:
network_mode: "service:vpn"
# 三层网络隔离架构 ✅
services:
nginx:
networks:
- public
api:
networks:
- public
- private
db:
networks:
- private
redis:
networks:
- private
networks:
public:
driver: bridge
private:
driver: bridge
internal: true # db和redis无法访问外网
# 连接外部数据库
services:
api:
extra_hosts:
- "db.external.com:10.0.0.100"
dns:
- 8.8.8.8
- 8.8.4.4
# 使用已存在的网络
services:
app:
networks:
- shared-network
networks:
shared-network:
external: true
❓ 服务间访问报"connection refused"?排查步骤:①确认两个服务在同一网络 ②使用服务名而非localhost ③确认目标服务已启动且监听 ④检查
internal: true是否阻断了出站连接 ⑤用docker compose exec api ping db测试连通性。| 要点 | 说明 | 最佳实践 |
|---|---|---|
| 资源规划 | 根据业务负载合理分配CPU/内存 | 先压测再上线,设置requests和limits |
| 监控告警 | 设置关键指标阈值和告警规则 | Prometheus + AlertManager,5分钟P99延迟告警 |
| 备份策略 | 定期备份关键数据和配置 | 自动化备份脚本 + 异地存储 + 定期恢复演练 |
| 灰度发布 | 新版本逐步放量降低风险 | 金丝雀发布5%→20%→50%→100% |
| 回滚预案 | 部署前确认回滚方案和步骤 | 保留前一版本镜像,数据库迁移向前兼容 |
| 文档更新 | 配置变更必须同步更新文档 | GitOps管理配置,变更即文档 |
| 安全基线 | 遵循CIS Docker Benchmark | 非root运行、只读FS、最小能力 |
| 日志规范 | 结构化日志,统一格式 | JSON格式日志,包含traceId |
# 容器生命周期
docker create/start/stop/restart/rm/pause/unpause
docker logs/top/stats/inspect/exec diff
# 镜像构建
docker build/pull/push/tag/rmi/images/history
docker save/load/import/manifest
# 网络与存储
docker network create/ls/inspect/connect/disconnect/rm/prune
docker volume create/ls/inspect/rm/prune
# Docker Compose
docker compose up/down/ps/logs/build/exec
docker compose config/stop/start/scale/top/cp
# 系统维护
docker system df/prune/info
docker builder prune
docker container/prune/image prune
# K8s常用命令
kubectl get/describe/logs/exec/apply/delete
kubectl rollout status/undo/history
kubectl scale/autoscale/set
version: "3.8" # Compose文件版本
services: # 服务定义(必填)
web:
image: nginx:alpine # 或 build: .
ports:
- "8080:80"
environment:
- KEY=VALUE
volumes:
- ./html:/usr/share/nginx/html:ro
networks:
- frontend
depends_on:
- api
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost/"]
interval: 30s
deploy:
replicas: 2
resources:
limits:
cpus: '0.5'
memory: 256M
restart: unless-stopped
networks: # 网络定义
frontend:
driver: bridge
volumes: # 卷定义
db-data:
secrets: # 密钥定义
db_password:
file: ./db_password.txt
configs: # 配置定义
nginx_config:
file: ./nginx.conf
# Step 1: 检查容器状态
docker ps -a # 查看所有容器
docker inspect <container> # 查看详细配置
# Step 2: 查看日志
docker logs --tail 100 <container> # 最近100行日志
docker logs --since 1h <container> # 最近1小时
# Step 3: 进入容器排查
docker exec -it <container> sh # 进入容器shell
# Step 4: 检查资源
docker stats --no-stream # 资源使用概览
docker system df # 磁盘使用
# Step 5: 网络排查
docker network ls # 网络列表
docker exec <c> ping <target> # 网络连通性
docker exec <c> nslookup <svc> # DNS解析
# Step 6: 检查健康状态
docker inspect --format '{{.State.Health}}' <c>
docker inspect --format '{{.State.ExitCode}}' <c>
docker inspect --format '{{.State.OOMKilled}}' <c>
# 通用最佳实践模板
FROM alpine:3.19 AS builder
# ... 构建步骤 ...
FROM alpine:3.19
RUN addgroup -S appgroup && adduser -S appuser -G appgroup
WORKDIR /app
COPY --from=builder /app/output .
USER appuser
HEALTHCHECK --interval=30s --timeout=3s --retries=3 \
CMD wget -qO- http://localhost:8080/health || exit 1
EXPOSE 8080
CMD ["./app"]
services:
app:
build: .
healthcheck:
test: ["CMD", "wget", "-qO-", "http://localhost:8080/health"]
interval: 30s
timeout: 5s
retries: 3
start_period: 30s
restart: unless-stopped
deploy:
resources:
limits:
cpus: '1'
memory: 512M