4、集成环境搭建：集成测试环境硬件/软件要求、Docker容器化部署、网络与数据库配置

集成环境是Testing驱动审批流回放验证的物理基础。本章从硬件/软件选型、Docker容器化部署、网络与数据库配置三个维度，给出可复现的搭建规范。所有配置均以“单机多容器+外部数据库”为基准拓扑，兼顾成本与回放性能。

4.1 硬件与软件要求

集成测试环境不需要生产级冗余，但必须保证CPU、内存、磁盘IO能够支撑至少3路并发回放线程。下表列出推荐配置与最低配置的差异：

4.2 Docker容器化部署

采用Docker Compose编排四个核心服务：回放引擎、审批流模拟器、配置中心、监控代理。每个容器均绑定资源限制与健康检查。

version: '3.8'
services:
  replay-engine:
    image: registry.example.com/replay-engine:4.2.0
    container_name: re-engine
    ports:
      - "8080:8080"
    environment:
      - DB_HOST=postgres
      - DB_PORT=5432
      - DB_NAME=replaydb
      - REPLAY_THREADS=4
    deploy:
      resources:
        limits:
          cpus: '2.0'
          memory: 4G
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8080/actuator/health"]
      interval: 15s
      retries: 3
    depends_on:
      postgres:
        condition: service_healthy

  approval-simulator:
    image: registry.example.com/approval-sim:3.1.0
    container_name: app-sim
    ports:
      - "9090:9090"
    environment:
      - SIM_MODE=replay
      - APPROVAL_RULES_PATH=/config/rules.json
    volumes:
      - ./config/rules.json:/config/rules.json:ro
    deploy:
      resources:
        limits:
          cpus: '1.5'
          memory: 2G

  config-center:
    image: consul:1.18
    container_name: config-ctr
    ports:
      - "8500:8500"
    command: agent -server -bootstrap-expect=1 -client=0.0.0.0

  postgres:
    image: postgres:16-alpine
    container_name: pg-replay
    environment:
      POSTGRES_DB: replaydb
      POSTGRES_USER: replay
      POSTGRES_PASSWORD: ${DB_PASSWORD}
    volumes:
      - pgdata:/var/lib/postgresql/data
      - ./init-scripts:/docker-entrypoint-initdb.d:ro
    deploy:
      resources:
        limits:
          cpus: '2.0'
          memory: 8G
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U replay -d replaydb"]
      interval: 10s

volumes:
  pgdata:

下图展示容器间的调用关系与端口映射：

4.3 网络配置

集成环境使用Docker内置的bridge网络，并显式定义子网与网关，避免与宿主机网段冲突。同时为每个服务分配固定IP，方便监控系统采集指标。

networks:
  replay-net:
    driver: bridge
    ipam:
      config:
        - subnet: 172.28.0.0/16
          gateway: 172.28.0.1

services:
  replay-engine:
    networks:
      replay-net:
        ipv4_address: 172.28.0.10
  approval-simulator:
    networks:
      replay-net:
        ipv4_address: 172.28.0.11
  config-center:
    networks:
      replay-net:
        ipv4_address: 172.28.0.12
  postgres:
    networks:
      replay-net:
        ipv4_address: 172.28.0.20

4.4 数据库配置

PostgreSQL是回放数据的核心存储。以下配置在 postgresql.conf 中必须调整，以匹配回放场景的高吞吐写入：

初始化脚本 init.sql 需创建回放专用的表结构与索引：

CREATE TABLE IF NOT EXISTS replay_events (
    event_id BIGSERIAL PRIMARY KEY,
    flow_id VARCHAR(64) NOT NULL,
    step_name VARCHAR(128) NOT NULL,
    event_type VARCHAR(32) NOT NULL,
    payload JSONB,
    created_at TIMESTAMPTZ DEFAULT NOW()
);

CREATE INDEX idx_replay_events_flow_id ON replay_events(flow_id);
CREATE INDEX idx_replay_events_created_at ON replay_events(created_at);

CREATE TABLE IF NOT EXISTS replay_progress (
    flow_id VARCHAR(64) PRIMARY KEY,
    last_event_id BIGINT NOT NULL,
    status VARCHAR(16) DEFAULT 'running',
    updated_at TIMESTAMPTZ DEFAULT NOW()
);

-- 加载pg_stat_statements扩展
CREATE EXTENSION IF NOT EXISTS pg_stat_statements;

下图展示数据库连接池与回放引擎的交互流程：

4.5 环境验证清单

搭建完成后，按以下步骤验证环境可用性：

1. 执行 docker compose ps 确认所有容器状态为 Up (healthy)。
2. 使用 curl http://localhost:8080/actuator/health 检查回放引擎健康端点。
3. 连接数据库并执行 SELECT count(*) FROM replay_events; 返回0行。
4. 运行一次最小回放测试：docker exec re-engine ./replay --dry-run，观察日志无ERROR。
5. 检查 pg_stat_statements 视图，确认回放引擎的SQL调用已被记录。