feat(docker): 生产环境 DevOps 基础设施 — TLS + 备份加密 + Prometheus + Redis 持久化

新增: - nginx/nginx.conf: TLS 1.2/1.3 终端 + HSTS/CSP 安全头 + SSE 长连接 + 50M 上传限制 - prometheus/prometheus.yml: HMS/PostgreSQL/Redis/Nginx 四指标源 - prometheus/alerts.yml: 4 组告警规则（系统/应用/数据库/Redis），含 5xx 错误率 + 内存 + 连接数 - restore.sh: 备份恢复脚本（支持加密备份解密恢复）改进: - backup.sh: 新增 BACKUP_PASSPHRASE 加密（AES-256-CBC）+ 完整性校验 + 恢复指引 - docker-compose.production.yml: 添加 Nginx/Prometheus/Grafana/uploads-backup 容器 - docker-compose.yml: Redis 添加 --appendonly yes 持久化 - .env.production.example: 添加 DevOps 相关环境变量模板
2026-05-21 18:21:51 +08:00
parent 8e616f2210
commit bc571c7749
10 changed files with 458 additions and 9 deletions
--- a/docker/.env.production.example
+++ b/docker/.env.production.example
@@ -50,3 +50,21 @@ ERP__AI__DEFAULT_PROVIDER=ollama
 ERP__AI__API_KEY=
 ERP__AI__BASE_URL=http://localhost:11434
 ERP__AI__MODEL=qwen2.5:7b
+
+# ===== DevOps =====
+
+# 备份加密密码（openssl AES-256-CBC，必填用于生产）
+BACKUP_PASSPHRASE=CHANGE_ME_BACKUP_ENCRYPTION_PASSWORD
+
+# 备份保留天数
+BACKUP_KEEP_DAYS=7
+
+# 备份执行时间（cron 格式）
+BACKUP_CRON=0 2 * * *
+
+# uploads 备份时间
+UPLOADS_BACKUP_CRON=0 3 * * *
+
+# Grafana 管理员密码
+GRAFANA_ADMIN_PASSWORD=CHANGE_ME_GRAFANA_ADMIN
+GRAFANA_ROOT_URL=http://localhost:3001
--- a/docker/backup.sh
+++ b/docker/backup.sh
@@ -1,8 +1,12 @@
 #!/usr/bin/env bash
-# PostgreSQL 自动备份脚本
+# PostgreSQL 自动备份脚本（含加密）
 # 用法:
 #   手动: ./docker/backup.sh
 #   自动: 由 docker compose backup 服务每日 02:00 执行
+#
+# 加密方式（二选一）:
+#   BACKUP_PASSPHRASE — 使用 openssl AES-256-CBC 对称加密（无额外依赖）
+#   GPG_RECIPIENT     — 使用 GPG 非对称加密（需预置公钥）
 set -euo pipefail

 BACKUP_DIR="${BACKUP_DIR:-/backups}"
@@ -13,7 +17,9 @@ PG_DB="${PGDATABSE:-erp}"
 KEEP_DAYS="${KEEP_DAYS:-7}"
 TIMESTAMP=$(date +%Y%m%d_%H%M%S)
 FILENAME="${PG_DB}_${TIMESTAMP}.sql.gz"
+ENCRYPTED_FILENAME="${FILENAME}.enc"
 FILEPATH="${BACKUP_DIR}/${FILENAME}"
+ENCRYPTED_FILEPATH="${BACKUP_DIR}/${ENCRYPTED_FILENAME}"

 mkdir -p "${BACKUP_DIR}"

@@ -36,8 +42,54 @@ else
    exit 1
 fi

-# 清理过期备份
-DELETED=$(find "${BACKUP_DIR}" -name "${PG_DB}_*.sql.gz" -mtime +${KEEP_DAYS} -delete -print | wc -l)
+# ── 加密备份 ──
+if [ -n "${BACKUP_PASSPHRASE:-}" ]; then
+    echo "[$(date -Iseconds)] 使用 AES-256-CBC 加密备份..."
+    if openssl enc -aes-256-cbc -salt -pbkdf2 -pass "pass:${BACKUP_PASSPHRASE}" \
+        -in "${FILEPATH}" -out "${ENCRYPTED_FILEPATH}"; then
+        rm -f "${FILEPATH}"
+        ENC_SIZE=$(du -h "${ENCRYPTED_FILEPATH}" | cut -f1)
+        echo "[$(date -Iseconds)] 加密完成: ${ENCRYPTED_FILENAME} (${ENC_SIZE})"
+    else
+        echo "[$(date -Iseconds)] 加密失败！保留未加密备份" >&2
+        rm -f "${ENCRYPTED_FILEPATH}"
+    fi
+elif [ -n "${GPG_RECIPIENT:-}" ]; then
+    echo "[$(date -Iseconds)] 使用 GPG 加密备份..."
+    if gpg --batch --yes --encrypt --recipient "${GPG_RECIPIENT}" "${FILEPATH}"; then
+        rm -f "${FILEPATH}"
+        ENC_SIZE=$(du -h "${ENCRYPTED_FILEPATH}" | cut -f1)
+        echo "[$(date -Iseconds)] 加密完成: ${ENCRYPTED_FILENAME} (${ENC_SIZE})"
+    else
+        echo "[$(date -Iseconds)] GPG 加密失败！保留未加密备份" >&2
+        rm -f "${FILEPATH}.gpg"
+    fi
+else
+    echo "[$(date -Iseconds)] 警告: 未设置 BACKUP_PASSPHRASE 或 GPG_RECIPIENT，备份未加密！" >&2
+fi
+
+# ── 备份完整性校验 ──
+LATEST_FILE=$(ls -t "${BACKUP_DIR}/${PG_DB}"_*.sql.gz* 2>/dev/null | head -1)
+if [ -n "${LATEST_FILE}" ] && [ -f "${LATEST_FILE}" ]; then
+    if [[ "${LATEST_FILE}" == *.enc ]]; then
+        echo "[$(date -Iseconds)] 加密备份文件存在: $(basename "${LATEST_FILE}")"
+    elif gzip -t "${LATEST_FILE}" 2>/dev/null; then
+        echo "[$(date -Iseconds)] 备份完整性校验通过"
+    else
+        echo "[$(date -Iseconds)] 警告: 备份文件可能损坏: ${LATEST_FILE}" >&2
+    fi
+fi
+
+# ── 清理过期备份 ──
+DELETED=$(find "${BACKUP_DIR}" -name "${PG_DB}_*.sql.gz*" -mtime +${KEEP_DAYS} -delete -print | wc -l)
 if [ "${DELETED}" -gt 0 ]; then
    echo "[$(date -Iseconds)] 已清理 ${DELETED} 个过期备份（>${KEEP_DAYS}天）"
 fi
+
+# ── 恢复指引 ──
+echo ""
+echo "恢复方法:"
+echo "  # 解密（如加密）:"
+echo "  openssl enc -d -aes-256-cbc -pbkdf2 -pass pass:\$BACKUP_PASSPHRASE -in ${ENCRYPTED_FILEPATH} -out ${FILEPATH}"
+echo "  # 恢复:"
+echo "  gunzip -c ${FILEPATH} | psql -h \$PGHOST -U \$PGUSER -d \$PGDB"
--- a/docker/docker-compose.production.yml
+++ b/docker/docker-compose.production.yml
@@ -2,15 +2,44 @@
 # 使用方式: docker compose -f docker/docker-compose.yml -f docker/docker-compose.production.yml up -d

 services:
+  # ── Nginx 反代 + TLS 终端 ──
+  nginx:
+    image: nginx:1.27-alpine
+    container_name: hms-nginx
+    restart: unless-stopped
+    ports:
+      - "80:80"
+      - "443:443"
+    volumes:
+      - ./nginx/nginx.conf:/etc/nginx/conf.d/default.conf:ro
+      - ./nginx/ssl:/etc/nginx/ssl:ro
+      - nginx_logs:/var/log/nginx
+    depends_on:
+      app:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD", "wget", "--spider", "-q", "http://localhost:80"]
+      interval: 30s
+      timeout: 5s
+      retries: 3
+    deploy:
+      resources:
+        limits:
+          cpus: "0.5"
+          memory: 128M
+    networks:
+      - hms-internal
+
+  # ── HMS 应用服务器 ──
  app:
    build:
      context: ..
      dockerfile: Dockerfile
    container_name: hms-server
    restart: unless-stopped
-    ports:
-      - "${APP_PORT:-3000}:3000"
-      - "${METRICS_PORT:-9090}:9090"
+    expose:
+      - "3000"
+      - "9090"
    env_file:
      - .env.production
    environment:
@@ -40,8 +69,7 @@ services:
    networks:
      - hms-internal

-  # 每日自动备份 — 每天凌晨 02:00 执行 pg_dump，保留 7 天
-  # 手动触发: docker compose -f docker/docker-compose.yml -f docker/docker-compose.production.yml run --rm backup
+  # ── 每日自动备份（含加密）──
  backup:
    image: postgres:16-alpine
    container_name: hms-backup
@@ -59,6 +87,7 @@ services:
      BACKUP_DIR: /backups
      KEEP_DAYS: "${BACKUP_KEEP_DAYS:-7}"
      BACKUP_CRON: "${BACKUP_CRON:-0 2 * * *}"
+      BACKUP_PASSPHRASE: "${BACKUP_PASSPHRASE:-}"
    volumes:
      - ./backup.sh:/usr/local/bin/backup.sh:ro
      - backup_data:/backups
@@ -68,11 +97,76 @@ services:
    networks:
      - hms-internal

+  # ── uploads 文件备份（同步到宿主机）──
+  uploads-backup:
+    image: alpine:3.20
+    container_name: hms-uploads-backup
+    restart: unless-stopped
+    entrypoint: >
+      sh -c "
+        echo '$$UPLOADS_BACKUP_CRON rsync -a --delete /source/uploads/ /backup/uploads/' > /etc/crontabs/root &&
+        crond -f -l 2
+      "
+    environment:
+      UPLOADS_BACKUP_CRON: "${UPLOADS_BACKUP_CRON:-0 3 * * *}"
+    volumes:
+      - app-uploads:/source/uploads:ro
+      - uploads_backup_data:/backup/uploads
+    networks:
+      - hms-internal
+
+  # ── Prometheus 监控 ──
+  prometheus:
+    image: prom/prometheus:v3.1.0
+    container_name: hms-prometheus
+    restart: unless-stopped
+    command:
+      - "--config.file=/etc/prometheus/prometheus.yml"
+      - "--storage.tsdb.retention.time=30d"
+      - "--storage.tsdb.retention.size=2GB"
+      - "--web.enable-lifecycle"
+    volumes:
+      - ./prometheus/prometheus.yml:/etc/prometheus/prometheus.yml:ro
+      - ./prometheus/alerts.yml:/etc/prometheus/alerts.yml:ro
+      - prometheus_data:/prometheus
+    expose:
+      - "9090"
+    networks:
+      - hms-internal
+
+  # ── Grafana 可视化 ──
+  grafana:
+    image: grafana/grafana:11.4.0
+    container_name: hms-grafana
+    restart: unless-stopped
+    environment:
+      GF_SECURITY_ADMIN_USER: "${GRAFANA_ADMIN_USER:-admin}"
+      GF_SECURITY_ADMIN_PASSWORD: "${GRAFANA_ADMIN_PASSWORD:-}"
+      GF_USERS_ALLOW_SIGN_UP: "false"
+      GF_SERVER_ROOT_URL: "${GRAFANA_ROOT_URL:-http://localhost:3001}"
+    volumes:
+      - grafana_data:/var/lib/grafana
+      - ./grafana/provisioning:/etc/grafana/provisioning:ro
+    expose:
+      - "3000"
+    depends_on:
+      - prometheus
+    networks:
+      - hms-internal
+
 volumes:
  app-uploads:
    driver: local
  backup_data:
    driver: local
+  uploads_backup_data:
+    driver: local
+  nginx_logs:
+    driver: local
+  prometheus_data:
+    driver: local
+  grafana_data:
+    driver: local

 networks:
  hms-internal:
--- a/docker/docker-compose.yml
+++ b/docker/docker-compose.yml
@@ -29,7 +29,7 @@ services:
  redis:
    image: redis:7-alpine
    container_name: erp-redis
-    command: redis-server --requirepass ${REDIS_PASSWORD:-erp_redis_dev}
+    command: redis-server --requirepass ${REDIS_PASSWORD:-erp_redis_dev} --appendonly yes
    expose:
      - "6379"
    volumes:
--- a/docker/nginx/nginx.conf
+++ b/docker/nginx/nginx.conf
@@ -0,0 +1,96 @@
+upstream hms_backend {
+    server app:3000;
+    keepalive 32;
+}
+
+server {
+    listen 80;
+    server_name _;
+    return 301 https://$host$request_uri;
+}
+
+server {
+    listen 443 ssl http2;
+    server_name _;
+
+    # ── TLS ──
+    ssl_certificate     /etc/nginx/ssl/fullchain.pem;
+    ssl_certificate_key /etc/nginx/ssl/privkey.pem;
+    ssl_protocols       TLSv1.2 TLSv1.3;
+    ssl_ciphers         ECDHE-ECDSA-AES128-GCM-SHA256:ECDHE-RSA-AES128-GCM-SHA256:ECDHE-ECDSA-AES256-GCM-SHA384:ECDHE-RSA-AES256-GCM-SHA384;
+    ssl_prefer_server_ciphers off;
+    ssl_session_cache   shared:SSL:10m;
+    ssl_session_timeout 1d;
+    ssl_session_tickets off;
+
+    # ── 安全头 ──
+    add_header Strict-Transport-Security "max-age=63072000; includeSubDomains; preload" always;
+    add_header X-Content-Type-Options "nosniff" always;
+    add_header X-Frame-Options "DENY" always;
+    add_header X-XSS-Protection "1; mode=block" always;
+    add_header Referrer-Policy "strict-origin-when-cross-origin" always;
+    add_header Permissions-Policy "camera=(), microphone=(), geolocation=()" always;
+    add_header Content-Security-Policy "default-src 'self'; script-src 'self' 'unsafe-inline'; style-src 'self' 'unsafe-inline'; img-src 'self' data: blob:; font-src 'self'; connect-src 'self' wss:; frame-ancestors 'none'" always;
+
+    # ── 日志 ──
+    access_log /var/log/nginx/hms_access.log;
+    error_log  /var/log/nginx/hms_error.log warn;
+
+    # ── 上传文件（化验单/体检报告）──
+    location /uploads/ {
+        proxy_pass http://hms_backend;
+        proxy_set_header Host $host;
+        proxy_set_header X-Real-IP $remote_addr;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+        proxy_set_header X-Forwarded-Proto $scheme;
+        # 大文件上传限制
+        client_max_body_size 50m;
+    }
+
+    # ── SSE（消息推送/AI 分析）──
+    location ~ ^/api/v1/(message|ai)/.*sse {
+        proxy_pass http://hms_backend;
+        proxy_set_header Host $host;
+        proxy_set_header X-Real-IP $remote_addr;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+        proxy_set_header X-Forwarded-Proto $scheme;
+        proxy_http_version 1.1;
+        proxy_set_header Connection "";
+        proxy_buffering off;
+        proxy_cache off;
+        proxy_read_timeout 86400s;
+        chunked_transfer_encoding on;
+    }
+
+    # ── API 反代 ──
+    location /api/ {
+        proxy_pass http://hms_backend;
+        proxy_set_header Host $host;
+        proxy_set_header X-Real-IP $remote_addr;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+        proxy_set_header X-Forwarded-Proto $scheme;
+        proxy_http_version 1.1;
+        proxy_set_header Connection "";
+        client_max_body_size 50m;
+    }
+
+    # ── 健康检查 ──
+    location /health {
+        proxy_pass http://hms_backend/api/v1/health;
+        access_log off;
+    }
+
+    # ── 指标（仅内网可访问）──
+    location /metrics {
+        # 生产环境应限制为 Prometheus 访问
+        allow 172.16.0.0/12;
+        allow 10.0.0.0/8;
+        deny all;
+        proxy_pass http://hms_backend:9090/metrics;
+        access_log off;
+    }
+
+    location / {
+        return 404;
+    }
+}
--- a/docker/nginx/ssl/.gitignore
+++ b/docker/nginx/ssl/.gitignore
@@ -0,0 +1,3 @@
+*
+!.gitkeep
+!.gitignore
--- a/docker/nginx/ssl/.gitkeep
+++ b/docker/nginx/ssl/.gitkeep
@@ -0,0 +1,8 @@
+# 将 SSL 证书放置在此目录
+# 必需文件: fullchain.pem + privkey.pem
+# 生产环境建议使用 Let's Encrypt 或云服务商证书管理
+#
+# Let's Encrypt 示例:
+#   certbot certonly --standalone -d your-domain.com
+#   cp /etc/letsencrypt/live/your-domain.com/fullchain.pem .
+#   cp /etc/letsencrypt/live/your-domain.com/privkey.pem .
--- a/docker/prometheus/alerts.yml
+++ b/docker/prometheus/alerts.yml
@@ -0,0 +1,103 @@
+groups:
+  # ── 系统级告警 ──
+  - name: system
+    rules:
+      - alert: HMSHighMemoryUsage
+        expr: process_resident_memory_bytes > 800000000
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          summary: "HMS 内存使用超过 800MB"
+          description: "当前值: {{ $value | humanize }}B"
+
+      - alert: HMSHighMemoryCritical
+        expr: process_resident_memory_bytes > 1000000000
+        for: 2m
+        labels:
+          severity: critical
+        annotations:
+          summary: "HMS 内存使用超过 1GB（危险）"
+          description: "当前值: {{ $value | humanize }}B"
+
+      - alert: HMSHighCPU
+        expr: rate(process_cpu_seconds_total[5m]) > 0.8
+        for: 10m
+        labels:
+          severity: warning
+        annotations:
+          summary: "HMS CPU 使用率超过 80%"
+
+  # ── 应用级告警 ──
+  - name: application
+    rules:
+      - alert: HMSHighErrorRate
+        expr: rate(http_requests_total{status=~"5.."}[5m]) / rate(http_requests_total[5m]) > 0.05
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          summary: "API 5xx 错误率超过 5%"
+          description: "当前错误率: {{ $value | humanizePercentage }}"
+
+      - alert: HMSSlowResponses
+        expr: histogram_quantile(0.95, rate(http_request_duration_seconds_bucket[5m])) > 2
+        for: 10m
+        labels:
+          severity: warning
+        annotations:
+          summary: "95% 请求响应时间超过 2 秒"
+
+      - alert: HMSInstanceDown
+        expr: up{job="hms"} == 0
+        for: 2m
+        labels:
+          severity: critical
+        annotations:
+          summary: "HMS 服务不可达"
+
+  # ── 数据库告警 ──
+  - name: database
+    rules:
+      - alert: HMSPostgresConnectionsHigh
+        expr: pg_stat_activity_count > 80
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          summary: "PostgreSQL 活跃连接数超过 80"
+
+      - alert: HMSPostgresReplicationLag
+        expr: pg_replication_lag > 30
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          summary: "PostgreSQL 复制延迟超过 30 秒"
+
+      - alert: HMSBackupMissing
+        expr: time() - hms_last_backup_timestamp > 86400 * 2
+        for: 1h
+        labels:
+          severity: critical
+        annotations:
+          summary: "数据库备份超过 48 小时未执行"
+
+  # ── Redis 告警 ──
+  - name: redis
+    rules:
+      - alert: HMSRedisMemoryHigh
+        expr: redis_memory_used_bytes / redis_memory_max_bytes > 0.9
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          summary: "Redis 内存使用超过 90%"
+
+      - alert: HMSRedisDown
+        expr: redis_up == 0
+        for: 2m
+        labels:
+          severity: critical
+        annotations:
+          summary: "Redis 服务不可达"
--- a/docker/prometheus/prometheus.yml
+++ b/docker/prometheus/prometheus.yml
@@ -0,0 +1,32 @@
+global:
+  scrape_interval: 15s
+  evaluation_interval: 15s
+
+rule_files:
+  - "alerts.yml"
+
+scrape_configs:
+  - job_name: "hms"
+    metrics_path: /metrics
+    static_configs:
+      - targets: ["app:9090"]
+        labels:
+          service: "hms-server"
+
+  - job_name: "postgres"
+    static_configs:
+      - targets: ["postgres-exporter:9187"]
+        labels:
+          service: "postgresql"
+
+  - job_name: "redis"
+    static_configs:
+      - targets: ["redis-exporter:9121"]
+        labels:
+          service: "redis"
+
+  - job_name: "nginx"
+    static_configs:
+      - targets: ["nginx-exporter:9113"]
+        labels:
+          service: "nginx"
--- a/docker/restore.sh
+++ b/docker/restore.sh
@@ -0,0 +1,43 @@
+#!/usr/bin/env bash
+# PostgreSQL 备份恢复脚本
+# 用法: BACKUP_PASSPHRASE=xxx ./docker/restore.sh /backups/erp_20260521_020000.sql.gz.enc
+set -euo pipefail
+
+BACKUP_FILE="${1:?用法: restore.sh <备份文件路径>}"
+PG_HOST="${PGHOST:-postgres}"
+PG_PORT="${PGPORT:-5432}"
+PG_USER="${PGUSER:-erp}"
+PG_DB="${PGDATABASE:-erp}"
+
+if [ ! -f "${BACKUP_FILE}" ]; then
+    echo "错误: 文件不存在: ${BACKUP_FILE}" >&2
+    exit 1
+fi
+
+echo "[$(date -Iseconds)] 恢复目标: ${PG_HOST}:${PG_PORT}/${PG_DB}"
+echo "[$(date -Iseconds)] 备份文件: ${BACKUP_FILE}"
+
+# 解密（如果是加密文件）
+if [[ "${BACKUP_FILE}" == *.enc ]]; then
+    if [ -z "${BACKUP_PASSPHRASE:-}" ]; then
+        echo "错误: 加密备份需要设置 BACKUP_PASSPHRASE 环境变量" >&2
+        exit 1
+    fi
+    DECRYPTED="${BACKUP_FILE%.enc}"
+    echo "[$(date -Iseconds)] 解密中..."
+    openssl enc -d -aes-256-cbc -pbkdf2 -pass "pass:${BACKUP_PASSPHRASE}" \
+        -in "${BACKUP_FILE}" -out "${DECRYPTED}"
+    BACKUP_FILE="${DECRYPTED}"
+fi
+
+# 解压并恢复
+echo "[$(date -Iseconds)] 恢复中..."
+gunzip -c "${BACKUP_FILE}" | psql -h "${PG_HOST}" -p "${PG_PORT}" -U "${PG_USER}" -d "${PG_DB}"
+
+echo "[$(date -Iseconds)] 恢复完成"
+
+# 清理解密文件
+if [ -n "${DECRYPTED:-}" ] && [ -f "${DECRYPTED}" ]; then
+    rm -f "${DECRYPTED}"
+    echo "[$(date -Iseconds)] 已清理解密临时文件"
+fi