Step 21: 智能客服 SaaS 平台 —— 生产部署¶
目标：完成 Docker 容器化、K8s 部署、监控配置、性能优化
003e 难度：⭐⭐⭐⭐ | 代码量：约 800 行（配置）| 预计学习时间：3-4 小时
一、Docker 容器化¶
1.1 多阶段 Dockerfile¶
# ═══════════════════════════════════════════════════════════
# 阶段 1：构建环境
# ═══════════════════════════════════════════════════════════
FROM ubuntu:22.04 AS builder
# 安装编译依赖
RUN apt-get update && apt-get install -y \
build-essential \
cmake \
git \
libboost-all-dev \
libssl-dev \
libpqxx-dev \
&& rm -rf /var/lib/apt/lists/*
# 工作目录
WORKDIR /build
# 复制源码
COPY CMakeLists.txt .
COPY src ./src
COPY config ./config
# 编译（Release 模式，启用优化）
RUN mkdir -p build && cd build \
&& cmake .. \
-DCMAKE_BUILD_TYPE=Release \
-DCMAKE_CXX_FLAGS="-O3 -march=native -DNDEBUG" \
&& make -j$(nproc)
# 运行测试
RUN cd build && ctest --output-on-failure
# ═══════════════════════════════════════════════════════════
# 阶段 2：运行环境
# ═══════════════════════════════════════════════════════════
FROM ubuntu:22.04
# 安装运行时依赖（最小化）
RUN apt-get update && apt-get install -y \
libboost-system1.74.0 \
libboost-thread1.74.0 \
libssl3 \
libpq5 \
curl \
&& rm -rf /var/lib/apt/lists/* \
&& apt-get clean
# 创建非 root 用户
RUN useradd -m -u 1000 -s /bin/bash smartsupport
# 工作目录
WORKDIR /app
# 从构建阶段复制产物
COPY --from=builder /build/build/smartsupport ./
COPY --from=builder /build/config ./config
# 创建数据和日志目录
RUN mkdir -p /app/data /app/logs \
&& chown -R smartsupport:smartsupport /app
# 切换到非 root 用户
USER smartsupport
# 暴露端口
EXPOSE 8080 9090
# 健康检查
HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
CMD curl -f http://localhost:8080/health || exit 1
# 启动命令
ENTRYPOINT ["./smartsupport"]
CMD ["--config", "./config/production.yaml"]
1.2 Docker Compose 本地部署¶
# docker-compose.yml
version: '3.8'
services:
# ───────────────────────────────────────────────────────
# SmartSupport 主服务
# ───────────────────────────────────────────────────────
app:
build:
context: .
dockerfile: Dockerfile
image: smartsupport:latest
container_name: smartsupport-app
restart: unless-stopped
ports:
- "8080:8080" # HTTP API
- "9090:9090" # Metrics
environment:
- DATABASE_URL=postgresql://postgres:postgres@postgres:5432/smartsupport
- REDIS_URL=redis://redis:6379
- OPENAI_API_KEY=${OPENAI_API_KEY}
- LOG_LEVEL=info
volumes:
- ./logs:/app/logs
- ./data:/app/data
depends_on:
postgres:
condition: service_healthy
redis:
condition: service_healthy
networks:
- smartsupport-network
deploy:
resources:
limits:
cpus: '2'
memory: 2G
reservations:
cpus: '0.5'
memory: 512M
# ───────────────────────────────────────────────────────
# PostgreSQL（主数据库 + 向量扩展）
# ───────────────────────────────────────────────────────
postgres:
image: ankane/pgvector:latest
container_name: smartsupport-postgres
restart: unless-stopped
environment:
POSTGRES_USER: postgres
POSTGRES_PASSWORD: postgres
POSTGRES_DB: smartsupport
volumes:
- postgres-data:/var/lib/postgresql/data
- ./init-scripts:/docker-entrypoint-initdb.d:ro
ports:
- "5432:5432"
healthcheck:
test: ["CMD-SHELL", "pg_isready -U postgres"]
interval: 5s
timeout: 5s
retries: 5
networks:
- smartsupport-network
# ───────────────────────────────────────────────────────
# Redis（缓存 + 会话存储）
# ───────────────────────────────────────────────────────
redis:
image: redis:7-alpine
container_name: smartsupport-redis
restart: unless-stopped
command: redis-server --appendonly yes --maxmemory 256mb --maxmemory-policy allkeys-lru
volumes:
- redis-data:/data
ports:
- "6379:6379"
healthcheck:
test: ["CMD", "redis-cli", "ping"]
interval: 5s
timeout: 3s
retries: 5
networks:
- smartsupport-network
# ───────────────────────────────────────────────────────
# MinIO（对象存储）
# ───────────────────────────────────────────────────────
minio:
image: minio/minio:latest
container_name: smartsupport-minio
restart: unless-stopped
command: server /data --console-address ":9001"
environment:
MINIO_ROOT_USER: minioadmin
MINIO_ROOT_PASSWORD: minioadmin
volumes:
- minio-data:/data
ports:
- "9000:9000"
- "9001:9001"
networks:
- smartsupport-network
# ───────────────────────────────────────────────────────
# Prometheus（监控）
# ───────────────────────────────────────────────────────
prometheus:
image: prom/prometheus:latest
container_name: smartsupport-prometheus
restart: unless-stopped
volumes:
- ./monitoring/prometheus.yml:/etc/prometheus/prometheus.yml:ro
- prometheus-data:/prometheus
command:
- '--config.file=/etc/prometheus/prometheus.yml'
- '--storage.tsdb.path=/prometheus'
- '--web.console.libraries=/usr/share/prometheus/console_libraries'
- '--web.console.templates=/usr/share/prometheus/consoles'
- '--web.enable-lifecycle'
ports:
- "9091:9090"
networks:
- smartsupport-network
# ───────────────────────────────────────────────────────
# Grafana（可视化）
# ───────────────────────────────────────────────────────
grafana:
image: grafana/grafana:latest
container_name: smartsupport-grafana
restart: unless-stopped
environment:
- GF_SECURITY_ADMIN_PASSWORD=${GRAFANA_PASSWORD:-admin}
- GF_USERS_ALLOW_SIGN_UP=false
volumes:
- grafana-data:/var/lib/grafana
- ./monitoring/grafana/dashboards:/etc/grafana/provisioning/dashboards:ro
- ./monitoring/grafana/datasources:/etc/grafana/provisioning/datasources:ro
ports:
- "3000:3000"
networks:
- smartsupport-network
depends_on:
- prometheus
volumes:
postgres-data:
redis-data:
minio-data:
prometheus-data:
grafana-data:
networks:
smartsupport-network:
driver: bridge
二、Kubernetes 部署¶
2.1 Namespace 和 ConfigMap¶
# k8s/00-namespace.yaml
apiVersion: v1
kind: Namespace
metadata:
name: smartsupport
labels:
name: smartsupport
---
# k8s/01-configmap.yaml
apiVersion: v1
kind: ConfigMap
metadata:
name: smartsupport-config
namespace: smartsupport
data:
DATABASE_URL: "postgresql://postgres:postgres@postgres:5432/smartsupport"
REDIS_URL: "redis://redis:6379"
LOG_LEVEL: "info"
METRICS_PORT: "9090"
HTTP_PORT: "8080"
2.2 应用部署¶
# k8s/10-deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: smartsupport-app
namespace: smartsupport
labels:
app: smartsupport
component: api
spec:
replicas: 3 # 3 个副本实现高可用
strategy:
type: RollingUpdate
rollingUpdate:
maxSurge: 1
maxUnavailable: 0
selector:
matchLabels:
app: smartsupport
component: api
template:
metadata:
labels:
app: smartsupport
component: api
annotations:
prometheus.io/scrape: "true"
prometheus.io/port: "9090"
prometheus.io/path: "/metrics"
spec:
containers:
- name: app
image: smartsupport:latest
imagePullPolicy: Always
ports:
- name: http
containerPort: 8080
protocol: TCP
- name: metrics
containerPort: 9090
protocol: TCP
envFrom:
- configMapRef:
name: smartsupport-config
env:
- name: OPENAI_API_KEY
valueFrom:
secretKeyRef:
name: smartsupport-secrets
key: openai-api-key
resources:
requests:
memory: "512Mi"
cpu: "500m"
limits:
memory: "2Gi"
cpu: "2000m"
livenessProbe:
httpGet:
path: /health
port: http
initialDelaySeconds: 30
periodSeconds: 10
readinessProbe:
httpGet:
path: /ready
port: http
initialDelaySeconds: 5
periodSeconds: 5
volumeMounts:
- name: logs
mountPath: /app/logs
volumes:
- name: logs
emptyDir: {}
---
# k8s/11-service.yaml
apiVersion: v1
kind: Service
metadata:
name: smartsupport-service
namespace: smartsupport
labels:
app: smartsupport
spec:
type: ClusterIP
ports:
- port: 8080
targetPort: http
protocol: TCP
name: http
- port: 9090
targetPort: metrics
protocol: TCP
name: metrics
selector:
app: smartsupport
component: api
---
# k8s/12-ingress.yaml
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: smartsupport-ingress
namespace: smartsupport
annotations:
nginx.ingress.kubernetes.io/ssl-redirect: "true"
nginx.ingress.kubernetes.io/proxy-body-size: "10m"
nginx.ingress.kubernetes.io/rate-limit: "100"
spec:
ingressClassName: nginx
tls:
- hosts:
- api.smartsupport.io
secretName: smartsupport-tls
rules:
- host: api.smartsupport.io
http:
paths:
- path: /
pathType: Prefix
backend:
service:
name: smartsupport-service
port:
number: 8080
2.3 数据库（StatefulSet）¶
# k8s/20-postgres.yaml
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: postgres
namespace: smartsupport
spec:
serviceName: postgres
replicas: 1
selector:
matchLabels:
app: postgres
template:
metadata:
labels:
app: postgres
spec:
containers:
- name: postgres
image: ankane/pgvector:latest
env:
- name: POSTGRES_USER
value: postgres
- name: POSTGRES_PASSWORD
valueFrom:
secretKeyRef:
name: smartsupport-secrets
key: db-password
- name: POSTGRES_DB
value: smartsupport
ports:
- containerPort: 5432
volumeMounts:
- name: postgres-storage
mountPath: /var/lib/postgresql/data
volumeClaimTemplates:
- metadata:
name: postgres-storage
spec:
accessModes: ["ReadWriteOnce"]
resources:
requests:
storage: 10Gi
---
apiVersion: v1
kind: Service
metadata:
name: postgres
namespace: smartsupport
spec:
selector:
app: postgres
ports:
- port: 5432
2.4 HPA 自动扩缩容¶
# k8s/30-hpa.yaml
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
name: smartsupport-hpa
namespace: smartsupport
spec:
scaleTargetRef:
apiVersion: apps/v1
kind: Deployment
name: smartsupport-app
minReplicas: 3
maxReplicas: 20
metrics:
- type: Resource
resource:
name: cpu
target:
type: Utilization
averageUtilization: 70
- type: Resource
resource:
name: memory
target:
type: Utilization
averageUtilization: 80
behavior:
scaleUp:
stabilizationWindowSeconds: 60
policies:
- type: Percent
value: 100
periodSeconds: 15
scaleDown:
stabilizationWindowSeconds: 300
policies:
- type: Percent
value: 10
periodSeconds: 60
三、监控配置¶
3.1 Prometheus 配置¶
# monitoring/prometheus.yml
global:
scrape_interval: 15s
evaluation_interval: 15s
alerting:
alertmanagers:
- static_configs:
- targets: ['alertmanager:9093']
rule_files:
- /etc/prometheus/rules/*.yml
scrape_configs:
# SmartSupport 应用
- job_name: 'smartsupport'
kubernetes_sd_configs:
- role: pod
namespaces:
names:
- smartsupport
relabel_configs:
- source_labels: [__meta_kubernetes_pod_annotation_prometheus_io_scrape]
action: keep
regex: true
- source_labels: [__meta_kubernetes_pod_annotation_prometheus_io_port]
action: replace
target_label: __address__
regex: ([^:]+)(?::\d+)?;(\d+)
replacement: $1:$2
# PostgreSQL
- job_name: 'postgres'
static_configs:
- targets: ['postgres-exporter:9187']
# Redis
- job_name: 'redis'
static_configs:
- targets: ['redis-exporter:9121']
# 节点监控
- job_name: 'node'
static_configs:
- targets: ['node-exporter:9100']
3.2 告警规则¶
# monitoring/rules/alerts.yml
groups:
- name: smartsupport
rules:
# 高错误率告警
- alert: HighErrorRate
expr: rate(http_requests_total{status=~"5.."}[5m]) / rate(http_requests_total[5m]) > 0.05
for: 5m
labels:
severity: critical
annotations:
summary: "High error rate detected"
description: "Error rate is above 5% for more than 5 minutes"
# 高延迟告警
- alert: HighLatency
expr: histogram_quantile(0.95, rate(http_request_duration_seconds_bucket[5m])) > 2
for: 5m
labels:
severity: warning
annotations:
summary: "High latency detected"
description: "95th percentile latency is above 2 seconds"
# 服务不可用
- alert: ServiceDown
expr: up{job="smartsupport"} == 0
for: 1m
labels:
severity: critical
annotations:
summary: "Service is down"
description: "SmartSupport service is not responding"
# 数据库连接池耗尽
- alert: DBPoolExhausted
expr: db_pool_active_connections / db_pool_max_connections > 0.9
for: 5m
labels:
severity: warning
annotations:
summary: "Database connection pool almost exhausted"
四、性能优化¶
4.1 数据库优化¶
-- 添加索引
CREATE INDEX CONCURRENTLY idx_messages_tenant_time
ON messages(tenant_id, created_at DESC);
CREATE INDEX CONCURRENTLY idx_conversations_tenant_status
ON conversations(tenant_id, status)
WHERE status = 'active';
-- 向量索引（pgvector）
CREATE INDEX CONCURRENTLY idx_knowledge_embedding
ON knowledge_docs
USING ivfflat (embedding vector_cosine_ops)
WITH (lists = 100);
-- 分区表（按时间）
CREATE TABLE messages_2024_03 PARTITION OF messages
FOR VALUES FROM ('2024-03-01') TO ('2024-04-01');
4.2 应用层优化¶
// 连接池配置
class ConnectionPool {
public:
struct Config {
size_t min_size = 5; // 最小连接数
size_t max_size = 50; // 最大连接数
std::chrono::seconds max_idle_time{300}; // 最大空闲时间
std::chrono::seconds connection_timeout{5}; // 连接超时
};
};
// 缓存策略
class CacheManager {
public:
// 租户配置缓存（很少变化）
LRUCache<std::string, TenantConfig> tenant_config_cache{1000, std::chrono::minutes(5)};
// 知识库热点缓存
LRUCache<std::string, std::vector<KnowledgeDoc>> knowledge_cache{100, std::chrono::minutes(1)};
// 会话状态缓存（Redis）
std::shared_ptr<RedisClient> session_cache;
};
// 异步批量处理
class BatchProcessor {
public:
// 批量写入日志
void enqueue_log(const LogEntry& entry);
// 批量记录用量
void enqueue_usage(const UsageRecord& record);
// 定时刷新（每秒）
void flush_batch();
};
五、压测与调优¶
5.1 压测脚本¶
#!/bin/bash
# benchmark.sh
# 安装 k6
# wget -qO- https://dl.k6.io/key.gpg | gpg --dearmor | sudo tee /usr/share/keyrings/k6-archive-keyring.gpg
# echo "deb [signed-by=/usr/share/keyrings/k6-archive-keyring.gpg] https://dl.k6.io/deb stable main" | sudo tee /etc/apt/sources.list.d/k6.list
# sudo apt-get update
# sudo apt-get install k6
# 压测脚本
k6 run --vus 100 --duration 5m - <<EOF
import http from 'k6/http';
import { check, sleep } from 'k6';
export default function () {
const payload = JSON.stringify({
message: "你好，我想了解产品价格",
tenant_id: "test-tenant"
});
const res = http.post('http://localhost:8080/api/chat', payload, {
headers: { 'Content-Type': 'application/json' },
});
check(res, {
'status is 200': (r) => r.status === 200,
'response time < 3s': (r) => r.timings.duration < 3000,
});
sleep(1);
}
EOF
5.2 调优清单¶
压测结果分析：
┌─────────────────┬────────────┬────────────┬──────────┐
│ 指标 │ 目标 │ 当前 │ 状态 │
├─────────────────┼────────────┼────────────┼──────────┤
│ P99 延迟 │ < 3s │ 2.5s │ ✅ 达标 │
│ 吞吐量 │ > 1000 QPS │ 1200 QPS │ ✅ 达标 │
│ 错误率 │ < 1% │ 0.5% │ ✅ 达标 │
│ CPU 使用率 │ < 70% │ 65% │ ✅ 达标 │
│ 内存使用 │ < 2GB │ 1.8GB │ ✅ 达标 │
└─────────────────┴────────────┴────────────┴──────────┘
优化措施：
1. 数据库连接池从 10 调整到 50（减少连接等待）
2. 添加 Redis 缓存层（命中 60%）
3. 启用 HTTP Keep-Alive（减少 TCP 握手）
4. RAG Top-K 从 10 降到 5（减少向量检索时间）
六、本章小结¶
核心收获：
- Docker 化：
- 多阶段构建减小镜像体积
-
Docker Compose 本地部署
-
K8s 部署：
- Deployment/Service/Ingress
- StatefulSet 部署数据库
-
HPA 自动扩缩容
-
监控告警：
- Prometheus 采集指标
- Grafana 可视化
-
告警规则配置
-
性能优化：
- 数据库索引与分区
- 缓存策略
- 批量处理
- 压测调优
七、课程完结¶
SmartSupport 项目总结：
从 0 到 1 构建的 SaaS 智能客服平台
技术栈：
• C++17 / Boost.Asio / CMake
• PostgreSQL + pgvector
• Redis
• Docker / Kubernetes
• Prometheus / Grafana
核心能力：
• AI 自动回复（RAG + LLM）
• 人工客服接入
• 多租户隔离
• 计费与套餐
• 管理后台
生产就绪：
• 容器化部署
• 自动扩缩容
• 监控告警
• 性能优化
恭喜你完成了 NuClaw 全系列教程！
从 Step 0 的 Echo 服务器，到 Step 20 的生产级 SaaS 平台，你已经掌握了 AI Agent 开发的完整技能栈。
去构建属于你自己的产品吧！ 🚀