Nginx日志调优之关闭无用日志与日志级别调整策略

# 在 http 块中定义映射规则
map $request_uri $loggable {
    default 1;                    # 默认记录
    ~*^/health$ 0;                # 健康检查不记录
    ~*^/ready$ 0;                 # 就绪检查不记录
    ~*^/live$ 0;                  # 活跃检查不记录
    ~*^/metrics$ 0;               # prometheus 指标不记录
    ~*^/swagger-ui.html$ 0;       # swagger ui 不记录
    ~*^/v1/health$ 0;             # 微服务健康端点
}
# 在 server 块中应用条件日志
server {
    listen 80;
    server_name example.com;
    # 关键：只有 $loggable 为 1 时才写入 access log
    access_log /var/log/nginx/access.log combined if=$loggable;
    # 其他配置...
    location / {
        proxy_pass http://backend;
    }
    location /health {
        return 200 "ok";
    }
    location /metrics {
        stub_status on;
        access_log off;  # 也可单独关闭
    }
}

map $http_user_agent $log_user_agent {
    default 1;
    ~*googlebot 0;
    ~*baiduspider 0;
    ~*yandexbot 0;
    ~*semrushbot 0;
    ~*ahrefsbot 0;
    ~*mj12bot 0;
    ~*dotbot 0;
    ~*zoominfobot 0;
}
# 组合多个条件：必须同时满足 loggable=1 且 log_user_agent=1 才记录
map $loggable$log_user_agent $final_log {
    default 0;
    "11" 1;
}
server {
    access_log /var/log/nginx/access.log combined if=$final_log;
}

# 为 api 网关单独记录
server {
    listen 8080;
    server_name api.example.com;
    access_log /var/log/nginx/api-access.log combined if=$loggable;
    error_log /var/log/nginx/api-error.log warn;
    location / {
        proxy_pass http://api-service;
    }
}
# 为静态资源服务器关闭 access log
server {
    listen 8081;
    server_name static.example.com;
    access_log off;  # 完全关闭
    error_log /var/log/nginx/static-error.log error;
    location / {
        root /var/www/static;
        expires 1y;
    }
}

# 全局 error_log 配置（建议放在 nginx.conf 最上方）
error_log /var/log/nginx/error.log warn;
# 如果需要更细粒度控制，可为不同 server 设置不同级别
server {
    listen 80;
    server_name example.com;
    error_log /var/log/nginx/example-error.log warn;
    # 某个敏感服务，记录更详细（仅临时）
    # error_log /var/log/nginx/sensitive-error.log notice; # 临时调试用
}

// healthcheckcontroller.java
package com.example.controller;
import org.springframework.web.bind.annotation.getmapping;
import org.springframework.web.bind.annotation.restcontroller;
@restcontroller
public class healthcheckcontroller {
    @getmapping("/health")
    public string health() {
        return "up";
    }
    @getmapping("/ready")
    public string ready() {
        // 检查数据库、缓存、消息队列等依赖
        return "ready";
    }
    @getmapping("/live")
    public string live() {
        // 仅检查进程是否存活
        return "live";
    }
    @getmapping("/metrics")
    public string metrics() {
        // 返回 prometheus 格式指标
        return """
            # help http_requests_total total number of http requests
            # type http_requests_total counter
            http_requests_total{method="get",status="200"} 12345
            """;
    }
}

# nginx.conf - 全局配置
user nginx;
worker_processes auto;
error_log /var/log/nginx/error.log warn;  # ✅ 生产推荐级别
pid /run/nginx.pid;
events {
    worker_connections 1024;
}
http {
    # 定义是否记录日志的映射规则
    map $request_uri $loggable {
        default 1;
        ~*^/health$ 0;
        ~*^/ready$ 0;
        ~*^/live$ 0;
        ~*^/metrics$ 0;
        ~*^/swagger-ui.html$ 0;
        ~*^/v1/health$ 0;
    }
    # 定义是否记录爬虫
    map $http_user_agent $log_user_agent {
        default 1;
        ~*googlebot 0;
        ~*baiduspider 0;
        ~*yandexbot 0;
        ~*semrushbot 0;
        ~*ahrefsbot 0;
        ~*mj12bot 0;
        ~*dotbot 0;
        ~*zoominfobot 0;
    }
    # 组合条件：只有非健康且非爬虫才记录
    map $loggable$log_user_agent $final_log {
        default 0;
        "11" 1;
    }
    # 定义自定义日志格式（仅记录关键字段）
    log_format custom '$remote_addr - $remote_user [$time_local] '
                      '"$request" $status $body_bytes_sent '
                      '"$http_referer" "$http_user_agent" '
                      '$upstream_response_time $request_time';
    access_log /var/log/nginx/access.log custom if=$final_log;
    # 开启压缩，减少网络传输（间接降低日志体积）
    gzip on;
    gzip_types text/plain application/json application/javascript text/css;
    include /etc/nginx/conf.d/*.conf;
}

# /etc/nginx/conf.d/app.conf
server {
    listen 80;
    server_name api.example.com;
    # 关闭静态资源日志
    location ~* \.(jpg|jpeg|png|gif|ico|css|js|woff2)$ {
        root /var/www/static;
        expires 1y;
        access_log off;  # ✅ 关闭
    }
    # 代理到 java 应用
    location / {
        proxy_pass http://127.0.0.1:8080;
        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;
        # 优化：关闭不必要的代理日志
        proxy_cache off;
        proxy_buffering off;
    }
    # 明确关闭这些路径的访问日志
    location /health {
        return 200 "up";
        access_log off;
    }
    location /ready {
        return 200 "ready";
        access_log off;
    }
    location /live {
        return 200 "live";
        access_log off;
    }
    location /metrics {
        stub_status on;
        access_log off;
    }
}

# 精简版：只记录核心指标
log_format concise '$remote_addr - $remote_user [$time_local] '
                   '"$request_method $request_uri $server_protocol" '
                   '$status $body_bytes_sent '
                   '$request_time $upstream_response_time';
access_log /var/log/nginx/access.log concise if=$final_log;

access_log /var/log/nginx/access.log concise buffer=32k flush=5s if=$final_log;

# /etc/logrotate.d/nginx
/var/log/nginx/*.log {
    daily
    missingok
    rotate 14
    compress
    delaycompress
    notifempty
    create 0640 nginx adm
    sharedscripts
    postrotate
        [ -f /var/run/nginx.pid ] && kill -usr1 `cat /var/run/nginx.pid`
    endscript
}

sudo logrotate -d /etc/logrotate.d/nginx  # 模拟运行
sudo logrotate -f /etc/logrotate.d/nginx  # 强制执行

{
  "title": "nginx 日志调优监控",
  "panels": [
    {
      "title": "每日访问日志体积",
      "type": "graph",
      "query": "sum(increase(nginx_access_log_bytes_total[1d]))",
      "alert": "如果 > 50gb，触发磁盘告警"
    },
    {
      "title": "错误日志条目数（每分钟）",
      "type": "stat",
      "query": "sum(rate(nginx_error_log_lines[1m]))",
      "alert": "如果 > 100/min，检查后端服务"
    },
    {
      "title": "日志过滤率",
      "type": "gauge",
      "query": "1 - (sum(nginx_access_log_records_filtered) / sum(nginx_access_log_records_total))",
      "alert": "过滤率 < 70%，说明配置失效"
    }
  ]
}