Introduction to how to configure fastcgi cache in Nginx

This article brings you an introduction to the method of configuring fastcgi cache in Nginx. It has certain reference value. Friends in need can refer to it. I hope it will be helpful to you.

Caching mechanism

Nginx provides two caching methods based on proxy (proxy) and fastcgi-based. The instruction sets of the two are very similar, but there are differences in use. Not big. Since this blog adopts Nginx PHP-FPM architecture, it uses fastcgi-based cache, that is, fastcgi cache.

Configuration

The mechanism of fastcgi cache is to cache the page returned by the back-end CGI service. When subsequent requests come, the page will be returned directly, eliminating the need to communicate with the back-end service. And the consumption of generating pages. The principle of fastcgi cache is similar to that of squid, varnish, CDN, etc. Due to the reduction of back-end requests, in principle the performance will be much higher than various WordPress cache plug-ins.

fastcgi cache belongs to the fastcgi module, and the fastcgi module belongs to the http module, so cache-related configurations should appear in these two and subordinate configuration blocks (mainly http, server and location).

According to the official documentation, the cache configuration used by the blog site is as follows:

http {
  ...
  fastcgi_connect_timeout 30s;
  # 全局设置， fastcgi_cache_path只能出现在http配置块中
  # fastcgi_cache_path设置全局的缓存路径、文件大小等，可多次出现设置不同的缓存池
  fastcgi_cache_path /var/cache/nginx levels=1:2 keys_zone=WORDPRESS:10m inactive=10m max_size=500m;
  fastcgi_cache_key $scheme$request_method$host$request_uri;
  # 设置使用过期缓存的情形：后端错误、超时等
  fastcgi_cache_use_stale error timeout invalid_header http_500 http_503;
  ...
  server {
    ...
    # 缓存策略指示变量
    set $skip_cache 0;
    # 缓存策略
    if ($query_string != "") {
      set $skip_cache 1;
    }
    if ($request_uri ~* "/wp-admin/|/xmlrpc.php|wp-.*.php|/feed/|sitemap(_index)?.xml") {
      set $skip_cache 1;
    }
    # 登录用户或发表评论者
    if ($http_cookie ~* "comment_author|wordpress_[a-f0-9]+|wp-postpass|wordpress_no_cache|wordpress_logged_in") {
      set $skip_cache 1;
    }
    location ~ \.php$ {
      ...
      # 使用的keyzone，这里使用http中定义的WORDPRESS
      fastcgi_cache WORDPRESS;
      # 是否跳过缓存
      fastcgi_cache_bypass $skip_cache;
      fastcgi_no_cache $skip_cache;
      # 后端返回成功的情况下缓存10分钟
      fastcgi_cache_valid 200 10m;
      # 添加header字段，指示缓存命中状态（调试使用）
      add_header X-Cache $upstream_cache_status;
    }
  }
}

The above configuration does the following:

1. Set the cache pool information (fastcgi_cache_path );

2. Set cache key (fastcgi_cache_key);

3. Expired file processing (fastcgi_cache_use_stale)

4. Cache strategy (fastcgi_cache_bypass/fastcgi_no_cache)

5. Cache time when the backend request is successful (fastcgi_cache_valid)

6. Added debugging information header (add_header)

Restart Nginx, and then visit the page to check the header status and response time of the page. It can be seen that after one request (the number of times can be set through fastcgi_cache_min_uses), subsequent requests are made from the cache (X-Cache value is HIT), and the speed improvement is very obvious. For the management background page, the value of x-cache is BYPASS, which is exactly what we need.

Recommended related articles:

Operation steps of Nginx configuration file nginx.conf

Explanation of the Rewrite pseudo-static rule in Nginx ( Details)

The above is the detailed content of Introduction to how to configure fastcgi cache in Nginx. For more information, please follow other related articles on the PHP Chinese website!