nginx优化静态内容

server

nginx 基本知识

location 匹配url
  • 优先级:
    • [精确匹配(同等类型顺序越前,优先级越高)]
    • [开头匹配(同等类型长度越长,优先级越高)]
    • [正则匹配(同等类型顺序越前,优先级越高)]
    • [通用匹配(同等类型长度越长,优先级越高)]
  • 语法规则：location [=|^~|~|~*] /uri/ {…} 先匹配普通location，在匹配正则location
类型 操作符 含意
精确匹配 = = /hello :完全相等
开头匹配 ^~ ^~ /hello/yesno :
uri以某个常规字符串开头,理解为匹配url路径即可,无需考虑编解码
正则匹配 ~ ~ /hello/y[a-e][e-z] :
开头表示区分大小写的正则匹配
正则匹配 ~* location ~* /hello/y[a-e][a-z][1-9] :
开头表示不区分大小写的正则匹配
正则匹配 !~ 开头表示区分大小写的不匹配的正则匹配
正则匹配 !~* 开头表示不区分大小写的不匹配的正则匹配
通用匹配 / /hello/no :
任何请求都会被匹配到

设置静态资源的过期时间

  • 服务器静态资源是否过期 存在两个判断机制的：

    • 新鲜度检测阶段:需要依赖响应头的Cache-Control/Expires，通过才会返回200(from cache)
    • 资源二次校验阶段:校验资源一致性,Last-Modified/ETag 这两个数据用于确定数据是否改变.如果一致返回304,如果不一致返回200,当然，如果没有找到返回404

  • expires指令用来对浏览器本地缓存的控制,指令可以放置在http {}，server {}，location {}或位置{}块内的if语句中

    • 当没有设置expires指令时:
curl -I http://0.0.0.0/logo.png
HTTP/1.1 200 OK
Server: nginx/1.10.3 (Ubuntu)
Date: Fri, 22 Jun 2018 03:16:20 GMT
Content-Type: image/png
Content-Length: 18
Connection: keep-alive
Pragma: public
Cache-Control: public
    • 当设置expires指令时:
curl -I http://0.0.0.0/logo.png
HTTP/1.1 200 OK
Server: nginx/1.10.3 (Ubuntu)
Date: Fri, 22 Jun 2018 03:16:20 GMT
Content-Type: image/png
Content-Length: 18
Connection: keep-alive
Pragma: public
Cache-Control: public
Expires: Sun, 22 Jul 2018 03:16:20 GMT
Cache-Control: max-age=2592000

缓存服务器设置(配置proxy_cahe):只是针对API后端,所以才需要 proxy_pass或者 uwsgi_pass配置才生效

  • 参数说明:
指令 指令与参数值 含意
proxy_cache_path /tmp/nginx cache这个zone的文件要存放的目录,会在已经的路径下生成一个目录
proxy_cache_path levels=1:2 表示缓存目录的第一级目录是1个字符,第二级目录是2个字符.
即/tmp/nginx/a/1b这种形式
proxy_cache_path keys_zone=cache:10m zone名称为cache1,分配的内存大小为10MB
proxy_cache_path inactive=1d 这个zone中的缓存文件如果在1天内都没有被访问，那么文件会被cache manager进程删除掉
proxy_cache_valid 200 304 30m 设置状态码为200和304的响应可以进行缓存,并且缓存时间为30分钟
upstream_cache_status MISS 未命中,请求被传送到后端
upstream_cache_status HIT 缓存命中
upstream_cache_status EXPIRED 缓存已经过期请求被传送到后端
upstream_cache_status UPDATING 正在更新缓存，将使用旧的应答
upstream_cache_status STALE 后端将得到过期的应答
upstream_cache_status BYPASS 缓存被绕过了

  • 遇到的坑(也一直没有效果)

    • proxy_pass http://0.0.0.0/; # uwsgi启动的话,则将对应的proxy设置换成uwsgi即可

  • 查看是否生效

    curl -I http://0.0.0.0/logo.png
X-Cache: MISS
X-Cache: HIT  # 第二次请求,如果是HIT,则表示缓存生效

最终配置如下:

proxy_cache_path /tmp/nginx levels=1:2 keys_zone=cache:10m inactive=60m;

map $sent_http_content_type $expires {
    default                    off;
    text/html                  epoch;
    text/css                   max;
    application/javascript     max;
    image/png                  2d;
    ~*\.(jpg|jpeg|png|gif|ico|css|js|pdf)(\?|$) 7d;

}

server {
    listen       80;
    server_name 0.0.0.0 localhost ;
    location = /hello {
        return 302  https://www.baidu.com;    
    }

    # expires 1d;

    # location ~* .*\.(ico|css|js|gif|jpe?g|png|woff|woff2|eot|svg|ttf|)$ {
        # expires 1d;
    # }

    # location ~* \.(jpg|jpeg|png|gif){
        # expires 1d;
    # }

    # location = /logo.png {
        # root /usr/share/nginx/html;
        # expires 30d;
        # access_log off;
        # log_not_found off;
        # add_header Pragma public;
        # add_header Cache-Control "public"; # 任何系统都可以缓存它们
    # }

    # expires $expires; # 这种方式在chrome,firefox没有Expires,在命令行中就有
    location / {

        root /usr/share/nginx/html;
        index index.html;
        if ($request_uri ~* "\.(ico|css|js|gif|jpe?g|png)$") {
            expires 30d;
            access_log off;
            add_header Cache-Control "public";
        }


        proxy_cache cache;
        proxy_cache_valid 200 301 304 30m;
        proxy_pass    http://0.0.0.0:8000;

        # 缓存 key 的值
        proxy_cache_key $host$uri$is_args$args;
        # proxy_cache_key "$scheme$request_method$host$request_uri";
        proxy_cache_use_stale updating error timeout invalid_header http_500 http_502 http_503 http_504;

        expires 1m;
        # 将缓存的状态添加到 Header 中
        add_header X-Cache '$upstream_cache_status';
    }
}

缓存服务器设置发生在:API后端和Nginx静态文件分离情况的配置文件(适用后端接口不是实时的改变的)

uwsgi_cache_path /tmp/nginx_cache/ levels=1:2 keys_zone=mycache:10m max_size=10g inactive=10m use_temp_path=off;

# the upstream component nginx needs to connect to
upstream uwsgi {
    server unix:/code/app.sock; # for a file socket
}

# configuration of the server
server {
    # the port your site will be served on
    listen    80;
    server_name  0.0.0.0;
    charset     utf-8;
    # add_header Accept-Ranges bytes;
    # add_header Content-Encoding identity;
    # add_header Accept-Encoding identity;
    # add_header Range bytes=0-;

    client_max_body_size 75M;   # adjust to taste


    # Django media
    location /media  {
        alias /code/media;  # your Django project's media files - amend as required
        if ($request_uri ~* "\.(ico|gif|jpe?g|png|woff|woff2|eot|svg|ttf|mp4)$") {
            expires 3d;
            access_log off;
            add_header Cache-Control "public";

        }

    }

    location /static {
        alias /code/static; # your Django project's static files - amend as required

        if ($request_uri ~* "\.(css|js)$") {
            expires 1d;
            access_log off;
            add_header Cache-Control "public";
        }
    }

    location /favicon.ico {
        alias /code/media/logo.png; # your Django project's static files - amend as required
    }

    # / 必须放在最后,不能覆盖静态url配置
    location / {

        add_header      X-uWSGI-Cache $upstream_cache_status;

        # server cache
        uwsgi_cache  mycache;
        uwsgi_cache_valid  200 301 1m;
        slice 1m;
        proxy_set_header Range $slice_range;
        proxy_set_header Range $http_range;
        # proxy_cache_valid 200 301 304 30m;

        uwsgi_cache_key  $request_uri;

        # client cache
        expires 3d;

        uwsgi_pass  uwsgi;
        include     /etc/nginx/uwsgi_params; # the uwsgi_params file you installed
    }
}
Share Comments