前置文章:
linux(centos7) 下 nginx1.15.8 安装步骤
我们访问一个网址,服务器返回对应的资源。那么一个网址是如何对应一个资源的呢?
用 nginx 可以很好地帮我们实现路由功能,我们所有需要做的就是配置好 location 模块。
语法规则
location [=|~|~*|^~] /uri/ {… }
符号 | 含义 |
= | 精确匹配 |
^~ | 非正则匹配 |
~ | 正则匹配(区分大小写) |
~* | 正则匹配(不区分大小写) |
!~ | 正则不匹配(区分大小写) |
!~* | 正则不匹配(不区分大小写) |
普通匹配(这里没有符号的时候) |
匹配规则
1. 精准匹配命中时,停止location
2.一般匹配(普通和非正则)命中时,对比所有命中的一般匹配,选出最长的一条
3.如果最长的那一条为非正则匹配,直接匹配此条,停止location
4.如果最长的那一条为普通匹配,继续尝试正则location(以上至此都不存在代码顺序)
5.按代码顺序执行正则匹配,当第一条正则location命中时,停止location
示例:
想运行以下示例需先下载第三方模块echo-nginx-module:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
|
#下载到/usr/local/src目录 wget https: //github .com /openresty/echo-nginx-module/archive/v0 .61. tar .gz tar -zxvf v0.61. tar .gz #在nginx源码目录nginx-1.15.8下配置,--add-module指向模块目录即会安装插件到nginx中 . /configure --add-module= /usr/local/src/echo-nginx-module-0 .61/ make && make install #验证安装 nginx - v |
你还需要配置 c:\windows\system32\drivers\etc\hosts 文件,添加“虚拟机ip 域名”到最后:
192.168.100.14 test.loaction.com
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
|
server { listen 80; server_name test .location.com; #精准匹配测试 #第1,2条虽然匹配,但第三条是精准匹配,出第三条结果 #测试路径/equal/a/b/c location ~ /equal/ * { echo '/equal/*' ; } location /equal/a/b { echo '/equal/a/b' ; } location = /equal/a/b/c { echo '/equal/a/b/c' ; } #普通匹配测试 #第1,2条虽然匹配,第三条匹配更长,出第三条结果 #测试路径/match/a/b/c location /match/a { return 200 "/match/a" ; } location /match/a/b { return 200 "/match/a/b" ; } location /match/a/b/c { return 200 "/match/a/b/c" ; } location /match/a/b/c/d { return 200 "/match/a/b/c/d" ; } #正则匹配覆盖普通匹配,不会覆盖非正则匹配 #访问/re/a.htm,会被后面的正则覆盖 #访问/re/a/b开头的路径,不会被后面的正则覆盖 location /re/a .htm { echo 'match /re/a.htm' ; } location ^~ /re/a/b { echo 'math ^~/re/a/b*' ; } location ~ /re/ (.*)\.(htm|js|css)$ { echo "cover /re/$1.$2" ; } #正则匹配成功一条后,便不再走其它正则 #测试路径/rex/a/b/c.htm location ~ /rex/ .*\.(htm|js|css)$ { echo "match first" ; } location ~ /rex/a/ (.*)\.(htm|js|css)$ { echo "match second" ; } location ~ /rex/a/b/ (.*)\.(htm|js|css)$ { echo "match third" ; } } |
结果:
精准匹配最优先:
一般匹配选最长:
正则覆盖普通匹:
不能覆盖非正则
正则匹配选首位:
proxy_pass 代理转发
在上面的示例中,我们通过修改本机hosts文件,让本机访问 test.loaction.com 相当于访问 192.168.100.14。我们现在假设有一个服务在 192.168.100.15 这台机器上。我们能不能通过访问 test.loaction.com,访问到 192.168.100.15 上的资源呢?
这样的想法其实是在模拟实际做项目的时候,我们可以只购买一个域名,然后访问多个云服务器。我们要做的就是让域名对应的服务器具有代理转发的功能。这里是服务器作为代理,让客户端能通过访问代理服务器来访问其它服务器,所以是反向代理。
通过在 location 里配置 proxy_pass 就能实现代理转发功能,帮助我们将请求转发到别的服务器。
我们先看示例再看规则:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
|
server { listen 80; server_name test .location.com; 本机ip地址:192.168.100.104 #后端服务路径: http: //192 .168.100.105:8080 /user/query ? id =1 #规则一: #访问路径:http://test.location.com/user/query?id=1 location /user { #path1:/user path2:/query #ip:port 后面无 / proxy_pass http: //192 .168.0.105:8080; } 规则二: #访问路径 :http://test.location.com/a/user/query?id=1 location /a/user { #path1:/a/user path2:/query #ip:port 后面有 /xxx proxy_pass http: //192 .168.0.105:8080 /user ; } } |
解读:
后端服务路径都是 http://192.168.100.105:8080/user/query?id=1。用规则一,可以使访问路径为 http://test.location.com/user/query?id=1;用规则二,可以使访问路径为 http://test.location.com/a/user/query?id=1。
规则:
1. 访问路径格式为 sever_name + path1 + path2 + ?param
2. location 后面的路径就是 path1,对照访问路径格式,path1 和 ?param 之间的为path2
3. 如果 proxy_pass 后面的值形如 http: // ip : port; 那么实际访问的地址是 http: // ip : port + path1 + path2 + ?param
4. 如果 proxy_pass 后面的值形如 http: // ip : port / xxx; (末尾相比上面有 /xxx)
那么实际访问的地址是 http: // ip : port + path2 + ?param
规则二的作用:
如果我们访问其它服务器都使用规则一,那么我们将无法直观地从 url 确定这个服务属于那台机器或者哪种服务。
而如果使用规则二,我们可以通过合理的设置明确地知道,我们访问的是服务a。
root 与 index
nginx.conf 里的默认 location 如下:
1
2
3
4
|
location / { root html; index index.html; } |
意思是资源文件的根目录在 nginx 程序所在文件夹下的 html 文件夹:
如果我们没有写 uri,默认的资源是 html 文件夹下的 index.html
但是你如果这样配置:
1
2
3
4
5
6
7
8
|
location / { root html; index index.html; } location /index .html { echo "index" ; } |
最后访问 http://test.location.com/ 浏览器会输出 index:
也就是说,虽然前面设置了 index ,但是如果它对应的地址有被 location 匹配到的话,会执行 location 匹配后的结果。
但是,必须还是得有 index 对应的文件。
为了验证上面这一点,我把 html 目录下的 index.html 改成 index,再 reload 并访问:
root 与 alias
还是先看示例比较直观:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
|
server { listen 80; server_name test . local .com; #文件地址:/etc/nginx/html/static/a.html #访问路径:http://test.local.com/static/a.html location /static { #path1:/static path2:/a.html root /etc/nginx/html/ ; #文件地址:root + path1 + path2 } #访问路径:http://test.local.com/target/a.html location /target { #path1:/target path2:/a.html alias /etc/nginx/html/static/ ; #文件地址:alias + path2 } } |
解读:
文件地址一样,访问路径可以不同。
使用 root 声明的是根目录,经过 location 匹配后,直接去根目录下找访问路径中 sever_name 后面的文件地址。
使用 alias 声明的是别名目录,经过 location 匹配后,去别名目录下找访问路径中 sever_name 后面除去 location 匹配的地址的剩余地址。这就好像给 location 匹配的地址取了个别名一样。
规则:
1. 访问路径格式为 sever_name + path1 + path2
2. location 后面的路径就是 path1,对照访问路径格式,path1 后面的就是 path2
3. 如果使用 root ,文件地址为:
root + path1 +path2
4. 如果使用 alias ,文件地址为:
alias + path2
location 执行过程
结合以上内容,就很容易理解 location 执行过程了,如下图
rewrite 重定向
location 里还可以配置 rewrite 实现重定向:
rewrite regex replacement [flag];
其中:
regex:正则表达式
replacement :替换值flag:后续处理标识,可以为 break/last/permanent/redirect
重点在于 flag :
1. flag=break
发生 nginx 内部重定向,path值被更新,rewrite层面的命令会中断。原控制流程逻辑不变往下走
2. flag=last
发生nginx内部重定向,path值被更新,rewrite层面的命令会中断。控制流程刷新,重新进行整个location层的逻辑流程
3. flag= permanent/redirect
发生页面重定向(301永久重定向/302临时重定向),nginx流程结束,返回http响应到浏览器,页面url更新
4.flag为空
发生nginx内部重定向,path值被更新,rewrite层面的命令继续。最后一个rewrite完毕,刷新控制流程,重新进行location重匹配
示例:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
|
server { listen 80; server_name test .location.com; location /a .html { echo 'i am a.html' ; } location /b .html { echo 'i am b.html' ; } #此路径请求:http://test.location.com/aa.html location /aa .html { ##内部重定向 rewrite ^/ /a .html break ; ##不会执行下面的rewrite rewrite ^/ /b .html break ; root /etc/nginx/html/ ; } #此路径请求:http://test.location.com/ab.html location /ab .html { ##内部重定向 rewrite ^/ /a .html last; ##不会执行下面的rewrite,但重新location匹配 rewrite ^/ /b .html last; rewrite ^/ /c .html; root /etc/nginx/html/ ; } #此路径请求:http://test.location.com/ba location /ba { rewrite ^/ /b .html permanent; ##301永久重定向 root /etc/nginx/html/ ; } #此路径请求:http://test.location.com/bb location /bb { rewrite ^/ /b .html redirect; ##302临时重定向 set $aa 12; root /etc/nginx/html/ ; } #此路径请求:http://test.location.com/cc.html location /cc .html { rewrite ^/ /c .html; ##指令不停,继续往下 rewrite ^/ /b .html; rewrite ^/ /a .html; ##最后一条,生效的是这条 root /etc/nginx/html/ ; } } |
结果:
访问 http://test.location.com/aa.html,直接寻找 root 目录下的 a.html,我没有写这个 html,所以是 404
访问 http://test.location.com/ab.html,因为重新 location 匹配了,所以匹配到了上面的 /a.html,输出了一句话
访问 http://test.location.com/ba,永久重定向,可以看到地址栏的网址都变了,是外部重定向
访问 http://test.location.com/bb,临时重定向
访问 http://test.location.com/cc.html,不是 404,也经过了重新 location 匹配
nginx 处理请求的11个阶段
nginx 处理请求的全过程一共划分为 11 个阶段(如图),按阶段由上到下依次执行 (上一阶段的所有指令执行完毕,才进入下一阶段)
各阶段的含义如下:
- post_read: 接收到完整的 http 头部后处理的阶段,在uri重写之前。一般跳过
- server_rewrite: location匹配前,修改uri的阶段,用于重定向,location块外的重写指令(多次执行)
- find_config: uri 寻找匹配的location块配置项(多次执行)
- rewrite:找到location块后再修改uri,location级别的uri重写阶段(多次执行)
- post_rewrite:防死循环,跳转到对应阶段
- preaccess: 权限预处理
- access:判断是否允许这个请求进入
- post_access: 向用户发送拒绝服务的错误码,用来响应上一阶段的拒绝
- try_files: 访问静态文件资源
- content : 内容生成阶段,该阶段产生响应,并发送到客户端
- log:记录访问日志
到此这篇关于nginx配置 location模块实现路由(反向代理、重定向)功能的文章就介绍到这了,更多相关nginx location 路由反向代理和重定向内容请搜索服务器之家以前的文章或继续浏览下面的相关文章希望大家以后多多支持服务器之家!
原文链接:https://blog.csdn.net/weixin_44367006/article/details/101715799