varnish 示例详解

最新推荐文章于 2020-08-28 00:39:56 发布
最新推荐文章于 2020-08-28 00:39:56 发布
阅读量330 收藏
点赞数
分类专栏： Linux学习笔记 文章标签： Linux varnish
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/weixin_44983653/article/details/102471668
版权

varnish 示例详解

1、缓存命中次数示例

obj.hits 是内建变量，用于保存某缓存项的从缓存中命中的次数。

基本语法如下：

sub vcl_deliver {
	if (obj.hits>0) {                                  
		set resp.http.X-Cache = "HIT via " + server.ip;
	} else {
		set resp.http.X-Cache = "MISS via " + server.ip;
	}
}

1.1 绘制拓扑

在这里插入图片描述

1.2 varnish 服务器配置

[root@Neo_Neo varnish]# vim varnish.params 
RELOAD_VCL=1

VARNISH_VCL_CONF=/etc/varnish/default.vcl

VARNISH_LISTEN_PORT=80

VARNISH_ADMIN_LISTEN_ADDRESS=127.0.0.1
VARNISH_ADMIN_LISTEN_PORT=6082

VARNISH_STORAGE="file,/data/varnish/cache,1G"

VARNISH_USER=varnish
VARNISH_GROUP=varnish

[root@Neo_Neo varnish]# ll /data/varnish/cache/ -d
drwxr-xr-x. 2 varnish varnish 6 Oct  9 10:22 /data/varnish/cache/

[root@Neo_Neo varnish]# vim default.vcl 
vcl 4.0;

backend default {
    .host = "192.168.10.12";
    .port = "80";
}

sub vcl_deliver {
    if (obj.hits>0) {
         set resp.http.X-Cache = "HIT via " + server.ip;
    } else {
         set resp.http.X-Cache = "MISS via " + server.ip;
    }
}

1.3 varnish 配置重载

[root@Neo_Neo varnish]# varnishadm -S secret -T 127.0.0.1:6082
200        
-----------------------------
Varnish Cache CLI 1.0
-----------------------------
Linux,3.10.0-957.el7.x86_64,x86_64,-sfile,-smalloc,-hcritbit
varnish-4.0.5 revision 07eff4c29

Type 'help' for command list.
Type 'quit' to close CLI session.

varnish> vcl.list                    # 使用 varnishadm 工具进行配置重载
200        
active          0 boot

varnish> vcl.load tang1 default.vcl 
200        
VCL compiled.

varnish> vcl.list
200        
active          0 boot
available       0 tang1

varnish> vcl.use tang1
200        
VCL 'tang1' now active

varnish> vcl.list
200        
available       0 boot
active          0 tang1

1.4 用户进行访问

1.4.1 varnish 服务器对后端服务器进行访问

[root@Neo_Neo varnish]# alias ipinfo='ifconfig | grep "inet 192" -B 1'
[root@Neo_Neo varnish]# ipinfo
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.1.11  netmask 255.255.255.0  broadcast 192.168.1.255
--
ens37: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.10.11  netmask 255.255.255.0  broadcast 192.168.10.255
[root@Neo_Neo varnish]# curl http://192.168.10.12
<h1>HTTP backend-web-server-1 192.168.10.12</h1>

1.4.2 用户对后端服务器进行访问

[root@Neo ~]# alias ipinfo='ifconfig | grep "inet 192" -B 1'
[root@Neo ~]# ipinfo
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.1.10  netmask 255.255.255.0  broadcast 192.168.1.255

### 第一次进行访问时，不会命中缓存 ###
[root@Neo ~]# curl -I http://192.168.1.11
HTTP/1.1 200 OK
Date: Wed, 09 Oct 2019 14:26:36 GMT
Server: Apache/2.4.6 (CentOS) OpenSSL/1.0.2k-fips
Last-Modified: Wed, 09 Oct 2019 14:19:45 GMT
ETag: "31-5947afa0a0130"
Content-Length: 49
Content-Type: text/html; charset=UTF-8
X-Varnish: 2
Age: 0
Via: 1.1 varnish-v4
X-Cache: MISS via 192.168.1.11
Connection: keep-alive

### 第二次访问时，命中缓存成功 ###
[root@Neo ~]# curl -I http://192.168.1.11
HTTP/1.1 200 OK
Date: Wed, 09 Oct 2019 14:26:36 GMT
Server: Apache/2.4.6 (CentOS) OpenSSL/1.0.2k-fips
Last-Modified: Wed, 09 Oct 2019 14:19:45 GMT
ETag: "31-5947afa0a0130"
Content-Length: 49
Content-Type: text/html; charset=UTF-8
X-Varnish: 32770 3
Age: 23
Via: 1.1 varnish-v4
X-Cache: HIT via 192.168.1.11
Connection: keep-alive

### 查看访问内容 ###
[root@Neo ~]# curl http://192.168.1.11
<h1>HTTP backend-web-server-1 192.168.10.12</h1>

2、强制对某类资源的请求不检查缓存

基本语法格式：

### (?i) 代表 不区分大小写 ###
sub vcl_recv {
	if (req.url ~ "(?i)^/(login|admin)") {         
		return(pass);
	}
}

2.1 绘制拓扑

在这里插入图片描述

2.2 varnish 服务器配置

[root@Neo_Neo varnish]# vim varnish.params 
RELOAD_VCL=1

VARNISH_VCL_CONF=/etc/varnish/default.vcl

VARNISH_LISTEN_PORT=80

VARNISH_ADMIN_LISTEN_ADDRESS=127.0.0.1
VARNISH_ADMIN_LISTEN_PORT=6082

VARNISH_STORAGE="file,/data/varnish/cache,1G"

VARNISH_USER=varnish
VARNISH_GROUP=varnish

[root@Neo_Neo varnish]# ll /data/varnish/cache/ -d
drwxr-xr-x. 2 varnish varnish 6 Oct  9 10:22 /data/varnish/cache/

[root@Neo_Neo varnish]# vim default.vcl 
vcl 4.0;

backend default {
    .host = "192.168.10.12";
    .port = "80";
}

sub vcl_recv {
    if (req.url ~ "(?i)^/(login|admin)") {
        return(pass);
    }           
}

sub vcl_deliver {
    if (obj.hits>0) {
         set resp.http.X-Cache = "HIT via " + server.ip;
    } else {
         set resp.http.X-Cache = "MISS via " + server.ip;
    }
}

2.3 按 1.3 进行重载

2.4 用户进行访问

2.4.1 varnish 服务器对后端服务器进行访问

[root@Neo_Neo ~]# ipinfo
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.1.11  netmask 255.255.255.0  broadcast 192.168.1.255
--
ens37: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.10.11  netmask 255.255.255.0  broadcast 192.168.10.255
[root@Neo_Neo ~]# curl http://192.168.10.12/login/
<h1>HTTP backend-web-server-1 192.168.10.12(login)</h1>
[root@Neo_Neo ~]# curl http://192.168.10.12/admin/
<h1>HTTP backend-web-server-1 192.168.10.12(admin)</h1>

2.4.2 用户对后端服务器（login 目录）进行访问

[root@Neo ~]# alias ipinfo='ifconfig | grep "inet 192" -B 1'
[root@Neo ~]# ipinfo
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.1.10  netmask 255.255.255.0  broadcast 192.168.1.255

### 第一次进行访问时，不会命中缓存 ###
[root@Neo ~]# curl -I http://192.168.1.11/login/
HTTP/1.1 200 OK
Date: Wed, 09 Oct 2019 15:11:12 GMT
Server: Apache/2.4.6 (CentOS) OpenSSL/1.0.2k-fips
Last-Modified: Wed, 09 Oct 2019 15:03:04 GMT
ETag: "38-5947b94fa215b"
Accept-Ranges: bytes
Content-Length: 56
Content-Type: text/html; charset=UTF-8
X-Varnish: 26
Age: 0
Via: 1.1 varnish-v4
X-Cache: MISS via 192.168.1.11
Connection: keep-alive

### 第二次访问时，依旧无法命中缓存 ###
[root@Neo ~]# curl -I http://192.168.1.11/login/
HTTP/1.1 200 OK
Date: Wed, 09 Oct 2019 15:11:39 GMT
Server: Apache/2.4.6 (CentOS) OpenSSL/1.0.2k-fips
Last-Modified: Wed, 09 Oct 2019 15:03:04 GMT
ETag: "38-5947b94fa215b"
Accept-Ranges: bytes
Content-Length: 56
Content-Type: text/html; charset=UTF-8
X-Varnish: 32796
Age: 0
Via: 1.1 varnish-v4
X-Cache: MISS via 192.168.1.11
Connection: keep-alive

### 查看访问内容 ###
[root@Neo ~]# curl http://192.168.1.11/login/
<h1>HTTP backend-web-server-1 192.168.10.12(login)</h1>

2.4.3 用户对后端服务器（admin 目录）进行访问

[root@Neo ~]# alias ipinfo='ifconfig | grep "inet 192" -B 1'
[root@Neo ~]# ipinfo
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.1.10  netmask 255.255.255.0  broadcast 192.168.1.255

### 第一次进行访问时，不会命中缓存 ###
[root@Neo ~]# curl -I http://192.168.1.11/admin/
HTTP/1.1 200 OK
Date: Wed, 09 Oct 2019 15:10:23 GMT
Server: Apache/2.4.6 (CentOS) OpenSSL/1.0.2k-fips
Last-Modified: Wed, 09 Oct 2019 15:03:47 GMT
ETag: "38-5947b97895005"
Accept-Ranges: bytes
Content-Length: 56
Content-Type: text/html; charset=UTF-8
X-Varnish: 32784
Age: 0
Via: 1.1 varnish-v4
X-Cache: MISS via 192.168.1.11
Connection: keep-alive

### 第二次访问时，依旧无法命中缓存 ###
[root@Neo ~]# curl -I http://192.168.1.11/admin/
HTTP/1.1 200 OK
Date: Wed, 09 Oct 2019 15:10:24 GMT
Server: Apache/2.4.6 (CentOS) OpenSSL/1.0.2k-fips
Last-Modified: Wed, 09 Oct 2019 15:03:47 GMT
ETag: "38-5947b97895005"
Accept-Ranges: bytes
Content-Length: 56
Content-Type: text/html; charset=UTF-8
X-Varnish: 20
Age: 0
Via: 1.1 varnish-v4
X-Cache: MISS via 192.168.1.11
Connection: keep-alive

### 查看访问内容 ###
[root@Neo ~]# curl http://192.168.1.11/admin/
<h1>HTTP backend-web-server-1 192.168.10.12(admin)</h1>

3、禁止使用 curl 命令进行访问

当然也可以禁止其它方式进行访问。
基本语法格式：

### (?i) 代表 不区分大小写 ###
sub vcl_recv {
	if (req.http.User-Agent ~ "(?i)curl") {
		return(synth(403));
	}
}

3.1 绘制拓扑

在这里插入图片描述

3.2 varnish 服务器配置

[root@Neo_Neo varnish]# vim varnish.params 
RELOAD_VCL=1

VARNISH_VCL_CONF=/etc/varnish/default.vcl

VARNISH_LISTEN_PORT=80

VARNISH_ADMIN_LISTEN_ADDRESS=127.0.0.1
VARNISH_ADMIN_LISTEN_PORT=6082

VARNISH_STORAGE="file,/data/varnish/cache,1G"

VARNISH_USER=varnish
VARNISH_GROUP=varnish

[root@Neo_Neo varnish]# ll /data/varnish/cache/ -d
drwxr-xr-x. 2 varnish varnish 6 Oct  9 10:22 /data/varnish/cache/

[root@Neo_Neo varnish]# vim default.vcl 
vcl 4.0;

backend default {
    .host = "192.168.10.12";
    .port = "80";
}

sub vcl_recv {

	if (req.http.User-Agent ~ "(?i)curl") {
		return(synth(403));
	}

    if (req.url ~ "(?i)^/(login|admin)") {
        return(pass);
    }           
}

sub vcl_deliver {
    if (obj.hits>0) {
         set resp.http.X-Cache = "HIT via " + server.ip;
    } else {
         set resp.http.X-Cache = "MISS via " + server.ip;
    }
}

3.3 按 1.3 进行重载

3.4 用户进行访问

3.4.1 varnish 服务器对后端服务器进行访问

[root@Neo_Neo ~]# ipinfo
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.1.11  netmask 255.255.255.0  broadcast 192.168.1.255
--
ens37: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.10.11  netmask 255.255.255.0  broadcast 192.168.10.255
[root@Neo_Neo ~]# curl http://192.168.10.12
<h1>HTTP backend-web-server-1 192.168.10.12</h1>

3.4.2 使用 curl 进行访问（禁止访问）

[root@Neo ~]# alias ipinfo='ifconfig | grep "inet 192" -B 1'
[root@Neo ~]# ipinfo
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.1.10  netmask 255.255.255.0  broadcast 192.168.1.255

### 访问时，被禁止 ###
[root@Neo ~]# curl http://192.168.1.11/
<!DOCTYPE html>
<html>
  <head>
    <title>403 Forbidden</title>
  </head>
  <body>
    <h1>Error 403 Forbidden</h1>
    <p>Forbidden</p>
    <h3>Guru Meditation:</h3>
    <p>XID: 46</p>
    <hr>
    <p>Varnish cache server</p>
  </body>
</html>

3.4.3 使用 Chrome 进行访问（可正常访问）

在这里插入图片描述

4、修改请求报文格式（携带用户原IP地址，便于分析日志）

4.1 varnish 服务器配置并按 1.3 进行重载

[root@Tang varnish]# cat default.vcl 
sub vcl_recv {
    if (req.restarts == 0) {
        if (req.http.X-Fowarded-For) {       
            set req.http.X-Forwarded-For = req.http.X-Forwarded-For + "," + client.ip;
        } else {
	    set req.http.X-Forwarded-For = client.ip;
        }
    }
}

4.2 后端服务器的 httpd 的 log

### httpd.conf log 配置 ###
[root@Tang conf]# vim httpd.conf
LogFormat "%h %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\"" combined

### 日志 log ###
[root@Tang ~]# tail /var/log/httpd/access_log
192.168.100.101 - - [10/Oct/2019:22:08:27 -0400] "GET / HTTP/1.1" 200 48 "-" "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.121 Safari/537.36"
192.168.100.101 - - [10/Oct/2019:22:08:28 -0400] "GET /favicon.ico HTTP/1.1" 404 209 "http://172.16.141.252/" "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.121 Safari/537.36"
192.168.100.101 - - [10/Oct/2019:22:18:56 -0400] "GET / HTTP/1.1" 200 48 "-" "curl/7.47.0"

### httpd.conf log 配置 ###
[root@Tang conf]# vim httpd.conf
LogFormat "%{X-Forwarded-For}i %h %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\"" combined

### 日志 log ###
### 包含了请求客户的 IP 地址和后端主机 IP 地址 ###
[root@Tang ~]# tail /var/log/httpd/access_log
172.16.141.199 192.168.100.101 - - [10/Oct/2019:22:21:19 -0400] "GET / HTTP/1.1" 200 48 "-" "Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36"
172.16.141.199 192.168.100.101 - - [10/Oct/2019:22:27:33 -0400] "GET / HTTP/1.1" 200 48 "-" "Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36"
172.16.141.199 192.168.100.101 - - [10/Oct/2019:22:27:33 -0400] "GET /favicon.ico HTTP/1.1" 404 209 "http://172.16.141.252/" "Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36"

5、多个主机的情况下，不同主机的缓存不会交叉命中

5.1 varnish 服务器配置并按 1.3 进行重载

sub vcl_hash {
    hash_data(req.url);            # 从 根 开始的 URL
    if (req.http.host) {           # 不同主机的缓存不会交叉命中（主机只有一个的话，不需要此配置）      
        hash_data(req.http.host);   
    } else {
        hash_data(server.ip);
    }
    return (lookup);
}

[root@Tang varnish]# cat default.vcl 
vcl 4.0;
import directors;

backend vhost1 {
    .host = "192.168.100.101";
    .port = "8080";
}

backend vhost2 {
    .host = "192.168.100.102";
    .port = "8080";
}

backend vhost3 {
    .host = "192.168.100.103";
    .port = "8080";
}

sub vcl_init {
    new v = directors.round_robin();
    v.add_backend(vhost1);
    v.add_backend(vhost2);
    v.add_backend(vhost3);
}

sub vcl_recv {
    set req.backend_hint = v.backend();
}

sub vcl_recv {
    if (req.restarts == 0) {
        if (req.http.X-Fowarded-For) {       
            set req.http.X-Forwarded-For = req.http.X-Forwarded-For + "," + client.ip;
        } else {
	    set req.http.X-Forwarded-For = client.ip;
        }
    }
}

sub vcl_hash {
    hash_data(req.url);      
    if (req.http.host) {        
        hash_data(req.http.host);
    } else {
        hash_data(server.ip);
    }
    return (lookup);
}

sub vcl_deliver {
	if (obj.hits>0) {
    	set resp.http.X-Cache-Tang = "HIT via " + server.ip;
    } else {
    	set resp.http.X-Cache-Tang = "MISS via " + server.ip;
   	}
}

5.2 用户在进行服务器访问时，同一 IP 的用户会一直命中同一后端主机

在这里插入图片描述

6、运用 purge 和 ban 修剪缓存

6.1 运用 purge 修剪缓存

因为 purge 是一个比较危险的操作，所以需要需要配置 ACL ，只允许部分地址段的 IP ，进行 purge 操作。

所以，我们的配置思路为：

  1. 先配置 ACL
  2. 在配置 VCL 时，引用 ACL
  3. 允许 PURGE 操作的，可以进行 OURGE 操作
  4. 不允许 PURGE 操作的，拒绝此操作，并回复相关页面

6.1.1 varnish 服务器配置并按 1.3 进行重载

基本语法说明：

acl purgers {                      # 定义一个 ACL ，名称为 purges
    "127.0.0.1";
    "172.16.141.0"/24;
}

sub vcl_recv {
    if (req.method == "PURGE") {   # 引用 ACL 
        if (!client.ip ~ purgers) {   
            return (synth(405, "Purging not allowed for " + client.ip));
        }
        return (purge);
    }
}

varnishi 配置文件：

[root@Tang varnish]# cat default.vcl 
vcl 4.0;
import directors;

backend vhost1 {
    .host = "192.168.100.101";
    .port = "8080";
}

backend vhost2 {
    .host = "192.168.100.102";
    .port = "8080";
}

backend vhost3 {
    .host = "192.168.100.103";
    .port = "8080";
}

sub vcl_init {
    new v = directors.round_robin();
    v.add_backend(vhost1);
    v.add_backend(vhost2);
    v.add_backend(vhost3);
}

sub vcl_recv {
    set req.backend_hint = v.backend();
}

sub vcl_recv {
    if (req.restarts == 0) {
        if (req.http.X-Fowarded-For) {       
            set req.http.X-Forwarded-For = req.http.X-Forwarded-For + "," + client.ip;
        } else {
	    set req.http.X-Forwarded-For = client.ip;
        }
    }
}

acl purgers {
    "127.0.0.1";
    "172.16.141.0"/24;
}

sub vcl_recv {
    if (req.method == "PURGE") {
        if (!client.ip ~ purgers) {
            return (synth(405, "Purging not allowed for " + client.ip));
        }
        return (purge);
    }
}

sub vcl_deliver {
    if (obj.hits>0) {
        set resp.http.X-Cache-Tang = "HIT via " + server.ip;
    } else {
        set resp.http.X-Cache-Tang = "MISS via " + server.ip;
    }
}

6.1.2 用户进行 purge 操作

### 第一次访问，无法命中缓存 ###
root@Neo:~# curl -I http://172.16.141.252
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 03:08:00 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Thu, 10 Oct 2019 03:12:21 GMT
ETag: "30-59485c50d3bef"
Content-Length: 48
Content-Type: text/html; charset=UTF-8
X-Varnish: 163953
Age: 0
Via: 1.1 varnish-v4
X-Cache-Tang: MISS via 172.16.141.252
Connection: keep-alive

### 第二次访问，成功命中缓存 ###
root@Neo:~# curl -I http://172.16.141.252
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 03:08:00 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Thu, 10 Oct 2019 03:12:21 GMT
ETag: "30-59485c50d3bef"
Content-Length: 48
Content-Type: text/html; charset=UTF-8
X-Varnish: 313 163954
Age: 2
Via: 1.1 varnish-v4
X-Cache-Tang: HIT via 172.16.141.252
Connection: keep-alive

### 携带请求方法 PURGE 进行缓存清除 ###
root@Neo:~# curl -X PURGE http://172.16.141.252
<!DOCTYPE html>
<html>
  <head>
    <title>200 Purged</title>
  </head>
  <body>
    <h1>Error 200 Purged</h1>
    <p>Purged</p>
    <h3>Guru Meditation:</h3>
    <p>XID: 315</p>
    <hr>
    <p>Varnish cache server</p>
  </body>
</html>

### 再次访问，无法命中缓存 ###
root@Neo:~# curl -I http://172.16.141.252
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 03:34:48 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Thu, 10 Oct 2019 03:12:49 GMT
ETag: "30-59485c6ba946e"
Content-Length: 48
Content-Type: text/html; charset=UTF-8
X-Varnish: 163956
Age: 0
Via: 1.1 varnish-v4
X-Cache-Tang: MISS via 172.16.141.252
Connection: keep-alive

### 再次访问，依旧能够成功命中缓存 ###
root@Neo:~# curl -I http://172.16.141.252
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 03:34:48 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Thu, 10 Oct 2019 03:12:49 GMT
ETag: "30-59485c6ba946e"
Content-Length: 48
Content-Type: text/html; charset=UTF-8
X-Varnish: 317 163957
Age: 21
Via: 1.1 varnish-v4
X-Cache-Tang: HIT via 172.16.141.252
Connection: keep-alive

6.1.3 用户进行 purge 操作（不在 ACL 内的用户）

### 修改 vcl 配置文件 ###
acl purgers {
    "127.0.0.1";
#    "172.16.141.0"/24;
}
sub vcl_recv {
    if (req.method == "PURGE") {
        if (!client.ip ~ purgers) {
            return (synth(405, "Purging not allowed for " + client.ip));
        }
        return (purge);
    }
}

### 用户进行 PURGE 操作 ###
### 回复界面，就是在 vcl 配置的合成界面 ###
root@Neo:~# curl -X PURGE http://172.16.141.252
<!DOCTYPE html>
<html>
  <head>
    <title>405 Purging not allowed for 172.16.141.209</title>
  </head>
  <body>
    <h1>Error 405 Purging not allowed for 172.16.141.209</h1>
    <p>Purging not allowed for 172.16.141.209</p>
    <h3>Guru Meditation:</h3>
    <p>XID: 163959</p>
    <hr>
    <p>Varnish cache server</p>
  </body>
</html>

6.2、运用 ban 修剪缓存

6.2.1 ban 语法格式（需要在 varnishadm 下进行 ban 操作）

# 运用 ban 清除正则表达式匹配到的缓存（以 /javascripts 开头的 URL） 
varnish> ban req.url ~ ^/javascripts   
200 

# 运用 ban 清除正则表达式匹配到的缓存（以 .js 结尾的 URL）
varnish> ban req.url ~ .js$            
200 

# 清除全部缓存
varnish> ban req.url == /     
200  

# 清除访问 neotang.com 的全部缓存
varnish> ban req.url == / && req.http.host ~ "neotang.com"
200

# 清除主机 172.16.0.6 且 url 是 /javascripts/test.js 的缓存
varnish> ban req.http.host == 172.16.0.6 && req.url == /javascripts/test.js
200

6.2.2 ban 应用示例

  1. 客户端进行 test.js 内容访问，第一次访问，没有缓存

    root@Neo:~# curl -I http://172.16.141.252/test.js
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 07:22:42 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Fri, 11 Oct 2019 04:15:17 GMT
ETag: "5-5949ac402c30b"
Content-Length: 5
Content-Type: application/javascript
X-Varnish: 337
Age: 0
Via: 1.1 varnish-v4
X-Cache-Tang: MISS via 172.16.141.252
Connection: keep-alive

  2. 客户端进行 test.js 内容访问，第二次访问，命中缓存

    root@Neo:~# curl -I http://172.16.141.252/test.js
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 07:22:42 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Fri, 11 Oct 2019 04:15:17 GMT
ETag: "5-5949ac402c30b"
Content-Length: 5
Content-Type: application/javascript
X-Varnish: 163979 338
Age: 2
Via: 1.1 varnish-v4
X-Cache-Tang: HIT via 172.16.141.252
Connection: keep-alive

  3. 在 varnishadm 下进行缓存清除（ .js 结尾的缓存进行清除）

    varnish> ban req.url ~ .js$
200

  4. 客户端再次进行 test.js 内容访问，无法命中缓存，缓存被清除符合预期

    root@Neo:~# curl -I http://172.16.141.252/test.js
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 07:25:07 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Fri, 11 Oct 2019 04:15:17 GMT
ETag: "5-5949ac402bf23"
Content-Length: 5
Content-Type: application/javascript
X-Varnish: 163981
Age: 0
Via: 1.1 varnish-v4
X-Cache-Tang: MISS via 172.16.141.252
Connection: keep-alive

  5. 客户端再次进行 test.js 内容访问，命中缓存

    root@Neo:~# curl -I http://172.16.141.252/test.js
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 07:25:07 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Fri, 11 Oct 2019 04:15:17 GMT
ETag: "5-5949ac402bf23"
Content-Length: 5
Content-Type: application/javascript
X-Varnish: 98319 163982
Age: 37
Via: 1.1 varnish-v4
X-Cache-Tang: HIT via 172.16.141.252
Connection: keep-alive

  6. 运用 ban 再次进行缓存清除

    varnish> ban req.http.host == 172.16.141.252
200

  7. 客户端再次进行访问，无法命中缓存

    root@Neo:~# curl -I http://172.16.141.252/test.js
HTTP/1.1 200 OK
Date: Fri, 11 Oct 2019 07:46:15 GMT
Server: Apache/2.4.6 (CentOS) PHP/5.4.16
Last-Modified: Fri, 11 Oct 2019 04:15:17 GMT
ETag: "5-5949ac402bf23"
Content-Length: 5
Content-Type: application/javascript
X-Varnish: 98321
Age: 0
Via: 1.1 varnish-v4
X-Cache-Tang: MISS via 172.16.141.252
Connection: keep-alive

7、多个后端主机配置

配置思路：

  1. 定义后端主机
  2. 定义后端主机的集群服务，并指定调度方式
  3. 引用后端主机的集群服务名称

配置语法：

import directors;    # load the directors

backend server1 {
	.host = 
	.port = 
}
backend server2 {
	.host = 
	.port = 
}

sub vcl_init {
	new GROUP_NAME = directors.round_robin();
	GROUP_NAME.add_backend(server1);
	GROUP_NAME.add_backend(server2);
}

sub vcl_recv {
	# send all traffic to the bar director:
	set req.backend_hint = GROUP_NAME.backend();
}

7.1 后端主机配置（一个主机配置多个虚拟主机）

在这里插入图片描述

7.1.1 后端主机的 httpd 的 conf 文件配置

[root@Tang conf]# cat /etc/httpd/conf.d/test.conf 
<VirtualHost 192.168.100.101:8080>
	ServerName 192.168.100.101
	DocumentRoot "/data/web/vhost1"
	<Directory "/data/web/vhost1">
		Options FollowSymLinks
		AllowOverride None
		Require all granted
	</Directory>
</VirtualHost>

<VirtualHost 192.168.100.102:8080>
        ServerName 192.168.100.102
        DocumentRoot "/data/web/vhost2"
        <Directory "/data/web/vhost2">
                Options FollowSymLinks
                AllowOverride None
                Require all granted
        </Directory>
</VirtualHost>

<VirtualHost 192.168.100.103:8080>
        ServerName 192.168.100.103
        DocumentRoot "/data/web/vhost3"
        <Directory "/data/web/vhost3">
                Options FollowSymLinks
                AllowOverride None
                Require all granted
        </Directory>
</VirtualHost>

7.1.2 验证后端主机的各个虚拟主机

[root@Tang conf]# for i in {1..3}; do cat /data/web/vhost$i/index.html; done;
<h1>Backend Http-Server-01 192.168.100.101</h1>
<h1>Backend Http-Server-02 192.168.100.102</h1>
<h1>Backend Http-Server-03 192.168.100.103</h1>

[root@Tang conf]# for i in {1..3}; do curl http://192.168.100.10$i; done
<h1>Backend Http-Server-01 192.168.100.101</h1>
<h1>Backend Http-Server-02 192.168.100.102</h1>
<h1>Backend Http-Server-03 192.168.100.103</h1>

7.2 varnish 服务器配置并按 1.3 进行重载

[root@Tang varnish]# cat default.vcl 
vcl 4.0;
import directors;

backend vhost1 {
    .host = "192.168.100.101";
    .port = "8080";
}

backend vhost2 {
    .host = "192.168.100.102";
    .port = "8080";
}

backend vhost3 {
    .host = "192.168.100.103";
    .port = "8080";
}

sub vcl_init {
    new v = directors.round_robin();
    v.add_backend(vhost1);
    v.add_backend(vhost2);
    v.add_backend(vhost3);
}

sub vcl_recv {
    set req.backend_hint = v.backend();
}

sub vcl_recv {
    if (req.restarts == 0) {
        if (req.http.X-Fowarded-For) {       
            set req.http.X-Forwarded-For = req.http.X-Forwarded-For + "," + client.ip;
        } else {
	    set req.http.X-Forwarded-For = client.ip;
        }
    }
}

acl purgers {
    "127.0.0.1";
    "172.16.141.0"/24;
}

sub vcl_recv {
    if (req.method == "PURGE") {
        if (!client.ip ~ purgers) {
            return (synth(405, "Purging not allowed for " + client.ip));
        }
        return (purge);
    }
}

sub vcl_deliver {
    if (obj.hits>0) {
        set resp.http.X-Cache-Tang = "HIT via " + server.ip;
    } else {
        set resp.http.X-Cache-Tang = "MISS via " + server.ip;
    }
}

7.3 用户进行访问

使用一台用户主机进行访问，访问步骤：

  1. 进行 http 访问
  2. 缓存清除
  3. 再次进行 http 访问，看后端主机是否变化
  4. 缓存清除
  5. 再次进行 http 访问，看后端主机是否变化
### 通过访问，发现 varnish 服务器在轮流把请求发往后端主机 ###
root@Neo:~# curl http://172.16.141.252
<h1>Backend Http-Server-01 192.168.100.101</h1>

root@Neo:~# curl -X PURGE http://172.16.141.252 > /dev/null
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   238  100   238    0     0   114k      0 --:--:-- --:--:-- --:--:--  232k

root@Neo:~# curl http://172.16.141.252
<h1>Backend Http-Server-02 192.168.100.102</h1>

root@Neo:~# curl -X PURGE http://172.16.141.252 > /dev/null
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   238  100   238    0     0   112k      0 --:--:-- --:--:-- --:--:--  232k

root@Neo:~# curl http://172.16.141.252
<h1>Backend Http-Server-03 192.168.100.103</h1>

7.4 后端主机的健康状态检测配置

7.4.1 varnish 服务器配置并按 1.3 进行重载

[root@Tang varnish]# cat default.vcl
vcl 4.0;
import directors;

# 检测 .healthcheck.html 是否能正常访问，访问五次，有四次成功，说明状态健康 #
# 超时时间 1s， 检测间隔时间为 2s #
probe check {
    .url = "/.healthcheck.html";
    .timeout = 1s;
    .window = 5;
    .threshold = 4;
    .interval = 2s;
}

backend vhost1 {
    .host = "192.168.100.101";
    .port = "8080";
    .probe = check;
}

backend vhost2 {
    .host = "192.168.100.102";
    .port = "8080";
    .probe = check;
}


backend vhost3 {
    .host = "192.168.100.103";
    .port = "8080";
    .probe = check;
}

sub vcl_init {
    new v = directors.round_robin();
    v.add_backend(vhost1);
    v.add_backend(vhost2);
    v.add_backend(vhost3);
}

sub vcl_recv {
    set req.backend_hint = v.backend();
}

sub vcl_recv {
    if (req.restarts == 0) {
        if (req.http.X-Fowarded-For) {       
            set req.http.X-Forwarded-For = req.http.X-Forwarded-For + "," + client.ip;
        } else {
	    set req.http.X-Forwarded-For = client.ip;
        }
    }
}

acl purgers {
    "127.0.0.1";
    "172.16.141.0"/24;
}

sub vcl_recv {
    if (req.method == "PURGE") {
        if (!client.ip ~ purgers) {
            return (synth(405, "Purging not allowed for " + client.ip));
        }
        return (purge);
    }
}

sub vcl_backend_response {
}

sub vcl_deliver {
    if (obj.hits>0) {
        set resp.http.X-Cache-Tang = "HIT via " + server.ip;
    } else {
        set resp.http.X-Cache-Tang = "MISS via " + server.ip;
    }
}

7.4.2 httpd 服务器配置

[root@Tang conf]# for i in {1..3}; do echo "I am Healthy" > /data/web/vhost$i/.healthcheck.html; done;
[root@Tang conf]# for i in {1..3}; do cat /data/web/vhost$i/.healthcheck.html; done;
I am Healthy
I am Healthy
I am Healthy

7.4.3 varnishadm 查看后端服务器的状态

varnish> backend.list 
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Healthy 5/5

7.4.4 手动 down 掉一个 IP 地址（有一个后端主机状态为 sick）

[root@Tang ~]# ifconfig enp1s0:3 192.168.100.103 down

varnish> backend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Sick 3/5

varnish> backend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Sick 0/5

7.4.5 手动 up 后，在查看状态（检测四次成功后，状态变为 healthy）

varnish> backend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Sick 0/5

backend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Sick 1/5
backend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Sick 2/5

backend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Sick 3/5
backend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Healthy 4/5

backend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Healthy 5/5

7.4.6 手动设置状态示例

varnish> backend.list 
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      probe      Healthy 5/5

varnish> backend.set_health vhost3 sick
200        

varnish> backend.set_healtlist
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      sick       Healthy 5/5

varnish> backend.set_health vhost3 healthy
200        

varnish> backend.sbackend.list
200        
Backend name                   Refs   Admin      Probe
default(192.168.100.100,,8080) 3      probe      Healthy (no probe)
vhost1(192.168.100.101,,8080)  9      probe      Healthy 5/5
vhost2(192.168.100.102,,8080)  9      probe      Healthy 5/5
vhost3(192.168.100.103,,8080)  9      healthy    Healthy 5/5
Lee木木
关注
  • 0
    点赞
  • 踩
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
Varnish用法详解
小新没有蜡笔
11-05 1203
简介：This is Varnish Cache, a high-performance HTTP accelerator varnish相关varnish 相关文件路径/etc/varnish/varnish.params： 配置varnish服务进程的工作特性，例如监听的地址和端口，缓存机制； /etc/varnish/default.vcl：配置各Child/Cache线程的缓存策略；var
varnish基础概念详解
weixin_33978016的博客
07-21 149
varnish基础概念详解比起squid更加轻量级，大致有以下几个特点：·可以基于内存缓存，也可以在磁盘上缓存，但是就算存放在磁盘上，也不能实现持久缓存    只要进程崩溃，此前缓存统统失效，无论是在内存还是在磁盘，但是现在已经具备持久缓存功能，但是仍然在实验阶段，经常容易崩溃，而且最大大小不能超过1G    如果期望内存大小超过几十个G，比如图片服务器，纯粹使用内存，性能未必好，这时候可以使用磁...
varnish 详解
weixin_44983653的博客
10-10 718
varnish 详解相关概念1、Web Page Cache2、Cache-related Headers Fields3、请求报文和响应报文关于缓存的内容varnish1、概念2、程序架构3、程序环境4、缓存存储机制5、varnish 程序的选项6、重载 vcl 配置文件（两种方法）7、varnishadm 介绍8、vcl 介绍8.1 state engine8.2 vcl 的默认配置查看8.3...
varnish详解
weixin_34284188的博客
07-27 303
开源解决方案：squid：varnish:默认端口：6081、127.20.0.1:6082 varnish官方站点： http://www.varnish-cache.org/ Community Enterprise This is Varnish Cache, a high-performance HTTP accelerator....
Varnish配置文件详解(架构师之路)
09-15
主要介绍了Varnish配置文件详解(架构师之路),需要的朋友可以参考下
PHP 清空varnish 缓存的详解(包括指定站点下的)
10-27
本篇文章是对清空varnish的缓存(包括指定站点下的)进行了详细的分析介绍，需要的朋友参考下
Varnish Cahe v6.0.0
10-14
Varnish cache，或称Varnish，是一个开源、高性能的反向代理服务器（reverse proxy server）。 Varnish目前被用在挪威最大的报社Verdens Gang上。
varnish测试报告
11-20
varnish测试报告
Varnish详解
weixin_34351321的博客
09-25 102
Varnish cache，或称Varnish，是一套高效能的反向代理，缓存器（reverseproxy server）。varnish主要运行两个进程：Management进程和Child进程(也叫Cache进程)。安装varnish，管网下载rpm包，varnish-3.0.5-1.el6.x86_64.rpm varnish-docs-3.0.5-1.el6.x86_...
varnish安装及配置（二）
莫激动
01-10 190
  Varish Configuration Language 简称VCL，通过它我们可以完成一些复杂的逻辑处理。下面将详细介绍：   Backend declarations:   backend www { .host = "www.example.com"; .port = "80"; .connect_timeout = 1s; .first_byte_t...
varnish详解与功能搭建
wnccmyr的博客
08-28 722
varnish介绍 Varnish cache，或称Varnish，是一套高性能的反向网站缓存服务器（reverse proxy server） varnish拥有俩套配置文件；一套配置文件用于varnish自身进程的参数配置，另一套用于定义缓存规则；定义缓存规则需要使用灵活的语言来定义，这就是VCL（varnish语言）；应用时需要将VCL编写的规则送给VCC编译后才能运行，所以安装varnish需要依赖gcc编译器。 安装 下载相应的软件包 jemalloc-3.6.0-1.el7.x86_6
Varnish 进阶配置与实例
achuDk的博客
06-29 1117
Varnish 进阶配置与实例 注意：使用varnishadm命令的前提是在varnish的default.vcl文件所在目录下 yum install varnish cd /etc/varnish cp default.vcl{,.bak} cp varnish.params{,.bak}vim varnish.params RELOAD_VCL=1 VARNISH_VCL_CONF=/e
varnish入门理解
ThinkerWalker
08-26 2735
简介 Varnish 是一款高性能且开源的反向代理服务器和 HTTP 加速器,开源的软件。但是也有企业版，有更多的功能。 我们知道，在计算机界，缓存为王。我们访问百度时候，有没有觉得在访问时候相当的快，这就是CDN缓存的结果。以前，常用squid来进行缓存，但是squid已经是过去时了。varnish和squid的关系，相当于现在的nginx和Apache的关系，squid已经是负重前
varnish缓存三：VCL配置示例
weixin_33836874的博客
12-11 72
我公司当前线上varnish配置示例：#backend default {#.host = "127.0.0.1";#.port = "8080";#}backend s360_01 {.host = "192.168.0.114";.port = "18286";.probe = { .url = "/"; .interval = 5s; ...
Varnish+Nginx 配置----Varnish
热门推荐
anyx的专栏
04-17 1万+
最近项目引入反向代理和缓存，熟悉了一下Squid、Apache、Valish、Nginx，根据项目实际进行选择，客观来说，采用Linux系统部署最好，也没有什么难度，但实际情况必须采用Windows系统（本着方案要结合现实的原则,研究要以Windows平台为主）。 一  Varnish启动: e: cd e:\varnish\bin set PATH=%CD%;%PATH% var
centos 7.5 部署varnish缓存服务器
lvjianzhaoa的博客
10-29 253
博文大纲： 一、varnish简介 二、Varnish 如何工作 三、部署varnish缓存服务器 一、varnish简介 Varnish是高性能开源的反向代理服务器和HTTP缓存服务器，其功能与Squid服务器相似，都可以用来做HTTP缓存。可以安装 varnish 在任何web前端，同时配置它缓存内容。与传统的 squid 相比，varnish 具有性能更高、速度更快、管理更加方便等诸...
varnish log
最新发布
08-19
- *1* [varnishlog、Varnishstat详解](https://blog.csdn.net/weixin_30709809/article/details/97828606)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_...

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值