1. 故障现象

架构如下:

当时servera的容器oom了,开始不停重启,造成80端口是活着的,但没法正常提供http服务
原先的配置如下:

upstream aaa {
	server 192.168.31.17;
	server 192.168.31.27;
}
server {
	listen 192.168.31.67:80;
        server_name aa.qiuqin.com 192.168.31.67;
        location / {
            proxy_pass http://aaa;
        }
}
1
2
3
4
5
6
7
8
9
10
11

2. 安装和配置Nginx第三方模块

2.1 下载nginx_upstream_check_module

https://github.com/yaoweibin/nginx_upstream_check_module

2.2 下载nginx

https://nginx.org/en/download.html

2.3 安装依赖

yum install -y gcc‐c++ pcre pcre‐devel zlib zlib‐devel openssl openssl‐devel patch pcre-devel zlib unzip
1

2.4 上传nginx和nginx_upstream_check_module

mkdir -p /apps/install/
cd /apps/install/
unzip nginx_upstream_check_module-master.zip
tar xf nginx-1.20.2.tar.gz
1
2
3
4

2.5 为nginx更新模块补丁

cd /apps/install/nginx-1.20.2
patch -p1 < /apps/install/nginx_upstream_check_module-master/check_1.20.1+.patch
1
2

2.6 编辑安装nginx

由于是测试环境新装就直接configure了

./configure --with-http_stub_status_module \
            --add-module=/apps/install/nginx_upstream_check_module-master
            make -j 4 && make install
            ln -sf /usr/local/nginx/sbin/nginx /usr/bin/nginx

1
2
3
4
5

如果是原来的升级先执行

nginx -V
1

把原来的配置复制出来,再追加–with-http_stub_status_module --add-module=/apps/install/nginx_upstream_check_module-master

3. 配置文件修改

我单独配置了个conf,也可以直接写在nginx.conf里
/usr/local/nginx/conf/conf.d/aa.conf

upstream aaa {
	server 192.168.31.17;
	server 192.168.31.27;
        check interval=3000 rise=2 fall=5 timeout=1000 type=http;
        check_http_send "HEAD / HTTP/1.0\r\n\r\n";
        check_http_expect_alive http_2xx http_3xx;
}
server {
	listen 192.168.31.67:80;
        server_name aa.qiuqin.com 192.168.31.67;
        location / {
            proxy_pass http://aaa;
        }
}
1
2
3
4
5
6
7
8
9
10
11
12
13
14

3.1 测试效果

两台web都正常启动

关闭27的nginx,并启动80的监听

此时27上80端口可以被telnet到,但nginx已经不会再转发请求了