Prometheus使用nginx-lua监控Nginx

这里只是对prometheus监控Nginx的配置过程做个记录，方便以后查阅。

1.下载安装LuaJIT

http://luajit.org/download.html
wget http://luajit.org/download/LuaJIT-2.0.5.tar.gz
tar xf LuaJIT-2.0.5.tar.gz
cd LuaJIT-2.0.5
make && sudo make install

2.下载ngx_devel_kit模块

https://github.com/simplresty/ngx_devel_kit/releases
tar xf ngx_devel_kit-0.3.0.tar.gz -C /usr/local/src/

3.下载lua-nginx模块

https://github.com/openresty/lua-nginx-module/releases
wget https://github.com/openresty/lua-nginx-module/archive/v0.10.15.tar.gz
tar xf v0.10.15.tar.gz 
sudo mv -f lua-nginx-module-0.10.15/ /usr/local/lib/

4.安装nginx依赖包zlib，pcre

sudo yum -y install pcre-devel openssl openssl-devel
sudo yum install zlib-devel

5.下载安装nginx

wget http://nginx.org/download/nginx-1.13.5.tar.gz
tar -xvf nginx-1.13.5.tar.gz

安装nginx、nginx-lua 模块

export LUAJIT_LIB=/usr/local/lib
export LUAJIT_INC=/usr/local/include/luajit-2.1
cd nginx-1.13.5
./configure --prefix=/work/admin/nginx --add-module=/usr/local/lib/lua-nginx-module-0.10.15 --add-module=/usr/local/src/ngx_devel_kit-0.3.0
make
sudo make install

查看nginx模块，可能会报以下错误

./nginx/sbin/nginx -V
./nginx/sbin/nginx: error while loading shared libraries: libluajit-5.1.so.2: cannot open shared object file: No such file or directory

解决方法

sudo ln -s /usr/local/lib/libluajit-5.1.so.2 /lib64/libluajit-5.1.so.2
./nginx/sbin/nginx -V
nginx version: nginx/0.0.1
built by gcc 4.8.5 20150623 (Red Hat 4.8.5-36) (GCC) 
configure arguments: --prefix=/work/admin/nginx --add-module=/usr/local/lib/lua-nginx-module-0.10.15

6.下载nginx-lua-prometheus

wget https://github.com/knyar/nginx-lua-prometheus/archive/0.20181120.tar.gz
tar xf 0.20181120.tar.gz
mv -f nginx-lua-prometheus-0.20181120 /work/admin/nginx-lua-prometheus

7.nginx lua 配置

#nginx monitor lua
    lua_shared_dict prometheus_metrics 10M;
    lua_package_path "/work/admin/nginx-lua-prometheus/?.lua";
    init_by_lua '
        prometheus = require("prometheus").init("prometheus_metrics")
        metric_requests = prometheus:counter(
                "nginx_http_requests_total", "Number of HTTP requests", {"host", "status"})
        metric_latency = prometheus:histogram(
                "nginx_http_request_duration_seconds", "HTTP request latency", {"host"})
        metric_connections = prometheus:gauge(
                "nginx_http_connections", "Number of HTTP connections", {"state"})
        metric_requests_uri = prometheus:counter(
                "nginx_http_requests_uri_total", "Number of HTTP requests_uri", {"host","uri", "status"})';
    log_by_lua '
        metric_requests:inc(1, {ngx.var.server_name, ngx.var.status})
        metric_requests_uri:inc(1, {ngx.var.server_name,ngx.var.request_uri, ngx.var.status})
        metric_latency:observe(tonumber(ngx.var.request_time), {ngx.var.server_name})';


    server {

        listen       9145;
        server_name  nginx-metrics.feiersmart.local;
        allow 192.168.0.0/16;
        deny all;

        location /metrics {
        access_log off;
        content_by_lua '
          metric_connections:set(ngx.var.connections_reading, {"reading"})
          metric_connections:set(ngx.var.connections_waiting, {"waiting"})
          metric_connections:set(ngx.var.connections_writing, {"writing"})
          prometheus:collect()';
        }
    }

8.prometheus 拉取监控指标

- job_name: 'wan'
  static_configs:
  - targets: ['192.168.6.214:9100']
    labels: {cluster: 'product',type: 'basic',env: 'wan',job: 'wan1',export: 'wan1'}
  - targets: ['192.168.6.214:9145']
    labels: {cluster: 'product',type: 'middle',env: 'wan',job: 'wan1',export: 'nginx-lua-exporter'}

重启prometheus

如果开启了api管理功能，可以如下方式热加载配置

curl -X POST http://127.0.0.1:9090/-/reload

9.grafana出图

推荐使用9516

效果如下

上图监控面板我经过一些改动，仅供参考