nginx - Elixir Phoenix - 跟踪由对等错误重置的连接

Question

我正在运行一个非常高并发的 Phoenix 应用程序，我注意到大约 10% 的请求在读取上游时出现 readv() failed (104: Connection reset by peer)我的 nginx 错误文件中的错误。我在我的第一个插件的开头添加了一个日志语句来打印传入的请求，但我没有看到请求到达。我不知道为什么会收到此错误以及如何调试它。

附件是我的 nginx 配置文件的副本:

worker_processes 4;
user nginx nginx;

pid /var/run/nginx.pid;
error_log /var/log/nginx/error.log;

events {
  worker_connections 1024; # increase if you have lots of clients
  accept_mutex on; # "on" if nginx worker_processes > 1
  use epoll; # enable for Linux 2.6+
}

http {
  include mime.types;

  default_type application/octet-stream;
  access_log /var/log/nginx/access.log;
  error_log /var/log/nginx/error.log;
  sendfile on;
  tcp_nopush on;
  tcp_nodelay off;

  gzip on;
  gzip_http_version 1.0;
  gzip_proxied any;
  gzip_min_length 500;
  gzip_disable "MSIE [1-6]\.(?!.*SV1)";
  gzip_types text/plain text/xml text/css
             text/comma-separated-values
             text/javascript application/x-javascript
             application/atom+xml application/json;
  proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
  proxy_buffers 8 64k;
  proxy_buffer_size 128k;
  large_client_header_buffers 4 16k;

  upstream nolen {
    server 127.0.0.1:8080;
  }

  server {
      listen 80;
      server_name _;

      access_log /var/log/nginx/access.log;
      error_log /var/log/nginx/error.log;

      location / {
        try_files $uri @proxy;
      }

      location @proxy {
          proxy_redirect off;
          proxy_pass http://nolen;
      }
  }

  server {
    listen 443 ssl;
    server_name _;

    ssl_certificate /etc/ssl/selfsigned.crt;
    ssl_certificate_key /etc/ssl/selfsigned.key;
    ssl_ciphers HIGH:MEDIUM:!aNULL:!ADH:!DH:!EDH:!CAMELLIA:!KRB5:!IDEA:!EXP:!eNULL:!LOW:RC4+RSA:+HIGH:+MEDIUM;
    ssl_protocols TLSv1 TLSv1.1 TLSv1.2;

    access_log /var/log/nginx/ssl.access.log;
    error_log /var/log/nginx/ssl.error.log;

    location / {
      try_files $uri @proxy;
    }

    location @proxy {
        proxy_redirect off;
        proxy_pass http://nolen;
    }
  }

  include /etc/nginx/conf.d/cloudflare.conf;
}

我注意到采用不同查询参数的两个 url 以及静态 js 文件的服务都可以确认正常工作的错误。

任何人都可以帮助指导我完成调试此问题的过程吗？谢谢。

Answer 1

这通常是由于超时导致远程套接字上的连接丢失。对于您的测试，您可以在发送请求之前尝试 ping 检查状态，这就是为什么您能够隔离 10%