Question Nginx configuration for 1000 domains

[stas styler]

Hi pleskians,

I'm having trouble with my server its getting 504\2 too often and I assume that the nginx.conf is the problem.
But i cant find it.

My server is 48 cores 128 GB ram with NVME 2tb, running centos7, nginx as reversed proxy and apache, 1gbit/s link.
this is my nginx.conf

#user  nginx;
worker_processes  48;
worker_rlimit_nofile 49152;
#error_log  /var/log/nginx/error.log;
#error_log  /var/log/nginx/error.log  notice;
#error_log  /var/log/nginx/error.log  info;

#pid        /var/run/nginx.pid;

include /etc/nginx/modules.conf.d/*.conf;

events {
    worker_connections 24576;
    multi_accept on;
    use epoll;
}


http {

include       mime.types;
    default_type  application/octet-stream;

    #log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
    #                  '$status $body_bytes_sent "$http_referer" '
    #                  '"$http_user_agent" "$http_x_forwarded_for"';

    access_log  /dev/null;
    error_log   /dev/null;
    keepalive_timeout  15;
    server_tokens off;
    open_file_cache          max=10000 inactive=20s;
    open_file_cache_valid    30s;
    open_file_cache_min_uses 2;
    open_file_cache_errors off;
    client_body_buffer_size 128K;
    client_header_buffer_size 1k;
    client_max_body_size 8m;
    large_client_header_buffers 4 256k;
    fastcgi_buffers 8 16k;
    fastcgi_buffer_size 32k;
    fastcgi_connect_timeout 600;
    fastcgi_send_timeout 600;
    fastcgi_read_timeout 600;
    proxy_connect_timeout       600;
    proxy_send_timeout          600;
    proxy_read_timeout          600;
    sendfile on;
    tcp_nopush on;
    tcp_nodelay on;
    client_header_timeout 15;
    client_body_timeout 15;
    send_timeout 15;
    autoindex off;


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

# override global parameters e.g. worker_rlimit_nofile
include /etc/nginx/*global_params;

I'm trying to make that to work good but unfortunatelly i'm having problems.
I got 200 domains currently, some are high traffic, some are not (AVG 700Packets/s).

Is my nginx.conf configured somehow bad so its just timing out and crashes?

 

[Brujo]

Well there are a lot of possabillitys which can cause the 504 & 502 messages, even if you search the forum for "Nginx 502 or 504", you will find dozens of threads. I assume you are you talking about this error on customer domains?

therefore you also should check the /var/log/nginx/error.log and /var/www/vhosts/system/example.com/logs/proxy_error_log for detailed error message. Helpfull is also when you provide OS & plesk version you have.

see for example:
[HUB] 502 Bad Gateway
A website is not accessible with "Nginx 504 Gateway Time-out": upstream timed out (110: Connection timed out) while reading response header from upstream
Websites are unavailable with 502 Bad Gateway on Plesk: upstream sent too big header while reading response header

 

[Bitpalast]

You ought to make sure that Apache is set to "graceful" restarts as described in How to enable/disable graceful restart for Apache? . Ignore the red warning in this article.