Why my nginx handles requests sequentially and not in parallel?

Question

I've implemented a special (it has an artificial await) API route in my backend which takes 1 second to return data. Having nginx configured as below I expected 2 requests sent at the same time to complete after 1 second. They complete after 2 seconds.

After analysing logs I can see that in fact each instance of my backend receives 1 requests but one of them receives request ~1s later which indicates that NGINX does sequentiall and not parallel processing.

This is my reverse proxy conifguration (/etc/nginx/conf.d/default.conf):

upstream backend {
    server host.docker.internal:4001;
    server host.docker.internal:4002;
}


server {
    listen       80;
    listen  [::]:80;
    server_name  something.com;

    location /api/ {
        proxy_pass http://backend;
    }
}

This is my /etc/nginx/nginx.conf

worker_processes  auto;

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


events {
    worker_connections  1024;
    multi_accept on;
}


http {
    include       /etc/nginx/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  /var/log/nginx/access.log  main;

    sendfile        on;
    #tcp_nopush     on;

    keepalive_timeout  65;

    #gzip  on;

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

My VPS has 4 cores and upon nginx startup I can see 4 lines "Starting worker process" so this seems to be working fine. What else might be the case?

Answer 1

Turns out my browser was using only 1 agent to send requests. When using 2 browsers/agents it works as intended