В логах постоянно:
upstream timed out (110: Connection timed out) while reading response header from upstream
и
connect() failed (110: Connection timed out) while connecting to upstream

Что делать?
nginx.conf:

worker_process 4;
worker_limit_nofile 16384;

events {
worker_connections 2048;
}

http {
sendfile on;
tcp_nopush on;
tcp_nodelay on;
keepalive_timeout 30;
directio 10m;
expires max;
proxy_buffers 8 32k;
proxy_buffer_size 64k;
client_header_timeout 3m;
client_body_timeout 3m;
send_timeout 3m;
proxy_connect_timeout 1m;
proxy_send_timeout 1m;
proxy_read_timeout 1m;
gzip on;
}

Уже по всякому игрался с настройками - пофиг чаще всего происходит при POST. но бывает и на GET

httpd.conf

Timeout 60

KeepAlive On

MaxKeepAliveRequests 200

KeepAliveTimeout 15

<IfModule prefork.c>
StartServers 16
MinSpareServers 10
MaxSpareServers 64
ServerLimit 512
MaxClients 512
MaxRequestsPerChild 1000
</IfModule>

<IfModule worker.c>
StartServers 16
MaxClients 300
MinSpareThreads 25
MaxSpareThreads 75
ThreadsPerChild 100
MaxRequestsPerChild 1000
</IfModule>

Listen 8080


Не справляться нечему, нагрузка на сервер маленькая ( 2000-3000 хостов в сутки )


# top
top - 14:12:39 up 15 days, 3:13, 1 user, load average: 0.50, 0.45, 0.45
Tasks: 153 total, 1 running, 152 sleeping, 0 stopped, 0 zombie
Cpu0 : 1.9%us, 0.4%sy, 0.0%ni, 97.3%id, 0.4%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu1 : 1.9%us, 0.4%sy, 0.0%ni, 97.3%id, 0.4%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu2 : 2.2%us, 0.5%sy, 0.0%ni, 97.2%id, 0.1%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu3 : 2.2%us, 0.5%sy, 0.0%ni, 97.2%id, 0.1%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 3613708k total, 3324976k used, 288732k free, 266244k buffers
Swap: 4194300k total, 9444k used, 4184856k free, 2758048k cached