我正在使用 haproxy 和 Apache 来保持 EC2 服务器处于活动状态。有时 haproxy 会停止接受新请求。如果我重新加载 haproxy,它会开始正常工作。我已禁用 iptables 并增加打开文件限制,但没有成功。以下是我的配置。
请指导我我做错了什么。
global
log 127.0.0.1 local5
maxconn 190001
daemon
nbproc 2
stats bind-process 1
stats socket /tmp/haproxy level admin
debug
defaults
log global
timeout server 86400000
timeout connect 86400000
timeout client 86400000
#timeout queue 30s
#timeout tarpit 60s
#timeout client 30s
#timeout connect 30s
#timeout server 30s
#option abortonclose
#option httpclose
mode http
option httplog
#option dontlognull
option forwardfor
#option http-server-close
#timeout http-keep-alive 30s
stats enable
stats hide-version
stats show-desc Master Server
stats auth admin:password
#stats auth admin:admin
stats uri /haproxyStats
stats refresh 15s
#errorfile 503 /etc/haproxy/maintenance.html
frontend HTTP
bind 0.0.0.0:80
acl network_allowed src 1.1.1.1
acl restricted_page path_beg /blog/wp-admin
block if restricted_page !network_allowed
reqadd X-Forwarded-Proto:\ http
acl is_blog url_beg /blog
use_backend Blog if is_blog
default_backend Apache-Cluster
bind-process 2
frontend HTTPS
bind 0.0.0.0:443 ssl crt /etc/ssl/crt.com.pem
reqadd X-Forwarded-Proto:\ https
bind-process 2
default_backend Apache-Cluster
acl is_blog url_beg /blog
use_backend Blog if is_blog
backend Apache-Cluster
redirect prefix http://www.example.com code 301 if { hdr(host) -i example.com }
#redirect scheme https if !{ ssl_fc }
redirect scheme https if !{ ssl_fc } { path_beg -i /order/checkout }
balance source
option httplog
option abortonclose
option httpclose
option forwardfor
cookie LOADBAL insert
option httpchk GET /working.php
server APPA 10.0.0.11:80 cookie APPA weight 50 maxconn 40000 check inter 4000 rise 2 fall 2
server APPB 10.0.0.12:80 cookie APPB weight 50 maxconn 40000 check inter 4000 rise 2 fall 2
backend Blog
mode http
option httplog
option abortonclose
option httpclose
option forwardfor
cookie LOADBAL insert
server BLOG 10.0.0.12:80 cookie BLOG weight 50 maxconn 512