HAProxy: Équilibrage de charge HTTP 

 
Ce  document  montre  l’installation  d’un  cluster  HAProxy  afin  de  configurer  l'équilibrage 
de charge. Cet exemple basé sur l'environnement comme suit. 
 

 
 
1. HAProxy 
 
 
# systemctl stop firewalld 
# firewall-cmd --reload 
# yum -y install haproxy 
 
 
2. Configurer HAProxy 
 
 
# mv /etc/haproxy/haproxy.cfg /etc/haproxy/haproxy.cfg.org 
# vi /etc/haproxy/haproxy.cfg 
 
global 
log 127.0.0.1 local2 info
chroot /var/lib/haproxy
pidfile /var/run/haproxy.pid

# max per-process number of connections 

maxconn 256

# process' user and group 

user haproxy
group haproxy
 # makes the process fork into background 
daemon
defaults

# running mode 
mode http

# use global settings 

log global

# get HTTP request log 

option httplog

# timeout if backends do not reply 

timeout connect 10s

# timeout on client side 

timeout client 30s

# timeout on server side 

timeout server 30s

# define frontend ( set any name for "http-in" section ) 

 
frontend http-in
# listen 80 
bind *:80

# set default backend 

default_backend backend_servers 

# define backend 
backend backend_servers

# balance with roundrobin 

balance roundrobin

​# define backend servers 

server www01 IP:80 check
server www02 IP:80 check

 
 
3. Configurer Rsyslog pour log HAProxy 
 
 
# vi /etc/rsyslog.conf 
 
 # ligne 15,16: décommenter, ligne 17: ajouter 
$ModLoad imudp 
$UDPServerRun 514 
$AllowedSender UDP, 127.0.0.1 
 
# ligne 54: changer, ligne 55 ajouter 
*.info;mail.none;authpriv.none;cron.none,​local2.none ​ /var/log/messages 
local2.* /var/log/haproxy.log 
 
 
# systemctl restart rsyslog 
# systemctl restart httpd 
 
 
4. Test 
 

 
 

 
 
 
 
NB : Toutes les commandes sont effectués sur les 02 serveurs. 
 IP VIRTUEL AVEC KEEPALIVED 
 
1. Installer et configurer Keepalived 
 
Sur le serveur 01 
 
# yum install keepalived -y 
# vi /etc/keepalived/keepalived.conf 
 
vrrp_script chk_haproxy { 
script "killall -0 haproxy" # check the haproxy process 
interval 2 # every 2 seconds 
weight 2 # add 2 points if OK 
} 
vrrp_instance VI_1 { 
interface eth0 # interface to monitor 
state MASTER # MASTER on haproxy1, BACKUP on haproxy2 
virtual_router_id 51 
priority 101 # 101 on haproxy1, 100 on haproxy2 
virtual_ipaddress { 
0.0.0.0/24 # virtual ip address 
} 
track_script { 
chk_haproxy 
} 
} 
 
 
Sur le serveur 02 
 
# yum install keepalived -y 
# vi /etc/keepalived/keepalived.conf 
 
vrrp_script chk_haproxy { 
script "killall -0 haproxy" # check the haproxy process 
interval 2 # every 2 seconds 
weight 2 # add 2 points if OK 
} 
vrrp_instance VI_1 { 
interface eth0 # interface to monitor 
state BACKUP # MASTER on haproxy1, BACKUP on haproxy2 
virtual_router_id 51 
priority 100 # 101 on haproxy1, 100 on haproxy2 
virtual_ipaddress { 
0.0.0.0/24 # virtual ip address 
} 
track_script { 
 chk_haproxy 
} 
} 
 
 
2. Configurer le transfert IP 
 
Sur les 02 serveurs 
 
# vi /etc/sysctl.conf 
 
net.ipv4.ip_forward = 1 
net.ipv4.ip_nonlocal_bind = 1 
 
 
3. Redémarrage des services 
 
 
# systemctl enable keepalived 
# systemctl start keepalived 
# systemctl enable haproxy 
# systemctl start haproxy 
 
 
 
 
 
WEBOGRAPHIE​ :  
 
https://www.server-world.info/en/note?os=CentOS_7&p=haproxy 
 
https://access.redhat.com/documentation/en-us/red_hat_cloudforms/4.6/html/hi
gh_availability_guide/configuring_haproxy