Comprehensive Guide to Installing, Configuring, and Using HAProxy for L7/L4 Load Balancing and High Availability

What is HAProxy

HAProxy is a free, open‑source load‑balancing software that runs on most mainstream Linux distributions, offering both L4 (TCP) and L7 (HTTP) balancing with a rich set of features and performance comparable to commercial solutions.

Core Capabilities and Key Features

Load‑balancing algorithms: round‑robin, static‑RR, least‑connection, IP/URI/HEADER hash, etc.

Health checks: TCP and HTTP modes.

Session persistence via cookies.

SSL termination and HTTP request rewriting.

Web‑based statistics page for monitoring.

Performance highlights include a single‑threaded, event‑driven, non‑blocking architecture that can handle hundreds of requests per millisecond with only a few kilobytes of memory per session, and reported throughput of over 100 k requests/second.

Installation and Operation on CentOS 7

Create a dedicated user/group (e.g., ha ), download the source, extract, compile, and install:

wget http://www.haproxy.org/download/1.7/src/haproxy-1.7.2.tar.gz
tar -xzf haproxy-1.7.2.tar.gz
make PREFIX=/home/ha/haproxy TARGET=linux2628
make install PREFIX=/home/ha/haproxy

Adjust sysctl parameters for optimal performance, for example:

net.ipv4.tcp_tw_reuse = 1
net.ipv4.ip_local_port_range = 1024 65023
net.ipv4.tcp_max_syn_backlog = 10240
net.core.somaxconn = 10000

Register HAProxy as a system service using an init script placed in /etc/init.d/haproxy and control it with service haproxy start|stop|restart .

Adding Logging via rsyslog

Configure HAProxy to send logs to rsyslog (local0 for info, local1 for warnings) and create /etc/rsyslog.d/haproxy.conf :

$ModLoad imudp
$UDPServerRun 514
local0.* /var/log/haproxy.log
local1.* /var/log/haproxy_warn.log

Restart both services to apply the changes.

Building an L7 Load Balancer

Define a simple configuration that balances traffic based on URI prefixes, enables session persistence with cookies, and provides a statistics page:

global
    daemon
    maxconn 30000
    user ha
    pidfile /home/ha/haproxy/conf/haproxy.pid
    log 127.0.0.1 local0 info
    log 127.0.0.1 local1 warning

defaults
    mode http
    log global
    option http-keep-alive
    option forwardfor
    option httplog
    timeout connect 5000ms
    timeout client 10000ms
    timeout server 50000ms
    option httpchk GET /healthCheck.html

frontend http-in
    bind *:9001
    acl url_ms1 path_beg -i /ms1/
    acl url_ms2 path_beg -i /ms2/
    use_backend ms1 if url_ms1
    use_backend ms2 if url_ms2
    default_backend default_servers

backend ms1
    balance roundrobin
    cookie HA_STICKY_ms1 insert indirect nocache
    server ms1.srv1 192.168.8.111:8080 cookie ms1.srv1 maxconn 300 check
    server ms1.srv2 192.168.8.112:8080 cookie ms1.srv2 maxconn 300 check

backend ms2
    balance roundrobin
    cookie HA_STICKY_ms2 insert indirect nocache
    server ms2.srv1 192.168.8.111:8081 cookie ms2.srv1 maxconn 300 check
    server ms2.srv2 192.168.8.112:8081 cookie ms2.srv2 maxconn 300 check

backend default_servers
    balance roundrobin
    cookie HA_STICKY_def insert indirect nocache
    server def.srv1 192.168.8.111:8082 cookie def.srv1 maxconn 300 check
    server def.srv2 192.168.8.112:8082 cookie def.srv2 maxconn 300 check

listen stats
    bind *:1080
    stats uri /stats
    stats refresh 30s
    stats realm HAProxy\ Stats
    stats auth admin:admin

After reloading HAProxy, you can test the setup by accessing http:// host :9001/ms1/demo.html , /ms2/ , and /def/ and observing the cookies that enforce persistence.

Building an L4 Load Balancer

Switch the mode to tcp and use a simple round‑robin or source‑based balancing. Cookie‑based persistence is unavailable, but IP‑based persistence can be achieved by changing balance roundrobin to balance source .

Key Configuration Details

The HAProxy configuration file is divided into five sections: global , defaults , frontend , backend , and listen . Each section controls specific aspects such as process limits, logging, ACLs, binding ports, health checks, timeouts, and load‑balancing algorithms.

Using Keepalived for High Availability

Deploy two HAProxy instances on separate hosts and run Keepalived on each. Keepalived manages a virtual IP (e.g., 192.168.8.201 ) and uses a VRRP script to monitor HAProxy health. The instance with the highest priority holds the virtual IP and acts as MASTER; if it fails, the BACKUP automatically takes over.

global_defs {
    router_id LVS_DEVEL
}

vrrp_script chk_haproxy {
    script "killall -0 haproxy"
    interval 2
    weight 2
}

vrrp_instance VI_1 {
    state MASTER
    interface enp0s25
    virtual_router_id 51
    priority 101
    advert_int 1
    virtual_ipaddress {
        192.168.8.201
    }
    track_script {
        chk_haproxy
    }
}

Start both Keepalived services, verify which node holds the virtual IP, and test failover by stopping HAProxy on the MASTER; the BACKUP will acquire the IP and continue serving traffic.

Overall, the guide provides a complete workflow—from understanding HAProxy’s capabilities, through installation, configuration for both L7 and L4 scenarios, logging, monitoring, to building a resilient HA setup with Keepalived.