Master HAProxy: From Installation to High‑Availability Load Balancing

What is HAProxy

HAProxy is a free load‑balancing software that runs on most Linux distributions. It provides both L4 (TCP) and L7 (HTTP) balancing with a rich set of features.

Core Functions

Load balancing: L4 and L7 modes, many algorithms (RR, static‑RR, LC, IP‑Hash, URI‑Hash, URL_PARAM Hash, HTTP_HEADER Hash, etc.)

Health checking: supports TCP and HTTP health checks

Session persistence: Insert/Rewrite/Prefix cookies and various hash methods

SSL termination: HAProxy can decrypt HTTPS and forward plain HTTP to back‑ends

HTTP request rewriting and redirection

Monitoring and statistics: built‑in web stats page, can be scraped by custom monitors

Key Features

Performance

Single‑threaded, event‑driven, non‑blocking architecture processes hundreds of requests in <1 ms and uses only a few KB per session.

O(1) event checker, zero‑copy forwarding and other kernel‑level optimisations keep CPU usage low.

In tests a single HAProxy process handled over 100 k requests / s and saturated a 10 Gbps link.

Stability

HAProxy runs as a single process; the author reports no crash‑inducing bugs in 13 years of use. Stability depends on the underlying Linux kernel – a recent 2.6/3.x kernel with tuned sysctl parameters is recommended.

Use a Linux 2.6+ kernel.

Run HAProxy on a dedicated host.

Provide a standby node for hardware failures.

Initial sysctl tuning (example values):

net.ipv4.tcp_tw_reuse = 1
net.ipv4.ip_local_port_range = 1024 65023
net.ipv4.tcp_max_syn_backlog = 10240
net.ipv4.tcp_max_tw_buckets = 400000
net.ipv4.tcp_max_orphans = 60000
net.ipv4.tcp_synack_retries = 3
net.core.somaxconn = 10000

Installation on CentOS 7 (HAProxy 1.7.2)

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

Create a system user “ha”, then create /home/ha/haproxy/conf/haproxy.cfg with a minimal configuration:

global
    daemon
    maxconn 256
    pidfile /home/ha/haproxy/conf/haproxy.pid

defaults
    mode http
    timeout connect 5000ms
    timeout client 50000ms
    timeout server 50000ms

frontend http-in
    bind *:8080
    default_backend servers

backend servers
    server server1 127.0.0.1:8000 maxconn 32

Ensure ulimit -n is at least maxconn*2+18.

Running HAProxy as a Service

Create /etc/init.d/haproxy with start/stop/restart functions (script omitted for brevity) and enable it.

Logging with rsyslog

Add to the global and defaults sections:

log 127.0.0.1 local0 info
log 127.0.0.1 local1 warning

Configure /etc/rsyslog.d/haproxy.conf to write local0 to /var/log/haproxy.log and local1 to /var/log/haproxy_warn.log, then restart rsyslog.

Log Rotation

Create /root/logrotate/haproxy with daily rotation, keep 7 files, compress, and post‑rotate restart rsyslog.

L7 Load‑Balancing Example

Deploy six Nginx back‑ends (ms1, ms2, def groups) on two hosts, each serving a simple demo.html. Then use the following HAProxy configuration (excerpt):

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 auth admin:admin

Access http://<master_ip>:1080/stats to view real‑time metrics, test health checks, URI‑based routing, and session persistence.

L4 Mode

Switch mode tcp in the defaults section for pure TCP load balancing. Session persistence can be achieved with balance source or stick‑tables.

High Availability with Keepalived

Install Keepalived, configure a VRRP instance that tracks HAProxy health via killall -0 haproxy, and assign a virtual IP (e.g., 192.168.8.201). The master node holds the VIP; if it fails, the backup automatically takes over.

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 instances and verify that the virtual IP moves to the backup when the master HAProxy stops.