Installing and configuring the load balancer

The load balancer is a component of the high-availability cluster of Integration Servers that monitors the health of the primary Integration Server and, if it becomes unavailable, switches over the SVMs and Light Agents to use the reserve Integration Server.

To install and configure a load balancer:

1. Install the solution to be used as a load balancer on a Linux device.
2. Open port 7271 on the load balancer device.
3. Configure the load balancer: specify the settings for connecting to the primary and reserve Integration Servers and configure request forwarding.

As the load balancer, you can use a solution that meets the following requirements:

• The solution can be used as part of a high-availability cluster of two servers in an active/passive configuration.
• The solution supports the following scheme:
  • If the primary server is available, the load balancer redirects all requests to the primary server.
  • If the primary server becomes unavailable, the load balancer detects this and begins to redirect requests to the reserve server.
  • After the availability of the primary server is restored, the load balancer detects this and begins to redirect requests to the primary server.

For example, HAProxy meets these requirements. The rest of this section describes the procedures for installing and configuring the load balancer using HAProxy 2.8.5 as an example.

Installing the HAProxy solution

You need to install the solution on a Linux device. Installation requires root access.

To install HAProxy, run one of the following commands depending on your operating system:

• On CentOS, Red Hat Enterprise Linux, or RED OS:

  yum install haproxy

• On Ubuntu or Debian:

  apt-get install haproxy

Configuring the HAProxy solution

To configure HAProxy for use as a load balancer:

1. Open the /etc/haproxy/haproxy.cfg configuration file for editing.
2. Add the following lines and save the file:

   frontend VIIS

   bind 0.0.0.0:7271

   mode tcp

   log global

   use_backend VIIS

   backend VIIS

   balance first

   mode tcp

   option ssl-hello-chk

   server VIIS1 {address1}:{port1} check on-marked-up shutdown-backup-sessions

   server VIIS2 {address2}:{port2} check backup

   where:

   • {address1}:{port1} is the address and port of the primary Integration Server.
   • {address2}:{port2} is the address and port of the reserve Integration Server.
3. Restart the haproxy service by running the following command:

   service haproxy restart

Page top