High Availability

Use a high availability configuration to increase resilience against unexpected operating system failures or network outages. The picture below shows the high availability architecture.

High availability – architecture

A high availability configuration has the following parts:

• The primary server. This is the active server by default and controls the configuration. You can set up new channels and triggers here.
• The secondary server. It is not possible to change the configuration from this server. By default, it only monitors the primary server.
• The database. The database must be a SQL Server database that is part of a SQL Server HA group or a Microsoft Azure database. Both primary and secondary servers access this database.
• The load balancer (optional). This can be any TCP load balancer such as HAProxy or Nginx. You need a load balancer when using server channels such as the HTTP API or the SMPP Server channel. If you are running client configurations on different PCs you also need a UDP load balancer. Nginx supports both TCP and UDP load balancing.

Set up the primary server

Use the Configuration Wizard to set up or change your configuration type. The Configuration Wizard runs automatically after setup, but you can also start it manually from the Start menu or from the Manager (File → Configuration Wizard). To configure your server as the primary server select Primary Server (HA) as your configuration type.

Next enable high availability in the Server options page. You can find this setting in the High Availability tab. Open Server options from the Configuration → Options menu.

Note: Make sure to restart the service after enabling high availability.

Make sure to configure a notification email, phone number, and or URL. These are used when the primary server goes offline or is restored.

Set up a secondary server

To set up a secondary server use the Configuration Wizard and select Secondary Server (HA) as the configuration type.

It is not possible to modify the configuration from the secondary server. The secondary server only monitors the primary server.

If the primary server goes down, the secondary server automatically takes over all tasks. When the primary server is restored, the secondary server stops processing tasks and resumes monitoring.

If it is not possible to restore the primary server, you can promote the secondary server to a primary server or to a standalone server using the Configuration Wizard. This allows you to modify the configuration again.

Primary and secondary server status

You can view the server status from both the primary and secondary server. Open the service control view from the Manager by clicking File → Service start / stop.

Service control dialog – HA enabled

If both servers are online, both show green status. If there is an issue with the primary server, you can log into the secondary server to view the current status in this dialog as well.