About Highly Available Gateways
The Aviatrix High Availability feature enabled deployment of multiple gateway instances within a VPC or VNet to support high availability and scalability, aiming to minimize network downtime and improve overall network stability and performance. In this setup, each gateway consists of multiple gateway instances operating in Active-Active mode. Network traffic flow is load balanced across all available gateway instances within that VPC or VNet. The Aviatrix Controller automatically updates CSP route tables to distribute traffic among all active gateway instances.
Creating Highly Available Gateway Instances
Aviatrix Platform supports up to 15 highly available instances for Transit and Spoke Gateways. At least two instances are required for highly availability. The gateway instances are configured in Active-Active mode by default. A Transit Gateway is limited to two gateway instances in the following scenariios:- When deployed with the FireNet function
- When peered to AWS TGW
- When external connections are configured
- When BGP external connections are present
- When Site2Cloud external connections are configured, or customized SNAT and DNAT are enabled.

<gateway-name>-<instance-number>.

Creating Highly Available Spoke Gateway Instance Using Terraform
When deploying the Aviatrix Platform with Terraform, a highly available Spoke Gateway instance can be created using the Terraform resource aviatrix_spoke_ha_gateway.