Today we are announcing a new capability in Elastigroup for AWS that allows defining the termination policy of an Elastigroup. Setting a termination policy allows the user to define which instances will be terminated whenever scale-down activities are scheduled for an Elastigroup cluster.
Up until now, the only applicable termination policy was empowered by Spotinst’s algorithm, which scheduled instance scale-down based on a scoring calculation that took into consideration several criteria such as instance healthiness, spot availability, cluster strategy and more.
With this new capability, the user is able to select if the termination policy is the default, or the ‘Newest first’, which first scales down the latest instances.
The termination policy configuration is available for Elastigroup AWS, and is accessible via the Scaling section of the cluster configuration.
More information is available in our documentation.