Release Note - July 30, 2026

Configurable Spot Instance Fallback and Retry Controls

General
This release delivers new Spot instance controls for task execution, giving you explicit control over On-Demand fallback behavior and Spot interruption retries. You can now choose whether a task falls back to On-Demand when Spot capacity can't be secured or an interruption retry limit is reached, or fails instead to keep costs predictable. You can also set how persistently a task retries on Spot after an interruption, to prioritize lower cost or reliable execution for each task.

New Features

• Configurable On-Demand fallback: A new "Fall back to On-Demand instance" toggle is available under Task → Execution Settings when Spot instances are enabled. The toggle is on by default, preserving the current behavior of falling back to On-Demand when Spot capacity can't be secured. Turn it off to run a task Spot-only and avoid unexpected On-Demand charges. When fallback is off and Spot capacity can't be secured, the task reports failure immediately instead of waiting for the next retry interval.
• Configurable Spot interruption retries: A new "Spot Instance Interruption retry" setting lets you decide how many times a task retries on a new Spot instance if the cloud provider reclaims the running instance mid-execution. Retries stay bounded by the limit set, and once the limit is reached, the fallback setting decides whether the task moves to On-Demand or fails.

Your fallback and retry settings are preserved when you copy or rerun a task or copy a project.

Resources
If you have questions or experience issues with Spot instance behavior, please contact your Velsera Seven Bridges representative or email [email protected]. Learn more about Spot Instances.