On Failure & Rollback

In JFrog Connect, you can define what will happen if a step fails, and if a rollback is needed, how the rollback will take place. This ensures that your devices are left in a stable and known state.

Control and Stability

If a failure occurs during a software update, it is important to return your device to a stable and known state. You can have full control over the process by configuring On Failure instructions in each step in the flow and defining per step the rollback actions to be taken.

For each step, you can decide what kind of rollback is triggered (if any), and trigger actions such as a command or a custom script to run. When an update deployment ends, either successfully or due to a failure in the update flow, the outcome will be that your device is left in a stable state, and you will have full knowledge of that state.

On Failure & Rollback Configurations

You can specify the rollback behavior at two levels:

• Flow Level: Use Rollback on Failure to configure a rollback of the entire update flow.

• Step Level: In the On Failure configuration of each step, you can define rollback instructions that relate specifically to that step. If no step-level instructions are configured, the flow-level rollback will run if a failure occurs.

Free Memory Requirement

In order to enable rollback functionality, your device's free memory must be greater than the update size. To calculate the amount of free memory required, use the relationship below:

Free Memory Required > (Current Folder Size + Update Size + 10MB)

What's Next?

Learn more about the on failure and rollback configurations in Rollback on Failure and On Failure Policy for Step.