If a disaster impacts the region or the availability zone that hosts a serverless runtime environment, redirect jobs to a temporary serverless runtime environment in a stable region or availability zone as part of your organization's disaster recovery plan.
Disaster recovery procedure
During a disaster, all virtual machines in the serverless runtime environment shut down and jobs can no longer run in the environment.
To minimize data loss and downtime, complete the following tasks:
Create a temporary serverless runtime environment in a stable region or availability zone.
Make sure that the connections used in jobs are available in the stable region or availability zone.
Clean up data related to incomplete job runs. If data was partially loaded to a target, manually delete the data or update the mapping to truncate the target before writing new rows.
Redirect jobs to the temporary environment.
Restoring the primary environment
When the region or availability zone that hosts the primary serverless runtime environment has recovered, you can restore the primary environment.
To restore the primary environment, complete the following tasks:
Clean up the ENIs that were created in your AWS account for the primary environment.