Amazon Redshift clusters should have automatic snapshots enabled

How to fix it

Set the cluster's automated snapshot retention period to seven days or more, so a bad load or an accidental delete has a recovery point. A retention period of 0 means automated snapshots are off and fails for the same reason a short window does. The change applies immediately and does not restart the cluster.

In the console

Redshift → Clusters → the cluster → Edit → Backup → Snapshot retention

With the AWS CLI

aws redshift modify-cluster --cluster-identifier <cluster-id> --automated-snapshot-retention-period 7 --region <region>

With Terraform

resource "aws_redshift_cluster" "this" {
  cluster_identifier                  = var.cluster_identifier
  node_type                           = var.node_type
  automated_snapshot_retention_period = 7
}