Redshift clusters should have Multi-AZ deployments enabled
Confirms that Multi-AZ deployment is enabled for Redshift clusters. A Multi-AZ cluster runs its compute in two Availability Zones behind one endpoint, so it keeps processing workloads if one zone fails.
- service
- Redshift
- severity
- Medium
- scope
- evaluated in every region
How to fix it
- Affects another check: Redshift.16 passes a subnet group with two Availability Zones, but Multi-AZ needs three, so a group fixed only for that check still blocks this one.
- Can break clients: A Multi-AZ cluster cannot be paused or resumed, and monitoring queries on STL, SVCS, SVL, SVV or STV views stop working because only SYS_ views are supported.
In the console
Redshift → Provisioned clusters dashboard → Clusters → the cluster → Actions → Activate Multi-AZ
- Node type
- RG or RA3 with at least 2 nodes; resize a DC2 or single-node cluster first
- Cluster subnet group
- One with subnets in 3 or more Availability Zones
- Edit encryption
- On, if the cluster is unencrypted; Multi-AZ requires encryption
- Port
- Within 5431 to 5455 or 8191 to 8215
- Activate Multi-AZ
- Selected from Actions once the settings above are in place
With the AWS CLI
aws redshift modify-cluster --cluster-identifier <cluster-id> --multi-az --region <region>With Terraform
resource "aws_redshift_cluster" "this" {
cluster_identifier = var.cluster_identifier
node_type = var.node_type
multi_az = true
}
See which of your accounts this check is failing on.
Book a live scan