How do I move my provisioned Amazon Redshift cluster into a different subnet?

Short description

When a new Amazon Redshift cluster is launched, an Amazon Virtual Private Cloud (Amazon VPC), security group, and cluster subnet group are provided. During the launch, Amazon Redshift picks up random subnets from the associated cluster subnet group.

You have two options for moving your existing provisioned Amazon Redshift cluster into a different subnet: You can either use an existing subnet group, or you can create a new subnet group that contains the subnet where you will move the provisioned Amazon Redshift cluster. After you have a subnet group with the preferred subnet, you must restore your cluster from a snapshot. Be sure that there is only one subnet in the subnet group. Otherwise, Amazon Redshift might not launch the cluster in that specific subnet.

Resolution

To move your Amazon Redshift cluster into a different subnet group, perform these steps:

1.    Create a manual snapshot of the cluster.

2.    Restore the cluster using the snapshot that you just created.

3.    Under Additional configurations, turn off the Use defaults settings.

4.    Choose Network and security.

5.    Select the desired subnet group.

6.    Choose Restore cluster from snapshot.

For more information, see How do I move my Amazon Redshift cluster from one VPC to another VPC?

To confirm the subnet and subnet group that your Amazon Redshift cluster is using, perform these steps:

1.    Open the Amazon Redshift console.

2.    Choose the Clusters tab.

3.    Select your Amazon Redshift cluster.

4.    Choose the Properties tab.

5.    Navigate to Network and security to view network details. The Network and security information shows you the subnet group that's being used.

6.    Under the Properties tab, navigate to Node IP addresses. This section has information that can help you verify the subnet that's being used. The subnet listed must be from your subnet group.

Subnet best practices

When you move your Amazon Redshift cluster into a different subnet, consider the following best practices:

• Configure your cluster subnet group so that the subnets reside in a homogenous network. In a homogenous network configuration, all the subnets that belong to that group share similar networking rules, performing as a uniform network. A homogenous network configuration also minimizes any connectivity issues that can occur when moving your Amazon Redshift cluster into a different subnet.
• Be sure that there are enough IP addresses present in each of your subnets within the subnet group. Otherwise, Amazon Redshift provisions your cluster into two or more subnets within the same Availability Zone (AZ). If there are no available subnets within the same Availability Zone, or there aren't enough IP addresses present, your cluster launch might fail. To resolve a launch failure, delete any unused elastic network interfaces. Or, you can move your AWS resources into a different subnet by choosing a different subnet group.