By using AWS re:Post, you agree to the Terms of Use

DMS CDC to Redshift during cluster resize


A customer has a 28-node dc2.large RS cluster and a CDC DMS process continuously replicating data from an on-prem database. They plan to migrate the RS cluster to a 3-node ra3.4xlarge cluster and are worried about any possible failure on the CDC process during the migration.

The process they will follow is:

  1. Elastic resize the cluster from 28 dc2.large nodes to 7 ra3.4xlarge nodes.
  2. Classic resize the cluster from 7 ra3.4xlarge nodes to 3 ra3.4xlarge nodes.

This way they will keep the same cluster endpoint, so theoretically nothing should change regarding DMS configuration. Do you see any potential risk for the replication process during the migration?

Thank you in advance!

1 Answers
Accepted Answer

Is there a reason you aren't using Elastic re-size for both operations? Our docs say you can use Elastic re-size to change node types if the new/old node types are supported so it would be worth double checking that.

Whilst CDC will work after you've done the migration as you aren't changing DB configuration, during the re-size you won't be able to update the DB on the Redshift cluster, as per our docs "During the resize operation, the cluster is read-only"

For elastic re-size, this would be 10-15 minutes, but for classic resize this could be 2 hours - 2 days and it's something you'd want to prepare your customer for.

answered 2 years ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions