Skip to main content
A Delta Migration uses an initial data load, followed by , to . In this approach, you migrate most application data to the target using before stopping application traffic to the source database. You then use to keep the target database in sync with any changes in the source database (the migration delta), before finally halting traffic to the source and cutting over to the target after schema finalization and data verification.
  • All source data is migrated to the target .
  • This approach utilizes .
  • is supported, though this example will not use it. See for an example of a migration that uses failback replication.
This approach is best for production environments that need to minimize system downtime. This page describes an example scenario. While the commands provided can be copy-and-pasted, they may need to be altered or reconsidered to suit the needs of your specific environment. Delta migration flow

​
Example scenario

You have a small (300 GB) database that provides the data store for a web application. You want to migrate the entirety of this database to a new CockroachDB cluster. Business cannot accommodate a full maintenance window, but it can accommodate a brief (<60 second) halt in traffic. The application runs on a Kubernetes cluster. Estimated system downtime: 3-5 minutes.

​
Step-by-step walkthroughs

The following walkthroughs demonstrate how to use the MOLT tools to perform this migration for each supported source database:
For CockroachDB-to-CockroachDB migrations, contact your account team for guidance.

​
See also