[Feature request] DMS CDC - Improve handling of schema changing on source database

Question

On the case, but not limited to this origin:
Origin: Aurora-postgres
Destination: Redshift
When a varchar/nvarchar length changes in the source table, it does not get applied to Redshift; it appears on DMS log as a change incoming from the database, but it's dropped, making that table fail when data requiring that field length arrives.
I know that Redshift could not change several column types without recreating the tables, but it can execute a command like this without any issue
```
ALTER TABLE table_name ALTER COLUMN column_name TYPE VARCHAR(new_size);
```
This will greatly improve the experience with these kinds of simple field changes.
Thank you

re:Post Agent · Answer

Thank you for this feature request regarding AWS Database Migration Service (DMS) Change Data Capture (CDC) handling of schema changes.

I understand your concern about VARCHAR/NVARCHAR length changes in Aurora PostgreSQL not being automatically applied to Amazon Redshift during CDC replication. You've correctly identified that while DMS detects these schema changes in the logs, it doesn't propagate the ALTER COLUMN TYPE commands to the target, which can cause replication failures when data exceeding the original column length arrives.

Your suggestion to have DMS automatically execute ALTER TABLE commands for compatible schema changes like VARCHAR length increases is a valuable enhancement that would improve the CDC experience and reduce manual intervention.

For feature requests like this, I recommend submitting your feedback through the official AWS channels:

1. **AWS Support** - If you have an AWS Support plan, you can submit feature requests directly through your support case
2. **AWS re:Post** - Share your use case and feature request with the community and AWS service teams
3. **Your AWS account team** - If you have a Technical Account Manager (TAM) or Solutions Architect, they can help escalate product feedback

When submitting, include your specific use case, the business impact, and technical details you've provided here. Product teams prioritize features based on customer demand and use cases, so detailed feedback like yours is valuable for influencing the DMS roadmap.

In the meantime, you may need to implement monitoring for schema change events in DMS logs and handle these ALTER statements manually or through automation scripts as a workaround.

[Feature request] DMS CDC - Improve handling of schema changing on source database

Add your answer

Relevant content