Hi everyone 👋 I am working with Airbyte (MongoDB → ClickHouse) and automating schema refresh using API. I noticed that even when schema changes are detected (non_breaking), Airbyte does not automatically apply them — I still need to call the update connection API manually. My questions: 1. Why does Airbyte not auto-apply schema changes by default? 2. If I enable automatic schema propagation, can it break my pipeline in case of breaking changes? 3. Is there a recommended way to safely automate schema updates in production? Currently, I built a Python automation: Refresh → Detect → Update → Sync → Slack alert Just want to confirm if this is the correct production approach. Thanks!

You might want to check on connection page if you have something like this:

Detect and propagate schema changes

Thank you for you replay...but, I checked the connection settings under the Advanced Settings tab and verified the “Detect and propagate schema changes” option. However, in my current setup (Airbyte running locally via Docker), this feature is not behaving as expected. From what I understand, full support for automatic schema change handling may be more reliable in Airbyte Cloud. For now, since I am using a local deployment, I am handling schema detection and decision logic externally using Airflow to ensure proper control and automation. Let me know if I might be missing something in the local configuration......

IIRC they changed it so that schema discovery doesn't run before every sync anymore, only once a window has expired (I think in OSS this might be 24 hours, on Cloud it's shorter, maybe 15 minutes). You can force it to run every time by setting

DISCOVER_REFRESH_WINDOW_MINUTES

to

0

in the env vars for the

worker

(or customize it to suit your needs). (meaning if you do this, you may not have to handle the workflow yourself)