Hey everyone, this is Heitor from Collact.
We recently added a new schema to our database which serves as caching for our API. The mechanism creates thousands of tables a day and drops them some time later. This means that if Metabase scans this schema, every day thousands of new tables and columns are added to metabase_fields. That will slowdown queries such as autocomplete ones (which I have already openned a topic and discussed with Flamber hehe).
So, is that a way to make Metabase skip scanning a specific schema?
Thank you very much!
Yes, create a database user that doesn't have access to that schema and use that user credentials in the Metabase connection.
We're working on adding partial schema sync, which was already added to BigQuery in 0.42.
I think I can remove schema READ access, but pg_objects (we are on postgres) , pg_columns and information_schema tables will still be visible. Won't that be a problem and Metabase will continue to index?
@heits There are a lot of information about that on the internet, so try searching, since it's not really anything to do with Metabase.