Hi @muaaz,

On GCP Databricks, the SQL pattern you are using is fine, but the recommended best practice is to back it with a Unity Catalog external location instead of pointing tables directly at arbitrary gs:// paths.

In practice, that means first creating a storage credential and an external location over the relevant GCS prefix (for example, gs://bucket/tables), granting CREATE EXTERNAL TABLE on that external location, and then registering each table with CREATE TABLE catalogue.schema.table USING DELTA LOCATION 'gs://bucket/tables/UUID_or_table_path' where the path sits under that external location and is dedicated to a single table.

Databricks generally recommends using Unity Catalog managed tables for long-term workloads, so a common migration pattern is to initially register existing Delta data as external tables for a fast cutover, and then, once things are stable, convert those external tables to managed tables so that Unity Catalog controls the storage location, lifecycle, and automatic optimisations going forward.

If this answer resolves your question, could you mark it as “Accept as Solution”? That helps other users quickly find the correct fix.