Registering a catalog connects e6data to a metastore so it can discover and query your tables. The Create flow is the same across catalog types - only the connection fields differ. After registration, you refresh, edit, or delete the catalog from the Catalogs page.
Before you start
- Open your workspace’s Compute Plane.
- Have your metastore connection details ready (host, port, credentials).
- For Serverless deployments, complete the cloud-specific storage access setup so e6data can read the underlying data - see Deployment.
Create a catalog
- From the Compute Plane, go to Catalogs in the left navigation and click Create Catalog.
- Enter a Catalog Name - unique within the workspace, up to 253 characters. Mixed case, letters, numbers, hyphens, and dots are allowed; spaces and underscores are accepted but normalized to hyphens in the name e6data stores internally, while dots are preserved (for example,
My Glue Catalog becomes my-glue-catalog). See resource naming conventions for the full rules.
- Select a Catalog Type: Hive Metastore, AWS Glue, Databricks Unity, Microsoft Fabric, Apache Polaris, Iceberg REST, or AWS S3 Tables.
- (Optional) Enable Governance to apply fine-grained access control via Ranger, OPA, Unity, or Lake Formation.
- Fill in the connection details for your catalog type - see the per-catalog guides, such as AWS Glue and Hive Metastore.
- (Optional) Enable Cross-Account Access if the storage or metastore is in a different cloud account - see Cross-account access.
- Click Test Connection to verify connectivity.
- After a successful test, specify the schemas to include - pick the databases you want this catalog to expose.
- Click Create Catalog. The catalog appears in the catalogs list and is available to attach to clusters.
Test Connection
Test Connection validates network reachability to the metastore, that authentication credentials are accepted, and that e6data can list schemas. A successful test is required before you can select schemas. If it fails, the error explains the cause - typically a credentials issue, a network/firewall block, or insufficient permissions.
Schema selection
After a successful test, select only the schemas (databases) you want this catalog to expose - all schemas for an analytics-team catalog, a single schema for a tenant-specific catalog, or a curated subset. You can change the selection later by editing the catalog.
Refresh a catalog
When tables or schemas change in your metastore, refresh the catalog to pick up the changes:
- Open Catalogs, find the catalog, and click the Refresh icon.
- Choose Refresh all schemas or Refresh specific schemas, then confirm.
Refresh runs in the background; the catalog status shows progress. Common triggers: new tables added, tables dropped, columns added/dropped, or bulk changes after a migration. For automated periodic refresh, use scheduled refresh.
Edit a catalog
- Open Catalogs and click the row actions (⋮) for the catalog.
- Click Edit, change the Select Schemas dropdown (or other fields), and click Update Catalog.
Changing fields that affect connectivity requires a successful re-test before saving.
Delete a catalog
- Open Catalogs, click the row actions, click Delete, and confirm.
Deletion removes the catalog from this workspace; attached clusters lose access. The underlying metastore and data are not modified.
Scheduled refresh
Scheduled refresh keeps catalog metadata in sync with the metastore on a schedule, without manual intervention. Each run reconnects with the catalog’s existing credentials, re-discovers schemas/tables/columns within scope, updates e6data’s metadata cache, and logs the result. Refreshes run in the background and update metadata atomically - they don’t interrupt queries.
Enable it when creating or editing a catalog:
-
Toggle Scheduled refresh on.
-
Pick a schedule:
-
Optionally scope the refresh to specific schemas rather than the full catalog (lighter and more targeted for large catalogs).
-
Save the catalog.
If a scheduled refresh fails, the catalog stays usable with its previous metadata snapshot, the error is recorded in the refresh history, and e6data retries on the next interval (predictable cadence, no backoff). Investigate persistent failures (expired credentials, unreachable metastore, changed permissions) using the refresh logs - see Catalog errors.
Error handling
If a catalog operation fails, the catalog row shows an Error status. Use the Retry icon for transient failures, the Download log icon to capture request/response/error details for support, and the Refresh List button to update statuses without re-fetching metadata.
See also
Last modified on October 1, 2026
