Troubleshooting catalogs
Resolve common catalog connection, cross-account, and refresh failures in e6data.
When a catalog operation fails, the catalog row shows an Error status with a brief reason. Use the Retry icon for transient failures, Download log to capture details for support, and Refresh List to update statuses without re-fetching metadata. For specific error codes, see Catalog errors.
Connection failures
Test Connection fails immediately
Wrong host/port or unreachable metastore
Verify the connection details and that the metastore is reachable from your workspace
Authentication rejected
Bad credentials or expired token
Re-check credentials; for Unity/Databricks, confirm the access token hasn't expired
Network/firewall block
Metastore port not open to e6data
Allow the metastore port (for example, Hive Thrift 9083) through firewalls and NSGs
Insufficient permissions
The identity can't list schemas
Grant the identity read access to the metastore and underlying storage
Cross-account access
Role cannot be assumed (AWS)
Trust policy missing e6data's principal
Confirm the trust policy names the correct e6data account/role and includes sts:AssumeRole and sts:TagSession
Access denied to storage
The cross-account role lacks storage read
Grant read on the S3 bucket / ADLS Gen2 account
Wrong account/tenant values
Mismatched Target Account ID or Tenant/Client ID
Re-enter the cross-account values; for Azure, verify Tenant ID, Client ID, and Storage Account Name
See Cross-account access for the full patterns.
Schema and refresh issues
Tables missing after a metastore change
Catalog metadata is stale
Refresh the affected schemas - see Register and manage catalogs
Scheduled refresh shows Failed
Credentials expired or metastore unreachable
Check the refresh logs; the catalog keeps its previous metadata until the next successful run
Schema list empty after Test Connection
The identity can't see any databases
Verify the identity's grants on the metastore
See also
Last updated
Was this helpful?