Troubleshooting Snowflake Data Clean Rooms¶

Installation issues¶

• See the installation troubleshooting section.

• Confirm that you have updated your network policy to allow the UI to access your data.

Analysis issues¶

Error:

Failure during expansion of shared view <CLEAN ROOM VIEW NAME> as view owner: Insufficient permission to resolve external/iceberg table <TABLE_NAME> shared by application SAMOOHA_CLEANROOM_APP_<CLEAN ROOM ID>

Cause:

You are trying to access external or Iceberg tables, but External and Iceberg tables are not enabled in both the provider’s and consumer’s accounts.

Solution:

Ensure that both provider and consumer accounts have enabled external and Iceberg tables.

Error:

SQL compilation error: Failure during expansion of shared view '<CLEAN ROOM VIEW NAME>' as view owner: Object '<some object name>' does not exist or not authorized.

Cause:

Clean room grants no longer exist on the dataset that you are trying to access. Most likely this is because the source object has been renamed or replaced.

Solution:

• If the table was renamed, change the name back to what is linked in the clean room. You might need to re-register the object as well.

• If the table was recreated, register the object again in the clean room.

Error:

Query returns zero results and you think that’s wrong**

Possible causes and solutions:

• Confirm that neither side has a masking policy on the data that could prevent it from being joined or shown.

• Confirm that the join columns are formatted the same way.

• Confirm that you are not falling below any fixed threshold settings. The audience overlap has a default threshold of 5, meaning that fewer than 5 rows will be omitted from the results. Ask the provider what the threshold is, and confirm whether you have any overlaps greater than that number; temporarily modify the overlap specifications to guarantee large segment groups to see whether you get results then.

Cross-cloud issues¶

Error:

Analysis Execution Failure: 'SnowparkSQLException' due to Database Listing Conflict in a single-account testing clean room

Cause:

Cross-Cloud Auto-Fulfillment is not supported with single-account testing clean rooms.

Solution:

Disable cross-cloud auto-fulfillment in this clean rooms account by calling library.disable_laf_on_account during testing, or don’t try to make cross-cloud procedure calls in this clean room.

Cloud data connector issues¶

If you are having problems with an external data connector for AWS, Azure, or Google Cloud Storage, see Snowflake Data Clean Rooms: Troubleshooting external data connectors.

Request log issues¶

Error:

**Failure**: Request logs unable to be mounted. Try again.

Cause:

Mounting request logs succeeded for internal clean room, then fails for external clean room in same account. Internal clean rooms have fewer requirements than external clean rooms. Your installation met the requirements for using internal clean rooms, but not for external clean rooms.

Solution:

Confirm that your email was validated by clean rooms, and that you fulfill all the clean rooms account requirements.

Data access issues¶

You can get an error message that reports the inability to access a data source during several points in the usage flow:

If the error occurred during the registration process:

• You misspelled the table name or path, if using the API.

• If this is an external or Iceberg table, confirm that you have fulfilled the requirements and procedure for registering an external or Iceberg table.

• Confirm that your current role has OWNERSHIP privilege on the object being registered.

If the error occurred during the linking process:

• In the API, you might be using the wrong role.

• The object might not have been registered. In the API if you try to link an object that isn’t registered, you will see an error. In the UI, you should see only objects that have been registered as available for linking.

• Confirm that SAMOOHA_APP_ROLE has USAGE and SELECT privileges on your object.

• The table might have been moved, renamed, or had its permissions (or any Snowflake policy permissions) changed since registration. When this happens, you might also see the error SQL access control error: Insufficient privileges to operate on table...

If the error occurred after data was successfully registered and linked:

If using the API, confirm that you spelled the fully qualified table name correctly.

Error:

Object '<some_object_name>' does not exist or not authorized

Cause:

The source table might have been moved, renamed, or had its permissions (or permissions on a policy or ancestor object that it depends on) changed.

Solution:

Try re-registering and re-linking the object in your account, or move the object back to the old location, or revert any additional permissions added.

Error:

Insufficient permission to resolve external/iceberg table

Cause:

If an external or Iceberg table is involved in your query, then the table was not registered properly.

Solution:

See Enabling external and Apache Iceberg™ tables to ensure that you fulfill the requirements and procedures to use these table types. You can sometimes resolve this by explicitly granting SELECT on the table to SAMOOHA_BY_SNOWFLAKE.

Error:

not approved:unauthorized columns used error as a result of run analysis

Cause:

You are joining or projecting a collaborator’s column against the collaborator’s join or column policy.

Solution:

View the join and column policies set by your collaborator by calling consumer.view_provider_column_policy and consumer.view_provider_join_policy.

CALL samooha_by_snowflake_local_db.consumer.view_provider_join_policy($cleanroom_name);
CALL samooha_by_snowflake_local_db.consumer.view_provider_column_policy($cleanroom_name);

Copy

You might have also exhausted your privacy budget:

CALL samooha_by_snowflake_local_db.consumer.view_remaining_privacy_budget($cleanroom_name);

Copy

External and Iceberg table issues¶

Error:

Failure during expansion of shared view <CLEAN ROOM VIEW NAME> as view owner: Insufficient permission to resolve external/iceberg table <TABLE_NAME> shared by application SAMOOHA_CLEANROOM_APP_<CLEAN ROOM ID>

Cause:

External or Iceberg tables are not enabled in both the provider and consumer accounts.

Solution:

Ensure that both provider and consumer accounts have enabled external and Iceberg tables.

Error:

Consumer gets Invalid restricted feature 'external_data' when linking in an external or Iceberg table.

Cause:

The provider has not yet enabled external and Iceberg tables.

Solution:

The provider must complete the process of enabling external and Iceberg tables for their account. If they are doing this in code, the provider must check the security scan results and, if successful, must update the default release version.

Error:

Insufficient permission to resolve external/iceberg table error when running an analysis involving an external or Iceberg table.

Cause:

The table was probably not registered properly by both the provider and consumer.

Solution:

Read the external and Iceberg table registration information and be sure to follow all instructions on both the provider and consumer side.