The April 2026 release includes changes in behavior for the following connectors.
Amazon Redshift V2 Connector
When you read from external tables, the string data type is mapped to varchar data type.
Previously, the columns of string data type were ignored.
MongoDB V2 Connector
You can configure an error threshold in the Stop on Error session property in a mapping task.
When you set an error threshold, the mapping task continues to run until the cumulative errors reach or exceed the threshold. As the mapping task runs in batch mode, the cumulative errors might exceed the threshold before the task fails.
Previously, the error threshold was not honored.
This changed behavior doesn't apply to mapping tasks that are based on mappings in advanced mode.
ODBC Connector
When you use the DB2 subtype in the connection and select the Truncate Target option in the Target properties of a mapping, the mapping runs the TRUNCATE TABLE query first. If the TRUNCATE TABLE query fails, the mapping runs the DELETE query.
Previously, the DELETE query was always run when the Truncate Target option was selected.
Salesforce Data 360 Connector
Salesforce Data Cloud Connector is renamed to Salesforce Data 360 Connector.
Snowflake Data Cloud Connector
The Password/PAT connection property for standard authentication is renamed to the Password connection property. If you want to use a programmatic access token generated in Snowflake, use the Programmatic Access Token authentication type to connect to Snowflake.
Previously, if your connection used standard authentication with a programmatic access token, the connection is not impacted.
However, Informatica recommends that you use authorization code, key pair, client credentials, or programmatic access token authentication to connect to Snowflake instead of using standard authentication.