Fivetran Release Notes

We have made the following improvements to the connector:

• Added table re-sync support. Previously, we only supported connection-level re-syncs.{% note %} Event-based tables and tables with a retention period at the source don't support this feature. {% /note %}
• You can now select and deselect tables to be synced on the Schema tab of the connection details page in your Fivetran dashboard.

We have made the following improvements to the connector:

• Added table re-sync support. Previously, we only supported connection-level re-syncs.{% note %} Event-based tables and tables with a retention period at the source don't support this feature. {% /note %}
• You can now select and deselect tables to be synced on the Schema tab of the connection details page in your Fivetran dashboard.

We have made the following improvements to the connector:

• Added table re-sync support. Previously, we only supported connection-level re-syncs.{% note %} Event-based tables and tables with a retention period at the source don't support this feature. {% /note %}
• You can now select and deselect tables to be synced on the Schema tab of the connection details page in your Fivetran dashboard.

We now capture and store HTTP request headers from webhook events in a new headers column. This enhancement provides better observability and debugging capabilities by allowing you to access request metadata such as User-Agent, Content-Type, and custom headers sent with webhook events. For security, authentication headers and other sensitive headers containing keywords like auth, token, secret, password, and cookie are automatically filtered before storage. For more information, see the schema information in our Webhooks documentation.

We have made the following changes to our sync strategy:

• We now incrementally sync the ADDRESS and END_SHIPPER tables to capture new records.
• We also re-import the ADDRESS and END_SHIPPER tables monthly to capture updates and deletes.
• We now support priority-first sync for the SHIPMENT table. Priority-first syncs fetch your most recent data first so that it's quickly ready for you to use.
• We now perform a rollback sync with a 45-day rollback window to capture updates for the SHIPMENT table during every sync.

The Medallia connector now supports the Hybrid Deployment model. For more information, see our setup instructions.

We’ve updated the Tremendous connector to use the default OAuth 2.0 framework, from custom OAuth 2.0 setup. This update introduces a centralized OAuth app, simplifying authentication and improving security. Be sure to re-authenticate your Tremendous connection. For more information, see our setup instructions.

We now support the In operator for row filtering. You can filter rows by Integer, Serial, or String column values using a list of allowed values.For more information, see our row filtering documentation.

The extract_summary log event for database connectors now includes SQL query count information, in addition to the API call count for API-based connectors. We are gradually rolling out this change to all existing connections.See our extract_summary event documentation for more details.

We now automatically upgrade Quickstart Data Models to the latest version after a 30-day notification period. Once notified via email and in your dashboard, you have 30 days to prepare and review for each upgrade. This ensures you always have access to the most recent features and schema improvements. We add new output models to Quickstarts by default for new customers. Existing customers can choose to automatically allow or block new output models.Minor patch updates (e.g., vx.x.1 to vx.x.2) will continue to be applied immediately and do not follow the 30-day upgrade window.

We now support row filtering for incremental syncs. The filters are applied starting from the next sync after the filters are created.Previously, we supported row filtering for initial syncs and re-syncs only.For more information, see our row filtering documentation.

We've updated the following field names in log messages we send to Splunk:

• connector_id has been renamed to connection_id
• connector_name has been renamed to connection_name

We’ve made the following improvements to prevent data integrity issues:

• Improved how the list_work_item temp table filters work items to ensure cleaner and more efficient syncs. Project ID and team ID filters are now applied in the query body instead of the path parameters. This change eliminates duplicate work_item_ids that were appearing in some syncs.
• Added incremental sync support so that only new or updated work items are fetched during subsequent syncs.

We released Fivetran Proxy Agent 1.1.2, which is bundled with High-Volume Agent version 6.1.0/88. The proxy_host and proxy_port configuration settings now apply to all agent communications, including connections to the Fivetran control service. Previously, these settings applied only to data source connections.You can download the updated package from the Downloads page.

We now retain data in the REPORT_CAMPAIGN and REPORT_LINE_ITEM tables for up to 3 years.

We have added a new option, Intraday Export Type, to the connection setup form for the connections created on or after July 24, 2023. You can now select one of the following export types:

• Streaming - Sync a continuous export of events
• Fresh Daily - Sync periodic batch updates
• Daily Only (skip intraday events) - Sync only the full export of data that takes place daily

For more information, see our setup instructions and REST API config.

We unified each connector’s pre-built dbt Core–compatible data models into a single model. It standardizes and documents the Fivetran schemas created in your destination and outputs analytics-ready tables. Previously, most connectors used distinct source and transform models.

We now re-import the PRODUCT table daily to capture updates and deletes.

We now sync the sale_order entity data based on the facility code you provide in the connection setup form.

The SingleStore destination is now in beta. For more information, see the Singlestore destination documentation.