Fivetran Release Notes

We have fixed a bug that prevented us from syncing outstanding jobs data into the PARTNERSHIP_JOB table. Now, we correctly sync outstanding jobs data into this table.

We now support precision and scale for the money data type.

We now handle JSON columns with duplicate keys by retaining only the last key’s value when writing to a PostgreSQL destination. Previously, such columns were populated with an error message if duplicate keys were detected.

We now reference users by unique ID instead of email address to improve data security. We updated the relevant BigQuery and Snowflake sample queries that join the LOG and USER tables.

You can now specify the name of the Ordway webhook you want Fivetran to use. To facilitate this change, we have made the following changes:

• added a new field, Webhook name, to the connection setup form
• added a new REST API parameter, webhook_name, to the REST API connection configuration

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

We now support the Teradata Vantage destination for the following versions of dbt Core transformations: 1.7.19, 1.8.9, 1.9.10, and 1.10.11.

We now show a warning when the Google Cloud Storage bucket used for log events cannot be accessed from the Hybrid Deployment environment.

• The warning includes the underlying error message to help troubleshooting.
• Account- and destination-related metadata continues to sync; log event ingestion resumes automatically once access is restored.

If you see this warning, verify that connection from your Hybrid environment can reach the storage endpoint.

We now perform a rollback sync with a rollback window of 3 months to capture updates for the FORM_SUBMISSION table. We perform the rollback sync during every sync.

We have changed our connection re-sync strategy. During a re-import, we no longer delete historical records for tables with a retention period at the source because we won't be able to retrieve the older records.This change applies to the following tables:

• DOC_REPORT_FAILED_SEARCH
• DOC_REPORT_POPULAR_SEARCH
• DOC_REPORT_TOP_ARTICLE
• DOC_REPORT_TOP_CATEGORY
• DOC_REPORT

We are gradually rolling out this improvement to all connections.

We have changed our connection re-sync strategy. During a re-import, we no longer delete historical records for tables with a retention period at the source because we won't be able to retrieve the older records.This change applies to the following tables:

• AUDIT_LOG_CHANGE_RULE_CONDITION
• AUDIT_LOG_CHANGE_RULE
• AUDIT_LOG

We are gradually rolling out this improvement to all connections.

We have changed the sync strategy for the TIME_ENTRY table. We now re-import the table once a month to capture updates. Previously, we incrementally sync it to capture both new records and updates.

We now support dbt Core versions 1.9.10 and 1.10.11. You can use these versions to run your projects.{% note %}For these dbt Core versions, we don't support MySQL and Azure Synapse destinations. These destinations are not currently supported by the custom dbt Core adapters.{% /note %}

We now display replacement role name for the deprecated roles in the role field in the List all users within a group endpoint response.

We now re-import the VISIT_BY_TREND table once a week to capture updates.

We now capture deletes for the CUSTOM_OBJECTS table.

We now support priority-first sync for the EXPENSE table. Priority-first syncs fetch your most recent data first so that it's quickly ready for you to use.

Our Databricks Lakebase for PostgreSQL destination is now generally available.Read our Databricks Lakebase for PostgreSQL destination documentation.

We have made the following changes to our sync strategy for the LINK table and its child tables:

• We no longer re-import these tables to capture updates and deletes.
• We now sync only the last 6 month's worth of data during historical syncs.

We have updated our sync strategy. We now incrementally sync the RECORD table. Previously, we re-imported the RECORD table during every sync.