February 2020

Databricks

We have added Databricks as one of our supported destinations. You can now use Databricks as your destination with Fivetran connectors.

Read our Databricks documentation.

Zendesk Support Portal

Fivetran has introduced a new Zendesk-based Support Portal where you can update your tickets and track their progress and status.

To learn more, log in to your Fivetran account and go to the Fivetran Support Portal guide.

New connectors

Oracle E-Business Suite

Oracle E-Business Suite (EBS) is an enterprise resource planning system developed by Oracle.

Read our Oracle EBS connector documentation.

Schema changes

AppsFlyer

Data collection using Push API V1.0 and Data Locker V1.0 has been deprecated. We will continue to support all deprecation changes for the next three months, so be sure to make the required schema changes by May 31, 2020.

Use Push API V2.0 and Data Locker V2.0 to continue using the service. Connectors must register the Push API V2.0 specific URL generated by Fivetran, with AppsFlyer. For more details see AppsFlyer Push API V2.O documentation.

We now sync data from Push API V2.0 and Data Locker V2.0 into a common table EVENT.

We have deprecated the following tables:

• DATA_LOCKER_V1_EVENTS
• DATA_LOCKER_V2_EVENTS
• WEBHOOK_EVENTS

Asana

We have added the completed_by_id column to the TASK table. Asana recently added this field to the Asana API Task endpoint. It lets you know who completed the task.

Braze

We have added the following fields to the tables below:

• first_entry and last_entry to the CANVAS table
• conversion_behavior to the CAMPAIGN_CONVERSION_BEHAVIOR and CANVAS_CONVERSION_BEHAVIOR tables
• timezone to the CAMPAIGN_CONVERSION_EVENT, CAMPAIGN_ENROLLMENT_EVENT, CANVAS_ENTRY_EVENT, and EMAIL_EVENT tables

Stripe

We have added a new field, start_date, to the SUBSCRIPTION table. This field stores the actual subscription start date, while the start field in the same table stores the last modified date of the subscription.

We have added a new table, SUBSCRIPTION_SCHEDULE, which stores the historical change data of a subscription.

We have added a new table, PHASE, which stores the details about phases in the subscription schedule life cycle.

We have added a new table, PHASE_PLAN, which stores the subscription plan details during a specific phase.

We have added the following fields to the BALANCE_TRANSACTION table:

• payout_id. This field stores the automatic payout ID in which the balance transaction was paid out.
• exchange_rate. This field stores the exchange rate used for a transaction.

Twitter Ads

We have made multiple schema changes to support Twitter API v6 changes. You can read detailed API documentation on Twitter's website. We will continue to support all deprecated items for the next three months, so be sure to make the required schema changes by May 25, 2020.

• We have renamed the CITIES segment to METROS. However, the CITIES sub-string will be present in all analytics tables.

• We have added the following fields to the tables below:

  • media_key to the CARD_HISTORY table
  • poster_media_key and poster_media_url to the MEDIA_LIBRARY table

• We have added four new fields to the TWEET table because the Twitter API endpoint has changed from scoped_timeline to tweets:

  • tweet_type (DRAFT, PUBLISHED, SCHEDULED)
  • media_key
  • card_uri
  • full_text

• We have deprecated and replaced the following fields in the tables below:

  • text with full_text in the TWEET table
  • video_poster_url with poster_media_url in the CARD_HISTORY table
  • media_id and video_content_id with media_key in the ACCOUNT_MEDIA_HISTORY table
  • image, wide_app_image, and video_url with media_url in the CARD_HISTORY table

• We have deprecated the following fields from the tables below:

  • media_ids from the SCHEDULED_TWEETS table
  • video_hls_url from the CARD_HISTORY table

  We will stop syncing data from the VIDEO_APP_DOWNLOAD_CARD, VIDEO_CONVERSATION_CARD, VIDEO_DIRECT_MESSAGE_CARD, and VIDEO_WEBSITE_CARD tables to the CARD_HISTORY table. However, we will continue syncing the data from the VIDEO_POLLS table to the CARD_HISTORY table.

Xero

We have updated the data type of the inventory_asset_account_code column in the ITEM table from INT to STRING to match the type in the Xero API.

Improvements

Adobe Analytics

We have fixed a problem where customer records were overwritten in the destination, causing data mismatch. Fivetran will re-sync the connector to resolve the issues.

We now support aggregations at the hour and day level. Use the new Aggregation field in your connector setup form to fetch data based on your desired date granularity.

AdRoll

When setting up your connector, you can now select which advertisables you'd like to sync. Previously, Fivetran synced all advertisables by default.

Azure Blob Storage

We have improved the way we track which files we have already synced to make sure we only pull new or changed data from the source containers. Previously, we re-synced files that were created at the same time as the last observed cursor position. That ensured that we never missed any files that were created while we were syncing your data. That also meant that we sometimes synced the same files twice. Now, in addition to tracking the timestamp, we also track the names of the files we have already synced. We store up to 1,000 file names. We sync files created at the time of the last observed cursor position only if we don’t have the file in our list of synced files for that timestamp.

Braze

We have added primary keys and foreign keys to improve data integrity.

We have updated the delete mechanism to keep historical data. Now, when you delete a row from the source table, we set the _fivetran_deleted column value of the corresponding row in the destination to TRUE.

LinkedIn Ad Analytics

You can now choose the conversion window size for view-through attribution and post-click attribution in your Fivetran dashboard. The default values of the conversion windows are:

• View-Through Attribution: 7 days
• Post-Click Attribution: 30 days

See LinkedIn's API documentation for more details.

Marketo

We now use a separate API endpoint to retrieve deleted lead activities, because Marketo’s BULK API does not support these activities. We can’t retrieve deleted lead activities older than 14 days because Marketo doesn’t retain them.

MongoDB

We have updated our MongoDB connector to improve sync reliability. As a part of this update, we have begun rolling out support for BINARY and BSON_BINARY data types.

We have updated our MongoDB connector to support syncing data from MongoDB versions 2.6 to 4.2.

MySQL

We have added a new test, Check Binlog Retention Period, to the Amazon RDS for MySQL connector setup form. The test checks your binlog retention period and displays a warning if:

• Your MySQL database's binlog retention period is less than 24 hours
• You do not have permissions to access the mysql.rds_configuration table

Oracle

We have fixed a bug with FLOAT, REAL, and DOUBLE PRECISION column values. Previously, these values may have been truncated in the destination, but they are now copied to the destination exactly as they appear in the source database.

Pinterest Ads

When setting up your connector, you can now select which advertisers you'd like to sync. Previously, Fivetran synced all advertisers by default.

PostgreSQL

We now support partitioned tables. You will be able to see both partitioned tables and their partitions in your Fivetran dashboard.

Snowflake

We now support key pair authentication to connect to your Snowflake destination.

Stripe

You can now configure your Stripe connector using the Fivetran REST API. This feature is in BETA and available only for Standard and Enterprise accounts.

Xero

We now sync archived records from the TRACKING_CATEGORY table.