Matillion ETL for Snowflake Release Notes
Matillion ETL for Snowflake 1.37
Important Notice: The Data Transfer component is now the preferred way to move files/objects between storage providers. Existing components such as S3 Get/Put, GCS Get will continue to work in existing jobs, but new jobs should use Data Transfer.
- New Data Loading orchestration components:
- All Data Loading orchestration components can now display data directly sampled from their source, allowing users to easily check their current configuration.
- All Data Loading orchestration components can now display the SQL generated by the component.
- Azure SQL Query component connects to a Microsoft Azure SQL Database. (Only for users with Matillion ETL instances hosted on Microsoft Azure)
- JDBC Table Metadata to Grid component connects to many types of JDBC database and can export the metadata from a source table into a Matillion ETL Grid Variable.
- A new Migration Tool can help move any number of Matillion ETL assets directly from one Matillion ETL instance to another.
Note: The preferred (and safest) way to upgrade is now to launch a new copy of Matillion ETL running the latest version, use the Migration Tool to move and validate the new version, before deleting the existing instance. In-place upgrade may be removed in future versions.
Matillion ETL for Snowflake 1.36
- A selection of new components to connect to various services:
- Shopify Query connects to the user’s Shopify account.
- Survey Monkey Query loads data from a SurveyMonkey database.
- Zoho CRM Query retrieves Zoho CRM data.
- Dynamics 365 Sales Query connects to the Sales service in Dynamics 365.
- Dynamics 365 Business Central Query connects to the Business Central services in Dynamics 365.
- DynamoDB Query loads data from Amazon’s DynamoDB.
- Blob Storage Unload Component (for users hosted on Azure) unloads data from your Snowflake warehouse into Azure Blob Storage.
- ORC and PARQUET file formats now supported in S3 Load.
- Many UX improvements including automatically connecting components on the canvas, improved variables workflow and new keyboard shortcuts.
- Selected Environments are now user-specific. Users can now specify their environments independently of one another.
- Users can now freely copy, cut, and paste jobs within a project.
- Autocompletion prompts now appear in many places when using Matillion ETL variables in code.
- CloudFormation Launch options (for users hosted on AWS)
- Data Lineage
- Allows you to understand the effect that your complex transformation jobs will have on your data. Track a column backwards to its source to determine where and how calculations are applied.
- This is an Enterprise Feature and thus is available to customers using large and xlarge instance types.
- New wizards to easily create Incremental Load Shared Jobs:
Matillion ETL for Snowflake 1.35
- New “Data Transfer” Component that boasts all the functionality of the existing S3 Get, S3 Put and Cloud Storage Put components, plus additional source and target destinations (Azure Blob Storage).
- In addition to AWS and GCP credentials, environments can now reference Azure credentials to interact with Azure services such as Blob Storage.
- A new “Apache Hive Query” component connects to your Apache Hive data warehouse.
- A new “LinkedIn Query” component connects to your company’s LinkedIn apps.
- A new “Bing Search Query” component connects to the Bing Search API.
- A new “Bing Ads Query” component connects to the Bing Ads service.
- A new “Dynamics 365 Sales Query” component connects to the Dynamics 365
- Allow uploading the native Microsoft SQL Server JDBC Driver (the bundled jTDS driver is often the fastest in scenarios where it works)
- New ‘Extract To New Job’ function available by right-clicking a selection of multiple components on the canvas. Allows users to instantly create new jobs from a group of components, tidying up workflows and helping to create reusable jobs.
- Support for running on Azure platform
- Support for Snowflake’s AUTOINCREMENT option in the Create Table component. Particularly useful for easily creating a unique key column on a new table.
Matillion ETL for Snowflake 1.34.5
Matillion ETL for Snowflake 1.34
Important Notice: On upgrade, a background task will restructure the task history. During this time not all historic tasks will be available to view in the UI or API. The process only takes a few minutes in the general case but can take several hours if you have millions of run history items. This will require additional disk space (either on the instance or on RDS depending on your setup) so ensure you have at least 50% free space before attempting the upgrade.
- Shared Jobs:
- You can now turn your reusable orchestration jobs into their own components with their own parameters, help and Icon.
- Shared jobs can be packaged and distributed across multiple ETL instances with Import and Export.
- Historic Task Viewer:
- Previously completed tasks can be viewed on the canvas along with any parameter errors.
- You can understand the canvas state of a job and also see the jobs contained in a Shared Job.
- An “Unconditional” connector:
- Its now simpler to build orchestrations where the next orchestration step is run regardless of the success or failure of the prior step. This avoids use of extra “and” and “or” components to achieve the same thing.
- “Auto Debug” for all Data Loading components:
- Data Loaders come with the Auto Debug property. When switched on, allows users to choose between 5 levels of Debug Logging verbosity.
- Makes it easier to retrieve logging information without console access to the Matillion ETL Instance. Include these logs in your support requests for much faster turnaround!
- Warning: Can potentially consume large amounts of disk space. Do not leave this switched on unless directly in need of it!
- It is now possible to import, export and modify permissions via the API.
- You can now bulk export data to Amazon RDS sources with the RDS Bulk Output Component
- The Create Table component now supports Snowflake's Transient and Temporary tables.
- New data loading component Zuora Bulk Query:
- Loads data from Zuora into Snowflake using their Bulk API.
- OpenID Connect support for third party login providers:
- You can now configure Matillion ETL to authenticate with any Open ID Connect provider.
- Default support for Google, Microsoft and Okta plus a “Generic” option.
- The SQL Script component now supports multiple queries.
Matillion ETL for Snowflake 1.33.10
- Hot fixes for Salesforce OOM
Matillion ETL for Snowflake 1.33
Important: Queries using the Advanced Mode of the Google BigQuery Query Component will pass the SQL directly to BigQuery without any interpretation. If this causes any problem, please set the Connection Option 'Query Passthrough' to FALSE
Important: Users with the API role bypass some permissions checks (reads) during API calls.
- Open Exchange Rates Query component connects to the Open Exchange Rates API.
- Grid Iterator allow iterating the values of a Grid Variable, similarly to iterating through a table of values.
- SQL Editor (in all Query components) now shows available Tables/Columns and Variables to help you author and test SQL queries from source systems.
- A new “Notices” V1 API endpoint allows you to query the current system notifications and post new messages which notify all users.
- A new “User Configuration” V1 API endpoint allows you to do user management via the Matillion API.
- Matillion no longer requires “listAllBuckets” permission (although this is still recommended).
- Job Variables (scalar and grid) now have a “Visibility” that determines how they are used elsewhere.
- All variables now have a description.
- 100+ bug fixes across all areas of Matillion ETL.
- A new Construct Variant component to create Snowflake variants from data columns.
- Transpose Rows component to aggregate multiple rows of data into a single output row.
Data staging (from all Query components) may use INTERNAL Staging so you don’t need to care where intermediate files are stored.
- Sequence Support has been added from ‘Environments’ context menu within the client. Users can create their own Snowflake sequences that are accessible by components for easy generation of unique numbers across their datasets (for example, as a way of introducing primary keys).
Matillion ETL for Snowflake 1.32.8
- Change: Allow Tomcat to start even if there is a corrupt schedule attached to a job. A bug fix to prevent the corrupt schedules attached to jobs in Matillion ETL 1.32 is currently in progress.
- Bug Fix: Prevented inappropriate authentication errors when using basic OAuth in the Zuora Query component.
IMPORTANT : Ensure you have a backup before you upgrade. Security configuration changes are applied on upgrade. These changes cannot be reversed, so do not use “yum downgrade” (or similar) to attempt to get back to versions prior to 1.32.
Matillion ETL for Snowflake 1.32
Enterprise Only: This version of Matillion introduces a new Permissions system that allows users to:
- Setup users with fine grained permission sets that can limit the 100+ core functions of the tool
- Provides default permission groups:
- Reader - Read only user who can’t modify a project
- Reader with Comments - Reader with ability to add notes to jobs
- Runner - A user who can execute but not modify jobs
- Scheduler - A user who can execute, schedule and change related config
- Writer - A user who can create ETL jobs but not delete projects
- Additional permission groups can be added at any time and are organised hierarchically making them easy to set up.
- A new suite of Grid Variable components are now included to make populating and manipulating them simpler - often without requiring any scripting:
- A new “SendGrid Query” component connects to the sendgrid email delivery platform
- A new “ElasticSearch Query” component to connect to the elasticsearch search engine
- A new “Magento Query” component to connect to the Magento content eCommerce system
- A new “Zuora Query” component to connect to the Zuora subscription software platform
- A new “GMail Query” component to connect to Google’s email service
- A new “Run Now” action has been added when defining a schedule
- Double-clicking a component on the canvas now opens the components “default” editor, if it has one. For example, double-clicking a Bash Script component will begin editing the script.
- Internal User: When using the “internal” security option tomcat user passwords are now hashed when stored on disk.
- External (Domain-based) Login: You can now encrypt your Realm Password with the AWS Key Management Service (KMS)
Matillion ETL for Snowflake 1.31.8
The "Google AdWords Query" component has been updated to support the latest Google AdWords API's.
Matillion ETL for Snowflake 1.31.7
Important (possible breaking change): API Profiles ("RSD’s") that handle paging may need to be tweaked to disable “auto” paging. Please see here for more details.
Important (possible breaking change): API profile limits are now applied. Where the default of 100 is set it will now be applied. This could affect API Query Components which previously ignored that limit.
- Zendesk Query orchestration component for loading data from the Zendesk customer relationship system.
- Mixpanel Query orchestration component for loading data from Mixpanel product analytics system.
- Xero Query orchestration component for loading data from the Xero accounting system.
- Dynamics 365 Query orchestration component for loading data from Microsoft Dynamics CRM/ERP.
- API Profile RSD Generator
- Accelerate the development of API Profiles using a new tool that automatically generates a basic XML “RSD descriptor” for any API endpoint, based on a sample of data returned.
- REST API Version 1 - Matillion ETL now has full API coverage:-
- You can now read/write more assets (JDBC Drivers, credentials, SQS configuration) as well as allowing finer-control of which resources to include.
- A map of the v1 API is available here.
- The “v0” api is still available and unchanged.
- Grid Variables System
- In addition to “scalar” (single-valued) variables, you can now define grid variables to hold lists and grids of values; use them wherever a compatible list or grid of values is required.
- Grid variables can be manipulated/modified in Python.
- You can pass values for grid variables when starting a job via SQS and/or the V1 API.
- You can now disable parts of an Orchestration job.
- Improved Matching in column mappings - Many transformation component “Column Mapping” parameters can now be automatically mapped, even when the input and output column names are similar but not identical.
- Viewless Architecture
- You may remove old views that were generated by previous versions of Matillion ETL for Snowflake, via the right-click menu of each environment.
- You may optionally define a “Default Role” in the Environment. Whenever a connection is required, the given username is first authenticated and then switched to the given role.
Matillion ETL for Snowflake 1.30.6
- The Google BigQuery Query component now supports standard SQL.
- New Intersect and Except transformation components
- New S3 Load Generator can inspect S3 files and generate Create Table / S3 Load components.
- Redesigned "Scheduler" user interface to simplify the management of scheduled orchestration jobs.
- New "Task Info" panel and "Task" panel make it much easier to understand complex tasks both at run time and after job execution.
- Matillion variables can be defined and scoped at job level making jobs much more reusable. Variables can now be passed to and returned from jobs.
- New Quickbooks Online Query component to connect to the popular online accounting system.
- New Square Query component to connect to the payment system.
- New Google Custom Search component allows google search data to be ingested.
- All data-staging components can append rows to an existing table as well as creating new tables.
Matillion ETL for Snowflake 1.29.9
- Matillion ETL for Snowflake Introduces the ability to configure Matillion ETL in a highly available topology with fully active-active cluster. This feature is only available on large and xlarge instance types.
- Jobs run from SQS, the API or the built-in Scheduler will now fail-over in the event of an instance failure.
- Scheduled runs missed because a server is offline will be run when it becomes available again
- Once two or more members are in the cluster, a Cluster Info tab shows membership status and activity.
- OAuth tokens, Database Drivers and RSD Profiles are replicated via the persistence database (postgres)
- Logging from each node is sent to Cloudwatch
- New Jira Query component loads data from Atlassian's popular Software Development Platform.
- New PayPal Query component can load payment and other data from Paypal Business accounts.
- New ServiceNow Query component loads data from Servicenow’s IT Service Management (ITSM) platform.
- New Stripe Query component loads data from Stripe’s payment platform
- New Email Query component can query an IMAP based email system.
- New YouTube Analytics component can query data from the YouTube Analytics API.
- All query components now allow you to override the output table so you can specify an existing table to load or append to.
- Excel Query can now load files from Google Cloud Storage, as well as Amazon S3
- You only see S3 and/or GCS when you have credentials in the environment, otherwise they are hidden.
- New option to drop a schema from the Environment Tree.
- Specify a region in S3 Unload (to allow writing to buckets to other regions)
- S3 / Google Cloud Storage file browser enhancements
- Set advanced connection options during OAuth flow (e.g. to connect to a Salesforce Sandbox)
- Warning: Manage Backups and View Audit haven't been removed, they have been moved to the Admin menu
- Map Values component
- Additional support for ORC file types
IMPORTANT Upgrade Notes: All data-staging components now create a target table with a wider range of target data types. Mostly this will be transparent, however if your source data contains variables with the Boolean type, these may have an impact on downstream logic so please test jobs after upgrade.