Life Cycle Management: Automatically Delete Old Versions of Master Workspaces
The rollout brick, which is part of the Life Cycle Management (LCM) feature, now supports the option of deleting old versions of a segment's master workspace. You can define how many older versions you want to keep, and all the other versions will automatically get deleted at rollout. You no longer have to delete them manually.
To set up the automatic version deletion, use the keep_only_previous_masters_count
parameter when scheduling the rollout brick.
Learn more:
Rollout Brick
LDM Modeler: Datasets without Facts
We have simplified the process of creating a dataset that contains only attributes (no facts). Now, you do not have to set a primary key in an attribute-only dataset to be able to publish the logical data model (LDM).
Attribute-only datasets do not require a primary key for data to be loaded to them. In a star-schema LDM, you can use such datasets as a factless fact table. You can upload data to those datasets and analyze them individually.
Learn more:
Data Modeling in GoodData
Enhancements to Data Loading Processes with "Warning" Status
We have made a few enhancements to how you can manage data loading processes that end with a "Warning" status.
Configuring Process Sequence to Run after Process Ends with "Warning" Status
When configuring a sequence of data loading processes, you can now allow the triggering process to end with a "Warning" status in order to execute the target process.
Learn more:
Configure Schedule Sequences
Setting Notifications on Processes Ended with "Warning" Status
You can now create a notification rule for a data loading process to get notified if the process has ended with the "Warning" status (for example, a process that loads data to multiple workspaces failed to complete for some workspaces while the process has completed successfully for the remaining workspaces).
Learn more:
Create a Notification Rule for a Data Loading Process
GoodData.UI Version 8.6 Available: Data Sampling, Insight Contributors, and Improved Sorting
We have released Version 8.6 of GoodData.UI, our powerful library for building analytical applications powered by GoodData.
What's new in GoodData.UI Version 8.6
- Data sampling in visualizations and DashboardView
Data sampling is now supported in visualizations and the DashboardView component.
Data sampling allows you to quickly determine the viability of a particular visualization by processing only a percentage of the total available rows in the associated datasets instead of processing all of the data. For more information, see Data Sampling.
NOTE: Data sampling is a feature of GoodData Cloud Native (GoodData.CN). This feature is available in GoodData.CN Version 1.2 or later (see the GoodData.CN user documentation) and supported only for Vertica. For more information about integration of GoodData.UI and GoodData.CN, see this article.
- Information about contributors in insights
When getting an insight using thegetInsights()
andgetInsight()
methods, you can now have information about the user who created the insight (createdBy
) and the user who last updated it (updatedBy
) included in the returned insight's object.
To include this information, set theloadUserData
option in the methods totrue
, for example:getInsights({ loadUserData: true })
- Improved sorting
When a request for a list of objects returns more than one page of results and you need to sort those results by a certain property, you can now download all the returned records and sort them by any property locally by using the newall()
andallSorted()
methods of the objects returned by the original request.
UPCOMING: Prevention of ADS Data Exfiltration
To improve data security and reduce risks of data exfiltration via external JDBC connections, we are planning to limit the number of rows that are returned to an external SELECT
query submitted to the GoodData Agile Data Warehouse (ADS).
This limit will be initially set to 10,000 rows. For the existing customers, we will be rolling the limit out gradually, and we will notify you of the limit enforcement on your ADS instances in advance.
What queries are external?
A SELECT
query is treated as external when it comes via a connection established from an IP address outside of the GoodData platform. For example, when you connect to an ADS instance from your SQL tool to retrieve some data, this is an external query.
How will it affect you?
We do not expect any platform-wide impact because:
- By design, you should not export large tables from the ADS outside of the GoodData platform.
- SQL queries executed for debugging and diagnostics of data loading processes typically return a much smaller number of rows than 10,000.
You may be affected if you connect to ADS instances via the JDBC driver and need to regularly run SELECT
queries that return more than 10,000 rows.
Loading data from the ADS to your workspaces will not be affected!
Any data loading processes that load data from the ADS to your GoodData workspaces (for example, data loaded through the data preparation and distribution pipeline or by custom bricks that have been implemented by GoodData specialists) will not be affected. This limit does not apply to them.
ACTION REQUIRED!
We encourage you to review your typical ADS usage and help us set the limit to a number that allows your data engineers to do their daily job without any obstacles, but at the same time prevents a potential malicious insider from offloading large portions of your data by a single SELECT
statement. Please contact GoodData Support to have the limit set to whatever works best for you.
What will happen if you do not send a request to adjust the limit?
When the limit is implemented, your SELECT
queries submitted to the ADS from an external IP address will return 10,000 rows at most.
REMINDER: Upgrade Your GoodData.UI to Version 8.x
Version 7.x of GoodData.UI reached its end-of-development, and it is now in maintenance mode until March 31, 2022, when it will reach its end-of-support. The versions older than Version 7.x are no longer supported.
How does it affect you?
You are affected if you are using a version of GoodData.UI older than Version 8.x.
ACTION REQUIRED!
Upgrade your GoodData.UI to Version 8.x. We recommend that you do so as soon as possible.
- Review the breaking changes in Version 8.0.
- Follow our step-by-step Migration Guide that will walk you through the migration process.
What will happen if you do not upgrade to Version 8.x?
When Version 7.x reaches its end-of-support, it will receive neither new features nor bug fixes. No technical support will be provided. Although the version will still be available for use, we do not recommend that you use it.
Learn more:
To learn more about how different versions of GoodData.UI go through the life cycle phases and what level of support you can expect when using a specific version, see Supported Versions.
REMINDER: End-of-Support of Internet Explorer 11
GoodData is announcing the end-of-support of Microsoft Internet Explorer 11 effective November 19, 2021.
How does it affect you?
You are affected if you are using Internet Explorer 11 for accessing the GoodData Portal.
ACTION REQUIRED!
In preparation for the end-of-support, switch to one of our supported browsers, such as Microsoft Edge, Google Chrome, or Mozilla Firefox.
What will happen if you still use Internet Explorer 11 after November 19, 2021?
After this date, you will no longer be able to use the following:
- Analytical Designer
- Dashboards (Free, Growth) / KPI Dashboards (Enterprise)
Support for the following components will continue:
- Reports (Enterprise)
- Dashboards (Enterprise)
Major GoodData.UI versions released after this date will not support Internet Explorer 11.
Learn more:
See the list of supported browsers in System Requirements and Supported Browsers.
REMINDER: Use TLS Version 1.2 or Higher Only
Support for Transport Layer Security (TLS) version 1.0 and 1.1 is deprecated and will end in one of future releases. We will announce the exact date as soon as it is set.
How does this affect you?
All modern web browsers and programming languages already support the newer versions of TLS (1.2 and higher). Unless you use obsolete software or unsupported systems, you are not affected.
You may be affected if you are using an older version of the libraries or programming languages in your tools. To verify, check the documentation for your tool:
- Java: https://blogs.oracle.com/java/post/jdk-8-will-use-tls-12-as-default
- Microsoft Windows: https://docs.microsoft.com/en-us/security/solving-tls1-problem
- Python: https://docs.python.org/3/library/ssl.html
- Ruby: https://blog.rubygems.org/2018/02/24/tls-10-and-11-deprecation-notice.html
ACTION REQUIRED!
Make sure that all your components and integrations use TLS version 1.2 or higher.
What will happen if you do not upgrade to the newer versions?
When support for TLS version 1.0 and 1.1 ends, if you do not switch to a newer version of TLS, the integration with the GoodData platform will stop working.
Learn more:
- For more information about the deprecation of support for TLS version 1.0 and 1.1, see Support for TLS Version 1.0 and 1.1.
- For more information about TLS protocols, see Supported TLS Security Protocols and Ciphers.
- For information about all deprecated features, see Deprecated Features.