Zoomdata version 2.6 Release Notes
These release notes cover changes and fixes made to releases under Zoomdata version 2.6. The latest updates are available in v2.6.7. You can identify your release version at the Login screen.
With Zoomdata version 2.6, we introduced a new model for feature and patch updates. Version 2.6 started as a rapid release version that adds stability fixes and new functionality every four weeks. Version 2.6.6 onwards, Zoomdata 2.6 has become the long-term supported (LTS) release. For more information, see Zoomdata Release Vehicles and Licensing Management.
Zoomdata Version 2.6.7
- Fixed a couple of problems with uploading file based data with certain input values.
- Resolved a scrolling issue on IE11 when using the raw data table field selection controls.
- Addressed a bug with deletion of disabled users when OAuth tokens existed for them.
- Fixed an issue with deletion of accounts created in a previous version.
- Introduced a configurable maximum number of fields setting when using the file upload service to upload files with a large number of fields.
- Added user attribute in the request activity entries in zoomdata-access.log file to allow easy mapping of users to requests.
Zoomdata Version 2.6.6
With the release of 2.6.6, Zoomdata version 2.6 has been declared the long-term supported (LTS) release of Zoomdata. The previous LTS release, Zoomdata version 2.5, will continue to be supported until August 1, 2018. Customers using earlier versions of Zoomdata are encouraged to upgrade to Zoomdata 2.6. For questions on upgrading, contact Zoomdata Application Support.
- Keysets is a brand new Zoomdata feature available for preview in 2.6.6. With this feature, data explorers can significantly expand on use cases where data exploration requires tying multiple data sources together. Analysts can now create a set of keys using the result of one chart and feed the keys into a completely different chart or dashboard, which uses different (or the same) data sources. Imagine use cases, such as in healthcare, where a hospital wants to analyze readmission rates for a cohort of patients who were in the same drug trial where data on readmission rates and drug trials was stored in different data sources. Contact your Zoomdata representative to learn more about this feature and give it a test drive.
- All charts that allow multiple groupings now support grouping on multiple time fields, as long as the underlying data source supports this functionality.
- Dashboards can now have a Refresh Data button for users to bypass any dashboard cache and get the latest data from the underlying data source.
- SQL Server now supports derived fields (row level calculations). For more information, see Creating Calculations in Zoomdata.
- Impala connector now supports accessing Kudu tables in addition to Impala tables.
- Hive 2.x connector now supports the Hive on LLAP configuration.
- Zoomdata Query Engine can now be deployed using an externalized Spark on Yarn configuration on Hortonworks clusters.
- Numeric filters now support the “NOTEQUALS” operator for specifying a filter condition.
- SQL Server data sources in Zoomdata now support the use of Unicode characters.
- Fixed an issue where UNIX timestamp fields stored as a float or decimal number in SQL Server were not handled correctly for grouping.
- Addressed a problem with logging of user authentication events in certain cases.
- Resolved a bug where multiple connections were created to SQL Server when using Windows authentication.
- Addressed a bug where derived fields are not visible on some data sources.
- Fixed an issue where the time range shown on the time bar could be sometimes incorrect.
- Resolved a UI problem where the scrollbar style wasn’t consistent on a Windows machine.
- Addressed an issue with an AVG calculation when dealing with Decimal types on a data source using the Spark SQL connector.
- Fixed a bug with using Japanese characters in fused attribute names inside a Fusion source.
- Resolved an issue with type detection for certain fields when uploading file data into Zoomdata.
- Support for older versions (2.x) of Elasticsearch has now been deprecated since Elastic has dropped support for this version of Elasticsearch.
- Removed the option for a user to specify whether a dashboard would “Remember Time Position” when it’s saved. Time position is now automatically remembered every time a dashboard is saved. Users can specify dynamic ranges on their time filters to auto-update time filters on their dashboards.
Zoomdata Version 2.6.5
- HDFS and S3 now support derived fields (row level calculations). For more information, see Creating Calculations in Zoomdata.
- Elasticsearch 6.x connector is now installed by default.
- SQL Server connector no longer requires a separate JDBC driver.
- Added ability to search in attribute, metric and filter pickers when working with large number of fields.
- Enhancements to cache handling to support larger number of dashboard viewers.
- Fixed an issue related to log rotation in connector logs.
- Deprecated support for older versions of Spark (1.5/1.6) in the Spark SQL connector.
- Elasticsearch 2.x connector is no longer installed by default, it must be installed manually.
- Removed the ability to set and change fonts on charts.
- Removed support for exporting to XLS file format.
Zoomdata Version 2.6.4
- Smart axis formatting is now applied to charts to automatically roll up large numbers to the most appropriate scale, for example millions and billions.
- You can now group by numeric fields within your charts.
- Elasticsearch 6, Spark SQL and Hive on Tez now support derived fields (row level calculations). For more information, see Creating Calculations in Zoomdata.
- Fixed an issue related to handling of PreviousPeriod() in calculations.
- Resolved a problem related to displaying HTML formatted text in raw data table.
- Addressed an issue related to rendering lines based on calculations in line trend chart.
Zoomdata version 2.6.3
- Zoomdata now offers a command line interface (CLI) for developers to create custom charts in a development environment of their choice. The CLI is an alternative to the Chart Studio web-based interface and helps developers manage all aspects of custom charts created within your environment.
- You can now use non-Latin characters in expressions and calculations. To see what characters are supported, see Creating Calculations in Zoomdata.
- The Elasticsearch connector has been updated to support ES versions 5.6.x and 6.0.x
- The Spark version bundled with Zoomdata is upgraded to Spark version 2.2.1.
- Data sources public REST API was updated to v2 and the prior v1 public data sources API was deprecated. Note this change requires updates to client applications to pass in v2 content type header.
- Resolved a problem with the Lines Trend chart style where using calculated values on the Y axis of the chart cased the chart to not render properly.
- Fixed a bug related to the Raw Data Table chart style where HTML formatted strings were no longer automatically converted to hyperlinks within the chart.
- Addressed an issue with using the time transform calculation where applying a shifted timezone to a particular time field in the datasource resulted in charts that grouped by time to not properly display data for the timezone shift.
- Zoomdata’s Chart Studio is now deprecated. You can continue to make custom charts and carry out other developer tools using the new custom chart CLI. If you want to create custom charts in Chart Studio, a Supervisor can toggle it on.
Zoomdata Version 2.6.2
- You can now filter on metrics within your dataset, as well as previously created calculations. In addition, filtering capabilities have been split into filtering on datasource fields and filtering on metrics and calculations. Filtering on datasource fields are now referred to as row level filters and filtering on metrics/calculations are called group filters. For more information about using group filters, see Filtering data.
- Administrators can define the type of formatting for number sets within a dataset, which is done at the level of the data source. This allows for greater control over the formatting of decimal points, numeric units, and negative numbers. In addition, Administrators can now format the following:
- Plain number formats
- Computer storage units
- Scientific notation
- When defining the format of fields, Administrators can also set the required locale for groups of users/users to display the correct format for their geographical location. For more information, see Users and Groups.
- The Solr data connector in Zoomdata has been upgraded to include support for Solr 6.4+ and Solr 7 versions.
- When exporting chart data from visualizations from your environment, they are now exported in CSV file formats instead of Excel sheets format(.xls). For more information, see Exporting Charts and Dashboards.
- Improved tooltips for visualizations by updating the positioning and color indicator.
When upgrading from previous versions, the default port for Zoomdata has changed from 8443/HTTPS to 8080/HTTP. HTTPS can be configured using the steps in Configuring Zoomdata Property Files. In addition, prior releases of Zoomdata supported running both https and http simultaneously, this dual port configuration is no longer supported by Zoomdata but can be achieved using a proxy front ending Zoomdata.
- Resolved an issue with OAUTH token handling.
- Fixed a bug related to the Upload API where concurrent REST API requests were not supported and caused HTTP errors to be thrown.
- Addressed a problem with loading visualizations.
- Resolved a problem with previewing data for a CSV upload data source where different columns in preview mode were incorrectly shown as identical.
- Addressed a problem with color selection in visualizations where changing the color palette from gradient to distinct colors would not update the color scale.
- Fixed a bug related to Line & Bar and Floating Bubbles chart styles where in-visualization tooltips would not disappear after expanding the chart to full screen.
Resolved a problem with MongoDB Atlas where data collections were not retrieved after the initial connection.
- HATEOAS links are no longer provided with the public API endpoints.
- HTTP redirect has been deprecated.
Zoomdata Version 2.6.1
- Controls inside a dashboard widget are now responsive to widget size and are minimized when not needed. The toolbar controls are now only shown on the dashboard canvas when hovering over the widget. Also, dashboard widget controls and group/metric pickers collapse or disappear depending on the size of the widget.
- Zoomdata Administrators now receive earlier notifications about the status of the license expiration date for their environment. For more information about requesting a new license key, see Managing Licensing for your Environment.
- An SAP Hana connector is now available from Zoomdata (subject to license), which supports Row Level Calculations. For more information, see Managing your SAP Hana connectors.
- Added a REST API endpoint to programmatically create personal OAUTH tokens for use in embedded scenarios.
- Updated the public REST API for connections.
- Resolved an issue around calculations yielding null values where the null values were displayed on the color legend range, instead of including numeric values.
- Addressed a problem around the Raw Data Table chart type in data source configuration where modifying or removing and re-adding the index of the underlying data source would cause the chart to error out when trying to render.
- Fixed a bug related to upgrading metadata from version 2.3 to version 2.6.
Zoomdata version 2.6.0
Charts and Dashboards
- You can now create derived fields within calculations for your charts and dashboards. This allows you to build visualizations using both the fields in your data source and also derived values of your data to enrich your insights. You can use derived fields and row-level expressions as arguments or aggregation functions in calculations, allowing you to create advanced calculations. This functionality is supported on Impala and PostgreSQL. For steps and information, see Creating Calculations in Zoomdata.
- Dashboards have an improved chart layout, make the responsiveness faster and increasing the ease of interaction. The layout improvements also enables smaller sized chart widgets within a dashboard, meaning you can view more data at one time.
Server and Client Application
- Zoomdata now uses Apache Spark as its internal processing engine. For more information, see How Zoomdata uses Spark.
- The Query Engine is now a standalone service within the Zoomdata environment and offers its own deployment and configuration options. For more information about how Query Engine performs, see Managing your Query Engine.
- If you have an existing externalized Spark cluster, specifically Spark on Yarn on Cloudera clusters, you can use your existing cluster for use with Zoomdata.
- You can now leverage FluentD as a unified logging layer in conjunction with Zoomdata, meaning that you can direct your log files to a central repository using FluentD. For more information, see Setting up Unified Logging.
- The Upload API connector no longer requires MongoDB for data storage. Data for Upload API is now stored in PostgreSQL. For more information, see Using Upload API.
- The Flat File (CSV) upload functionality now utilizes PostgreSQL as a data store.
- Amazon S3 and HDFS data sources now use their own embedded Spark instance for data processing.
- The SparkIt capability for data sources is now fully removed and is no longer available.
- Amazon S3, HDFS, and MongoDB data sources are now migrated over to use Zoomdata's own microservice architecture.
Security and Account Management
- For users imported in to Zoomdata via SAML or LDAP, the auto-creation of groups that map to users existing groups is now configurable.
- Security scanning for your Zooomdata environment is improved.
- Custom visualizations can now be defined to use up to 10 group-bys within a chart.
- REST API to append data to an existing Upload data source is now supported.
Known Limitations and Issues
- Upload API and File Upload interpret values which include a currency symbol such as $10.00 as attributes (text), not as numbers.
- Upload API and File Upload do not calculate date and time properly when your PostgreSQL server is not in UTC time format.
- HDFS and S3 connectors have known limitations around grouping time fields that were converted from strings.
- Upload API and File Upload only support upload of individual files less than 200MB. For larger file sizes, split the files up into smaller files and add them incrementally to an existing File Upload source.
- Derived fields have the following limitations in 2.6.0:
- Play mode, Live mode, and data sharpening are not supported.
- Data profiling is not supported.
- Derived fields cannot be used on fusion sources made of Impala and PostgreSQL.
- The following chart types are not supported: Histogram, Boxplot, Percentile, and Last Value.
Was this topic helpful?