このセクションの記事

もっと見る

Yellowfin BI 9.17.1 - build 20260619

Avatar
徳田 真奈美
  • 更新
フォローする

Yellowfin BI 9.17.1 - build 20260619
Published: June 23, 2026
国内公開日: 2026/06/26

製品リリースにおけるナンバリングのルールを変更しています。
従来：V8.0.1→V8.0.2→V8.0.3 …
Ver9：V9.1.0→V9.2.0→V9.3.0 …
※必要に応じてパッチリリースされる場合があり、その際のナンバリングは9.x.1→9.x.2…となります。
詳細につきましてはサポート提供規約をご確認ください。

ダウンロードはこちら
※ダウンロードにはログインが必要です。

 

Yellowfinの古いバージョンにおいて、セキュリティ脆弱性が確認され、最新バージョンにアップグレードしていないお客様に影響が出る可能性があります。
現在ご利用いただいおりますYellowfinのバージョンをできるだけ早く、影響を受けないバージョンへアップグレードしていただくようお願いします。
詳細につきましては【重要】セキュリティ情報をご確認ください。

 

V9.9からフィルター値を含むダッシュボードの埋め込みタグおよびURLリンクの仕様が一部変更となりました。
また、本変更との互換性保持のため、 既存ダッシュボードの埋め込みタグとURLリンクにつきましても再生成されますので、該当機能をご利用の方はアップデート後に必ずご確認をお願いいたします。
詳細につきましてはv9.9の改修による影響について：ダッシュボードの埋め込みタグおよびダッシュボードのURLリンクをご確認ください。

 

Highlights（ハイライト）:

  • Modified some elements of the installer so that it can be white-labeled more easily. This includes using variables for replacing the word Yellowfin with the name of the application in text entries and genericising some image names to not contain the word Yellowfin.
  • Added the ability to display custom text labels against non-numeric data points in a Google Map.
  • Enhanced the bulk user creation REST endpoint, POST /api/admin/users, to respond with all users in the request even if they were already created in the system. This behavior is based on a new request parameter "includeExistingUsersInResponse". The default behavior of the endpoint is to fail, and to return only the newly created users when the request parameter "ignoreDuplicates" is true.
  • Added support for custom functions to have different internal implementations to cater for syntax differences across different database types. This allows implementations to use DB specific syntax, such as Group_concat for MYSQL and string_agg for SQLServer.
  • Implemented functionality to hide the "Customise Fields" (previously "Optional Fields") button for both reports and dashboards.
  • Added a usage parameter for data sources using Client Substitution to disable Cached Filter value refresh in Client Organisations when refreshing the filter values in the Primary Org.
  • When adding a report to a story, the display will now default to the canvas view—if one exists—instead of defaulting to the first chart (by index). This applies to reports added via both NLQ actions and the standard "Add Report" option.
  • Added an option to override Reference Codes defined in the Primary org in a Client org. This means that Client orgs can have their own unique set of references codes.
  • Added functionality to access PostgreSQL Foreign Tables in the View Builder.
  • Added a configuration option to customise the UI behaviour after cancelling long-running reports. Users can now remain on the Report Page to adjust filters and retry, instead of being redirected to the Browse Page automatically.
  • Added functionality to clear the entire CachedFilterCache via the REST API.
  • Added the ability to use Date-Extracted (DATEHIERARCHY) fields in calculated fields.
  • Added support for using a templating language to modify SQL queries within Virtual Tables. This allows for basic conditional logic based on filter and parameter values, the injection of filter values into the query, and varying SQL based on which database type is in use.

 

Enhancements（機能強化・拡張）:

AI Features

  • Added the following models to the External Connection model drop down - allowing these models to be used by Yellowfin AI features. Older deprecated models have been removed from the dropdown list.
    • OpenAI ChatGPT 5.5
    • Microsoft Azure GPT 5.5
    • Microsoft Azure GPT 5.4
    • Microsoft Azure GPT 5.4 mini
    • Anthropic Claude Opus 4.7

Data Sources

  • Added a usage parameter for data sources using Client Substitution to disable Cached Filter value refresh in Client Organisations when refreshing the filter values in the Primary Org.
  • Added functionality to enable table aliases to be loaded from AS400 DB2i datasources in the view builder.
  • Added functionality to access PostgreSQL Foreign Tables in the View Builder.

Stories

  • When adding a report to a story, the display will now default to the canvas view—if one exists—instead of defaulting to the first chart (by index). This applies to reports added via both NLQ actions and the standard "Add Report" option.

Reports and Charts

  • Added the ability to display custom text labels against non-numeric data points in a Google Map.
  • Added support for custom functions to have different internal implementations to cater for syntax differences across different database types. This allows implementations to use DB specific syntax, such as Group_concat for MYSQL and string_agg for SQLServer.
  • Implemented functionality to hide the "Customise Fields" (previously "Optional Fields") button for both reports and dashboards.
  • Replaced deprecated Google Maps Markers with the new Advanced Markers.
  • Added the ability to configure whether totals are shown on tooltips for Layered and Stacked charts - this helps prevent the display of incompatible data being added together.
  • Added a configuration option to customise the UI behaviour after cancelling long-running reports. Users can now remain on the Report Page to adjust filters and retry, instead of being redirected to the Browse Page automatically.
  • Added the ability to use Date-Extracted (DATEHIERARCHY) fields in calculated fields.

Views

  • Automatically remove the "Clone" suffix from the View name when publishing a new version of the View.

Administration

  • Modified some elements of the installer so that it can be white-labeled more easily. This includes using variables for replacing the word Yellowfin with the name of the application in text entries and genericising some image names to not contain the word Yellowfin.
  • Previously, updates to References Codes as part of a Yellowfin upgrade, would not work for some Docker deployments as the Upgrader process was not always run. These updates have now been moved to a migration job executed on application startup. This means that those docker deployments will now receive updates to References Codes.
  • Added an option to override Reference Codes defined in the Primary org in a Client org. This means that Client orgs can have their own unique set of reference codes.
  • Added the ability to search for users within the User Group Management screen.
  • Added a ServletFilter that can enforce only valid Host Headers, and thus prevent Host Header Poisoning.
  • Added a new configuration option - "HIDEMAILSHARELINK" that can hide the share link option. This disables the ability for a user to share an object as a url link.

REST API

  • Enhanced the bulk user creation REST endpoint, POST /api/admin/users, to respond with all users in the request even if they were already created in the system. This behavior is based on a new request parameter "includeExistingUsersInResponse". The default behavior of the endpoint is to fail, and to return only the newly created users when request parameter "ignoreDuplicates" is true.
  • Updated the content listing endpoints GET /reports, GET /dashboards, GET /presentations, GET /themes, GET /stories, GET /content and GET /admin/users/{id}/content to support filtering by CategoryUUID and parentCategoryUUID.
  • Improved REST API security by restricting login token creation to admin users only.
  • Added functionality to clear the entire CachedFilterCache via the REST API.
  • Enhanced the POST /api/admin/users REST API to return a 409 CONFLICT response when attempting to register a user with an email address that already exists.

Open Source Dependencies

  • Upgraded Apache Tomcat to 9.0.118.
  • Upgraded the Postgresql JDBC driver to 42.7.11.
  • Upgraded the Microsoft SQL JDBC drivers to 13.4.0.
  • Upgraded Log4j to 2.26.0.
  • Upgraded Apache Commons Lang3 to 3.20.0.
  • Upgraded Bouncy Castle to 1.84.
  • Upgraded JSch to 2.28.2.
  • Upgraded Axios to 1.16.0.
  • Upgraded Apache POI to 5.5.1.
  • Upgraded Snappy to 0.5.
  • Upgraded JSoup to 1.15.4.
  • Upgraded underscore.js to 1.13.8.
  • Upgraded Spring Framework to 5.3.51.
  • Upgraded Liquibase to 4.33.0.

 

Resolved Issues（不具合改修）:

AI Features

  • Resolved an issue where some charts on NLQ did not display correctly and showed a "your chart canvas is too small to render this chart" error message.

Data Connectivity

  • Resolved an issue where client source substitution was unstable when using Athena datasources.

Views

  • Resolved an issue where bookmarked drills were not working after the View was cloned.
  • Resolved an issue where date category separators were not shown in the View's automation settings popup.
  • Resolved an issue where schedules for View caching were not being saved correctly.
  • Resolved an issue where the Data Preview page in the View Builder would not update properly after clicking the "Refresh Now" button.
  • Resolved an issue where users could still select writable data sources when the "Advanced CSV Functionality" role was disabled.

Reports and Charts

  • Resolved an issue where certain reports would never complete and would continue spinning and polling indefinitely.
  • Resolved an issue where access filters were still applied to queries after they had been removed from the View.
  • Resolved an issue where the filter popup for setting a custom timestamp range could remain open or in an inconsistent state after the user clicked away from the popup.
  • Resolved an issue where some Freehand SQL queries running against RedShift would generate a "transaction is read-only" error message.
  • Resolved an issue where UNION subqueries containing calculated fields failed after upgrading to version 9.17.0.3.
  • Resolved an issue with certain reports containing append subqueries, where the OR operator was incorrectly converted to AND.
  • Resolved an issue where date slider controls would sometimes show the wrong values.
  • Resolved an issue where calculated fields were not working correctly in some reports after upgrading to version 9.17.0.2.

Dashboards and Presentations

  • Resolved an issue where adding a report to the associated reports list prevented the report from being placed on the Dashboard canvas.
  • Resolved an issue where Dashboards actions were not working correctly for Numeric charts

Broadcasts

  • Resolved an issue where Broadcasts sending a link and using saved filter values were failing.

XLS/DOCX/PDF/CSV Export

  • Resolved an issue where the “Append Timestamp” option for Exports could not be set to “None” when using an Oracle database.
  • Resolved an issue where a gap appeared between a line and its associated arrowhead when exported to PDF using Fast PDF.
  • Resolved an issue where the Secure PDF settings were not being applied in FTP broadcasts using the Fast PDF exporter.
  • Previously, when performing an export of all of the report components on a dashboard, any duplicate copy of a report would be suppressed, even if each copy of the report was linked to different filters. This has been resolved and all copies of a report will be exported unless the filter set is identical - and thus the data is identical.
  • Resolved an issue where exporting reports to DOCX resulted in incomplete data.
  • Resolved an issue where transparent background logos showed a black background during an export.

User Interface

  • Resolved an issue where Folder names were not correctly translated in the Browse page.
  • Removed dependency on the deprecated page unload event, which sometimes resulted in the "Return to Dashboard" or other content button appearing unexpectedly.

Import/Export

  • Resolved an issue where importing content with Skip Datasource Validation enabled prevented OnDemand filters from being initialised properly.

Administration

  • Resolved an issue where editing or deleting a Reference Code did not immediately update the screen.
  • Resolved an issue where connection information was no longer being logged to source.log.
  • Resolved an issue where the Reference Code name was not displayed correctly when a value from another Reference Code was entered in the ‘Name’ field.
  • Enhanced logging so that reports that fail are easier to identify. If there is an out-of-memory event during report execution, the report results will be purged from memory in an attempt to allow the system to log information about the failed report.
  • Resolved an issue where the Reference Code Usage tab displayed deleted views.
  • Added a system check to prevent the creation of Users with duplicate email IDs. A database cleanup job has also been added to remove duplicate user records upon upgrade.
  • Resolved an issue where Yellowfin could not start up after upgrading to 9.17.0 due to the upgrade causing issues in the Yellowfin database.
  • Resolved an issue where database migrations were not working on Java 24 or later.

Data Transformation

  • Resolved an issue where certain ETL text fields were automatically being converted into boolean types.

REST API

  • Resolved an issue where REST APIs did not correctly handle user group visibility across different organisation contexts.