Anzo 5.3 Releases

To view the release notes for an Anzo 5.3 version, select the version from the list below. The release notes for each version describe the product changes from the previous version.

If you are upgrading to 5.3.x from a previous release, see Anzo 5.3 Installation and Upgrade Notes for important notes about upgrades to Anzo platform components.

Anzo Version 5.3.5

This section describes the issues that were fixed in Anzo Version 5.3.5.

Import was Slow for an Artifact with Many Versions

Version 5.3.5 significantly improves performance when an artifact with several versions is imported.

Could not Export a Version from the UI Because the Export Button Remained Disabled

Version 5.3.5 corrects an issue where the Related Entities table on the Versions tab could not be displayed. Since the table did not finish loading, the Export button was never enabled on the screen and the user could not export the version.

Could not Version and Export an Ontology

Version 5.3.5 fixes an issue that prevented a valid version from being created for particular ontology. As a result, the ontology could not be exported.

Spark ETL Pipeline Generated Incorrect Values for an Impala Data Source

Version 5.3.5 includes an updated sdi-deployment-utils-5.1.1.0.jar file that resolves an issue where incorrect values were output in the target after a Spark pipeline was run to ingest data from an Impala data source.

A User Could not Access a Migration Package That They Created

Version 5.3.5 resolves an issue where a user with the Manage Migration Packages permission was unable to access a migration package that they created earlier.

Unrelated Artifacts were Included in a Migration Package for a Mapping

Version 5.3.5 corrects an issue that caused unrelated entities to be included in a migration package for a mapping artifact.

Anzo Version 5.3.4

This section describes the new features and issues that were fixed in Anzo Version 5.3.4.

New Features and Improvements

Fixes

Patches for Security Vulnerabilities

Added a Step for Enriching Elasticsearch JSON Documents before Indexing

Version 5.3.4 adds a JSON Update Step that can be used to enrich Elasticsearch JSON documents with data from a Graphmart before the documents are indexed. For more information and instructions on enabling the step, refer to your Cambridge Semantics Customer Success manager.

Removed the "(g)" Suffix from Dashboard Labels

Version 5.3.4 removes the "(g)" suffix from the labels on Graphmart Dashboards since it is no longer necessary to distinguish between dashboard types.

Prevented Debug Messages from Being Displayed in Browser Developer Tools

Version 5.3.4 prevents debug messages from being displayed when browser developer tools are used to inspect the Anzo and Admin applications. In Version 5.3.4, only errors and warnings are displayed in the browser console.

Added Absolute Values to Percentages in Data Profile Results

To make Data Profile results easier to interpret, Version 5.3.4 displays the absolute value next to the percentage value for all metrics that return percentages.

Added Support for Substring Searches on Profiled Data

Version 5.3.4 adds support for searching for substrings in Data Profile results.

Replaced NaN Values with a Message when a Metric is not Applicable for Profiled Data

Version 5.3.4 returns the message "Not an Applicable Metric" in place of "NaN" in Data Profile results if a metric is not applicable for a property.

Made the Data Models Field Optional for Elasticsearch Indexing Steps

Version 5.3.4 changes the Data Models field to an optional field for Elasticsearch Indexing Steps. It was previously required but not needed.

Streamlined the Configuration of Settings with Checkboxes

In previous releases, certain settings, such as the Advanced settings for Unstructured Pipelines, required the user to first make the setting editable, then select or deselect the checkbox, and then save the change. Version 5.3.4 streamlines the configuration of these types of settings by allowing them to be changed with a single click.

Added the Ability to Denormalize All Results for Data on Demand Endpoints

Version 5.3.4 adds the ability to denormalize the results for all multi-valued properties in both Auto-Generated and Custom Data on Demand endpoints. For JSON, XML, and CSV output formats, denormalization expands the multi-valued properties into new rows so that they can be viewed in BI tools that do not support arrays or multi-valued properties. For more information, see Creating Data on Demand Endpoints in the Anzo Deployment & User Guide.

Pipeline Sharing Tab was Read-Only for a User with Modify Permissions

Version 5.3.4 corrects an issue that caused the Pipeline Sharing tab to be read-only for a user even though that user was assigned Modify permissions.

New Regex Annotator Used the Regex Rule from a Preexisting Annotator

When a user added a new Regex Annotator to an Unstructured Pipeline that already included a Regex Annotator, the new annotator contained the Regex Rule from the existing annotator and the rule could not be changed. Version 5.3.4 resolves this issue.

Workflow Manager Ignored the Load Timeout Setting

Version 5.3.4 fixes an issue that caused the Load Timeout setting to be ignored when Workflows were run. Workflows could continue to run longer than the configured timeout value.

No Error Returned when an Elasticsearch Indexing Step Failed

Version 5.3.4 resolves an issue that made an Elasticsearch Indexing Step appear to succeed when in reality the index update failed.

This issue is resolved by an update to the AnzoGraph GDI extension. If you do not upgrade AnzoGraph and require this fix, replace your existing GDI .jar file with the latest version. For more information, see Update the AnzoGraph Graph Data Interface Extension.

Unexpected Results for Query Step that Modified Data in Another Layer

When a Query Step ran a query to delete data from another layer, the data failed to be deleted even though the step was correctly configured to operate against the other layer. Version 5.3.4 resolves this issue.

Query with OPTIONAL and MINUS Clauses Returned Unexpected Results

Version 5.3.4 corrects an issue where running a query that included an OPTIONAL clause with a MINUS clause returned only the results that were specified in the MINUS clause rather than all results except for the data referenced in the MINUS clause.

Network Navigator Legend Remained Out of View on a Small Screen

Version 5.3.4 resolves an issue where users with small screen sizes were unable to scroll to the legend for a Network View.

Network Navigator Legend did not Show Conditional Nodes

Version 5.3.4 fixes an issue that caused conditional nodes not to be displayed in the Network Navigator legend.

Resized Panels in the User Interface Retained Their Size after Closing

Version 5.3.4 corrects an issue that caused panels in the user interface to take up the same space but remain blank if they were resized before they were closed.

Could not Save the Temporary Edition after Editing the Default Edition

Version 5.3.4 fixes an issue where a user could not save the temporary edition that was created after editing the default edition in a Graphmart.

Refresh Required after Importing New Database Schema

When a user imported a new schema for a database data source, the schema was not displayed in the schema list on the Tables screen until the user refreshed the browser. Version 5.3.4 corrects this issue.

Unable to Update Image Versions in Cloud Locations for AKS Clusters

When using an Azure AKS cluster, a user was unable to update a Cloud Location to change the image versions for the Anzo Agent, Anzo Unstructured operator, Anzo Unstructured Leader, and Anzo Unstructured Worker images. Version 5.3.4 resolves this issue.

Unknown Function Error for Query with xsd:dateTime Function

When a Data Layer query included the xsd:dateTime function instead of the DATETIME function, the query failed with an "Unknown Function" error. Version 5.3.4 resolves this issue so that xsd:dateTime can be used in queries.

The Query Builder did not Recognize the UNBOUND Function

Version 5.3.4 adds support for using the UNBOUND function in Query Builder queries.

Error with RDF Handler when Exporting a Version of a Model

Version 5.3.4 resolves an issue that caused an export of a version of a model to fail with an "Error with RDF Handler" message.

Searching Data Profile Results Returned the Model Name if the Search Value was not Found

Version 5.3.4 corrects an issue that caused the model name to be displayed if a user searched Data Profile results for a value that did not exist. Anzo now returns a "No results found" message.

The Value Present Metric was not Displayed for DateTime Properties in Profiled Data

Version 5.3.4 resolves an issue that prevented the Value Present metric from being displayed in Data Profile results for properties with dateTime data types.

NPE when an Auto-Generated Data on Demand Endpoint was Refreshed

Version 5.3.4 fixes an issue that caused an Auto-Generated Data on Demand Endpoint to return a null pointer exception when it was refreshed.

Failure to Create an OData Schema Resulted in an NPE

Version 5.3.4 corrects an issue that caused OData Schema generation to fail for a Data on Demand endpoint. As a result, a null pointer exception occurred with the endpoint was accessed.

Auto-Generated Data on Demand Endpoint Appeared Disabled after Graphmart Refresh

Version 5.3.4 resolves an issue that caused an Auto-Generated Data on Demand Endpoint to appear offline after the Graphmart was refreshed multiple times.

Data on Demand $expand Results were not Returned when Server Side Paging was Enabled

Version 5.3.4 corrects an issue that prevented expand values from being returned from Data on Demand endpoints if OData server side paging was enabled.

Incorrect Results for OData Query with Nested $expands

Version 5.3.4 resolves an issue that caused incorrect results to be returned for an OData query that included multiple $expands for properties of the same type. Only the values for one of the nested expands was returned.

Detached Properties in a Model were not Selectable in Custom Data on Demand Endpoints

Version 5.3.4 fixes an issue that prevented detached properties from being displayed in Custom Data on Demand endpoints.

Incorrect Data on Demand Results if Multi-Valued Property Values were not Distinct

Version 5.3.4 fixes an issue that caused incorrect results to be returned from a Data on Demand endpoint if the resultset included multi-valued properties for which the values were not distinct.

Sparkler Jobs Failed for Database Sources when AES was Enabled

Version 5.3.4 resolves an issue that caused Sparkler jobs to fail for database data sources when Advanced Encryption Standard (AES) was enabled.

Could not Search for Text when Choosing Formula Values in Custom Data on Demand Endpoints

When creating a Custom Data on Demand endpoint, the property and function selection drop-down list covered the Formula Value field, making it impossible to type search values into the field.

Drill Down Lens Breadcrumbs Rendered Incorrectly if they Included Brackets

Version 5.3.4 corrects an issue that caused breadcrumbs with brackets ([ ]) to render incorrectly in Drill Down lenses in the Hi-Res Analytics application.

The Query Manager Showed Script Errors when used with Network Navigator Dashboards

Version 5.3.4 resolves an issue where the Hi-Res Analytics Query Manager panel showed script errors when it was used with a Network Navigator dashboard.

Subclasses Appeared to be Disabled in Network Navigator Filters

Version 5.3.4 fixes an issue where the classes and subclasses in a Network Navigator dashboard filter were grayed out and appeared as if they could not be selected even though they were enabled.

Dashboard Query Ran Against the Wrong Data Layer

Version 5.3.4 resolves an issue that caused the Hi-Res Analytics application to run queries against the incorrect Data Layer when one layer was selected in the Data Layers panel on a dashboard.

Query-Driven Templated Step with Many Iterations Caused Anzo to Stop Responding

Version 5.3.4 resolves an issue that caused Anzo to become unresponsive when a Query-Driven Templated Step with ~70k iterations was run. The unresponsiveness was caused by the significant amount of logging that was produced during the query iterations.

ASK Queries did not Return Results from the SPARQL Endpoint

Version 5.3.4 resolves an issue that prevented ASK queries from returning a response when they were run against the SPARQL endpoint.

Irrelevant Error Returned for Edition Creation Failure

Version 5.3.4 corrects an issue that caused Anzo to return an "Edition with title <title> already exists" error message if edition creation failed regardless of the real reason the edition could not be created.

Anzo Version 5.3.3

This section describes the new features and issues that were fixed in Anzo Version 5.3.3.

New Features

Fixes

Added the Ability to Integrate Graph Data with Elasticsearch Indexes

Version 5.3.3 introduces the ability to load classes and properties from knowledge graphs into Elasticsearch indexes. Creating graph-based indexes gives you the scalability and performance to execute free-text and pattern searches on Graphmart data. You can also use graph-based indexes to augment document-based indexes created by unstructured pipelines. For more information, see Creating an Elasticsearch Index from Graphmart Data in the Anzo Deployment & User Guide.

Added Support for AES Encryption Types

Version 5.3.3 adds support for the Advanced Encryption Standard (AES). For more information and instructions on enabling AES, refer to your Cambridge Semantics Customer Success manager.

Could Not Activate a Graphmart after Adding a Data on Demand Endpoint

Version 5.3.3 resolves an issue that prevented a Graphmart from being activated after a Data on Demand Endpoint was created for it.

Incremental ETL Jobs Failed for an Oracle Data Source

Version 5.3.3 resolves an issue where Anzo incorrectly added double quotes around an integer Increment Value in a schema query. The quotes caused an error when run against the data source and the ETL jobs failed.

Anzo Version 5.3.2

This section describes the improvements and issues that were fixed in Anzo Version 5.3.2.

Improvements

Other Changes and Fixes

Patches for Security Vulnerabilities

Enhanced the Generated Direct Data Loading Queries to Improve Readability

Version 5.3.2 includes improvements to the auto-generated queries that are created when the Direct Data Loading workflow is used to onboard a Data Source. The enhancements include reducing the use of blank nodes and improving the formatting to make the generated queries easier to read by users. For more information about the Direct Data Loading workflow, see Directly Loading Data Sources via Graphmarts in the Anzo Deployment & User Guide.

Improved Usability of Data Profiling Metrics

Version 5.3.2 includes enhancements to the usability and layout of the Metrics that are returned when a Data Source, Dataset, or Graphmart is profiled.

Improved the Error Message Returned if LDAP Authentication Fails

Version 5.3.2 improves the error message that is returned if LDAP authentication fails by making it more specific about the cause of the failure. For example, if an incorrect username or password is entered, Anzo returns a message such as, "Invalid username and password combination."

Added an Alert if a User Attempts to Delete a Data Source that is Directly Loaded in a Graphmart

Version 5.3.2 adds an alert that is returned if a user attempts to delete a Data Source that is being directly loaded in a Graphmart.

Added the Ability to Delete Components from Datasets as well as from Editions

Version 5.3.2 adds the ability to delete Data Components from both the Managed Edition as well as Saved Editions. When browsing an Edition, the menu in the Data Components Actions column has the new options, Remove Component From Edition and Remove Component From Dataset (shown in the image below).

Selecting Remove Component From Edition is similar to the functionality that was introduced in Version 5.3.0. This option deletes the Component from the Edition but it does not remove the Component from the Dataset. Selecting the Remove Component From Dataset option deletes the Component from the Edition and deletes that Component's data from the file system.

Added the Option to Create a Graphmart Load Service Task that Refreshes a Graphmart

Version 5.3.2 adds to the Workflow Manager the ability to create a Graphmart Load Service Task that refreshes a Graphmart rather than reloads it. For information about the Graphmart Load Service Task, see Adding a Task that Refreshes or Reloads a Graphmart in the Anzo Deployment & User Guide.

Added the Ability to Limit the Number of anzo_full Logs and Total Size of all anzo_* Logs

Version 5.3.2 adds maxHistory and totalSizeCap properties to the Pax Logging SLF4j Listener Service. These properties can be set to limit the number of anzo_full.log files that are retained on disk and place a total size limit on all anzo_* logs. For more information, see Limiting the Size and Number of anzo_full Logs in the Anzo Deployment & User Guide.

Improved Usability of the Suggested Keys Tab in Data Sources

Version 5.3.2 improves the layout and usability of the Suggested Keys interface in the Anzo application by moving the Actions buttons into menu items under a menu in the Actions column.

Improved the Formula Selector for Custom Data on Demand Endpoints

Version 5.3.2 improves the formula selector to make it easier to select or type functions when adding a column to a custom Data on Demand endpoint.

Improved Logging for Custom Data on Demand Endpoints

Version 5.3.2 increases logging for Custom Data on Demand Endpoints to aid in debugging when errors are encountered.

Set the Dataset Format to TTL.GZ by Default

Version 5.3.2 automatically sets the Dataset Format to ttl.gz by default. This field is required when creating a Dataset and was previously left blank even though the only valid values are ttl or ttl.gz.

Improved Usability when Directly Loading Multiple Data Sources

Version 5.3.2 adds an explanation to the Ontology URI field under the Advanced options that are available when using the Direct Data Loading workflow to onboard multiple Data Sources at once. The text informs the user that a custom Ontology URI cannot be set when multiple Data Sources are selected.

An Incremental ETL Job Failed on the Second Run

Version 5.3.2 corrects an issue that caused the second run of an incremental job for a database Data Source to fail when the increment Value set for the schema matched the number of columns in the table.

An Incremental ETL Job Failed when the Counter was Manually Changed

Version 5.3.2 resolves an issue where an incremental job failed after the job was edited to change the counter Value.

The Counter Value Failed to Update after an Incremental Job was Run

Version 5.3.2 resolves in issue that caused the counter Value on an incremental job to fail to update after the job was run.

The Change was not Saved after Updating the Increment Value in a Schema Query

Version 5.3.2 fixes an issue that caused the change not to be saved when the Value was updated for an Incremental Schema query.

A Dashboard Dialog Type Hyperlink Failed to Load

Version 5.3.2 resolves an issue that occurred when a Dashboard included a hyperlinked column that was configured to open in a Dialog. In this case, the data would not load when the link was clicked unless the lens Designer was opened first.

Artifacts in a Cloned Graphmart Inherited Permissions from the Original Graphmart

Version 5.3.2 fixes an issue that caused the Data Layers and Steps in a cloned Graphmart to incorrectly inherit their permissions from the original Graphmart instead of the new copy of the Graphmart.

Artifacts in a Cloned Graphmart were Missing Permissions

Version 5.3.2 fixes an issue where the Data Layers and Steps in a cloned Graphmart could be missing permissions instead of inheriting them from the new copy of the Graphmart.

Related Entities were not Displayed after an Artifact was Imported

Version 5.3.2 increases to 400 the maximum number of related entities that can be shown in the Anzo application when importing, exporting, or creating versions of artifacts. Previously the limit was much lower and some users received a "Too many related entities found, unable to show the list" message.

The Create New Folder Option was Missing from the Create Dataset Dialog Box

Version 5.3.2 adds the Create New Folder option to the user interface when a new Dataset is being created.

Roles Created in Anzo were Included in the Groups List

Version 5.3.2 resolves an issue that caused Roles that were created in Anzo to be displayed in the list of Groups in the Anzo application.

A User was Unable to Delete an Edition

Version 5.3.2 corrects an issue where a user was unable to delete a Saved Edition due to an "Invalid Arguments" error.

Could Not Cancel Adding a Column to a Custom Data on Demand Endpoint

Version 5.3.2 corrects an issue where the Cancel button failed to cancel the action after a user clicked Add Column to add a column to a custom Data on Demand endpoint. The browser had to be refreshed in order to cancel the column addition.

A Custom Data on Demand Endpoint Formatted Date Values Differently for CSV Output

Version 5.3.2 fixes an issue that caused a Custom Data on Demand endpoint to format date values differently than other return types. For example, the preview of the values showed "2021-02-01" while CSV output returned "Mon Feb 01 00:00:00 GMT 2021."

The "Results Have Changed" Message Failed to Display for Dashboards

Version 5.3.2 resolves an issue that caused the "Results Have Changed" alert on Hi-Res Analytics Dashboards to be displayed only briefly, rather than remaining onscreen until the user updated the Dashboard.

The Dashboard Graphmart Panel Did Not Display the Graphmart Name

Version 5.3.2 fixes an issue that caused the Hi-Res Analytics application to fail to display the Graphmart name in the Graphmart panel on a Graphmart Dashboard.

A Migration Packages Configuration Field Tooltip Blocked Choosing an Option for the Field

Version 5.3.2 resolves a problem that prevented a tooltip from closing when a user clicked away. The tooltip blocked the user from choosing a value for the field.

A Network Navigator View with Many Resources Did Not Load when Opened From a Table Lens

Version 5.3.2 corrects an issue that caused a Network Navigator View to fail to load when it was opened from a grouped row in a Table lens and a large amount of resources were included.

An Empty Model was Generated when it was Overwritten by the Direct Data Loading Workflow

Version 5.3.2 resolves an issue that caused a Model to become blank after the original Model was overwritten using the Direct Data Loading workflow.

Deleting a Workflow Resulted in Several Exceptions

Version 5.3.2 fixes an issue that caused multiple exceptions to be logged when a Workflow was deleted. Now, Workflows are deregistered when they are deleted, preventing the exceptions.

Anzo Version 5.3.1

This section describes the improvements and issues that were fixed in Anzo Version 5.3.1.

Added Queries from Query-Driven Templated Steps to the Activity Log

In Version 5.3.1, all of the queries that are executed as part of Query-Driven Templated Step processing are now logged in the Activity Log.

Added Support for Elasticsearch Versions 7.11 and 7.12

Version 5.3.1 adds support for Elasticsearch Versions 7.11 and 7.12. Anzo Unstructured now supports Elasticsearch Versions 7.10.2 through 7.12.0.

Added Sparkler Support for Schema Queries with Tab Characters

Version 5.3.1 enhances the Sparkler ETL engine to support Schema queries that include tabs. Previously, the queries failed if the SQL included tabs.

Added Support for Special Characters in Usernames

Version 5.3.1 adds support for including special characters in usernames.

Added Registry Statements Validation on Migration Package Import if Export Included Them

Version 5.3.1 adds validation that will fail a Migration Package import if the export included registry statements and the statements are missing from the import file.

Improved Formula Selector for Custom Data on Demand Endpoints

Version 5.3.1 improves the formula selector to make it easier to select or type functions when adding a column to a custom Data on Demand endpoint.

Modified Migration Import CLI Command to Enforce ACL Validation

Version 5.3.1 modifies the migrationImport Anzo Admin CLI command so that when the -sk (skip validation) option is specified, validation is still performed on ACL templates to ensure that all placeholder ACL values are replaced in the template.

Numeric Values Incorrectly Shown on Chart Axis with String Values

Version 5.3.1 corrects an issue that caused a Chart lens using string values on the Y-Axis to incorrectly display numeric values above and below the range of strings.

Column Name Replaced Instance Data when Column Included CONCAT in Mapping

Version 5.3.1 corrects an issue that could cause the column name to appear in the Dataset in place of the instance data if the column included the CONCAT function in the Mapping and Spark was used to publish the job.

Long Artifact Titles Moved Action Buttons Off Screen

Version 5.3.1 corrects an issue that pushed the action buttons, like Ingest, Create Dashboard, and Profile Data, far to the right of the screen when artifact titles were very long. Long titles are now shortened with ellipses (...).

Unknown or Inaccessible Graphs Error During Graphmart Refresh

Version 5.3.1 resolves an issue that caused Anzo to return an "Unknown or inaccessible graphs included in query..." error when a sysadmin or Admin user refreshed a Graphmart.

Incremental File Ingestion Failed if File Name Had Spaces

Version 5.3.1 corrects an issue that caused a Sparkler job to fail if incremental processing was enabled and the file name had spaces in it.

JavaScript Error when Canceling All Queries from System Query Audit

Version 5.3.1 corrects a JavaScript error that could occur when a user clicked the Cancel All button on the Inflight Queries tab in the System Query Audit log.

Sparkler Did Not Limit Max Output File Size

Version 5.3.1 sets the default maximum file size limit of 100MB for files that are output by the Sparkler ETL engine. This default matches the default limit for the Spark ETL engine.

LDAP UUID Shown as User Name for New Workflow

Version 5.3.1 corrects an issue that caused the LDAP UUID to be shown as the Creator rather than the username if a user other than the sysadmin user created a new Workflow.

Server Connection Lost when Downloading an X-Ray from the Dynamic Resources Log

Version 5.3.1 resolves an issue that could cause the Administration application to lose the connection to the Anzo server when an X-Ray was downloaded from the Dynamic Resources log.

Graphmart Explore Tab Did Not Show Subclass Connections from Different Model

Version 5.3.1 corrects an issue that caused subclass connections not to be rendered on the Graphmart Explore tab if the connections were to another Model.

Could Not Cancel Adding Column to Custom Data on Demand Endpoint

Version 5.3.1 corrects an issue where the Cancel button failed to cancel the action after a user clicked Add Column to add a column to a custom Data on Demand endpoint. The browser had to be refreshed in order to cancel the column addition.

Sparkler Jobs Seemed to Run Indefinitely When Ingesting Files Incrementally

Version 5.3.1 corrects an issue where Sparkler continued to send job status updates to Anzo after incremental file ingestion jobs completed. As a result, the pipeline seemed to run indefinitely.

Unstructured Pipeline Ran Indefinitely if a Document Errored Out

Version 5.3.1 resolves an error that could cause an Unstructured Pipeline to run indefinitely if one or more documents had an error.

Anzo Version 5.3.0

This section describes the new features and changes to existing components that are introduced in Anzo Version 5.3.0.

New Features

Improvements to Existing Features

Improvements for Kubernetes-Based Deployments

Other Changes and Fixes

Patches for Security Vulnerabilities

Direct Data Load Automated Ingestion Workflow

Version 5.3.0 introduces a new type of automated ingestion workflow that follows the extract, load, and transform (ELT) process. In the ELT workflow, Data Sources are loaded directly to AnzoGraph via Graphmarts. Data Layers are automatically generated to transform and blend the data in memory to an analytics-ready knowledge graph. The AnzoGraph Graph Data Interface (GDI) Java plugin is used to connect to the Data Sources, create a Model, and generate the Data Layer queries. For more information, see Directly Loading Data Sources via Graphmarts in the Anzo Deployment & User Guide.

Incremental Onboarding of CSV and Parquet Files

Version 5.3.0 adds the ability to onboard CSV and Parquet files incrementally. For details about the new feature, see Configuring a CSV or Parquet Data Source for Incremental Processing in the Anzo Deployment & User Guide.

Migration Packages Administrative Tool

Version 5.3.0 introduces the Migration Packages tool for administrators. When migrating artifacts between environments, administrators can now perform a bulk export (and import) by assembling a Migration Package that includes any number and type of artifacts and their related entities. For more information, see Migration Packages in the Anzo Deployment & User Guide.

Workflow Manager Administrative Tool

Version 5.3.0 introduces the Workflow Manager tool for administrators. The Workflow Manager is used to manage tasks such as Structured or Unstructured Pipeline runs and Graphmart loads. Workflows can be triggered from the Anzo Admin CLI, and the CLI call can be automated by setting up cron jobs.

Hi-Res Analytics Query Manager

Version 5.3.0 introduces the Hi-Res Analytics Query Manager. The Query Manager is available from the Help menu on the main toolbar. When selected, the Query Manager panel is displayed on the right side of the screen. The panel has two tabs: an Active Queries tab that lists any queries that are in progress, and a Query History tab that shows completed queries.

The Active Queries tab shows the elapsed time for each query and includes the option to cancel queries. For example:

The Query History tab shows completed queries and statistics such as the total query runtime, start time, and end time. For example:

Added Hi-Res Analytics Features for Improved Multi-User Experience

Version 5.3.0 adds the following new features that improve usability when multiple users view and change the same Hi-Res Analytics Dashboard.

  • Dashboard changes are automatically cached for the entire user session. If a user closes the Hi-Res Analytics application browser tab, their Dashboard changes are preserved as long as they are still logged in to Anzo. Re-opening the browser will show the in-progress changes.
  • If multiple users are viewing a Dashboard and one user saves changes to it, the other users are alerted to the change and have the option to see the changes or continue to view the Dashboard as-is.
  • If a user makes changes to a Dashboard but wants to revert those changes, there is a new Reset to Original Dashboard option in the Dashboard menu on the main toolbar. Selecting that option resets the Dashboard view to the last saved version.

Added the Ability to Show and Hide Columns in Tables Lenses

Version 5.3.0 adds a Manage Columns button to the object toolbar for Table lenses in Hi-res Analytics Dashboards. The Manage Columns dialog box enables users to show or hide columns in the Table without having to use the lens Designer.

Added the Ability to Create Custom Anzo Groups

Version 5.3.0 introduces the ability to add new Groups that are internal to Anzo. Previously Groups could only be synced to Anzo from a directory server.

Added the Ability to Set the Default Anzo Data Store

Version 5.3.0 adds the option to designate the default Anzo Data Store for the server. The new option is available in Server Settings in the Administration application. For more information, see Configure the Default Anzo Data Store in the Anzo Deployment & User Guide.

Added View of System Functions and Extensions in AnzoGraph

Version 5.3.0 introduces the Extensions tab that is available when viewing the AnzoGraph connection configuration in the Administration application. The Extensions tab lists all of the builtin functions and extensions that are available in your version of AnzoGraph.

Added Option to Remove Components from the Managed Edition

In Version 5.3.0, users have the ability to remove components from a Dataset's Managed Edition. When viewing components in the Managed Edition, there is now a Remove option in the Actions column.

Enhanced Export Steps to Automatically Create New Editions

In Version 5.3.0, Export Steps automatically create new Dataset Editions when they are run. A new Do Not Create New Edition in Dataset on Export setting can be enabled to disable the creation of new Editions.

If Do Not Create New Edition in Dataset on Export is enabled and Overwrite FLDS is disabled, Editions will not be created but a new Component will be added to the Managed Edition each time the Export Step is run. To limit the number of Components that can be created when Do Not Create New Edition in Dataset on Export is enabled and Overwrite FLDS is disabled, you can specify the maximum number of Components to create in the Maximum Number of Components in Edition setting.

In addition to the changes described above, Version 5.3.0 also adds the ability to rename the Components that are created by Export Steps.

Added Option to Normalize DNs when Configuring an LDAP Connection

Version 5.3.0 adds a Normalize LDAP DNs setting to the Directory server configuration screen in the Administration application. To ensure that duplicate user accounts are not created in Anzo if an LDAP distinguished name has both a lowercase and uppercase version, you can configure the system to normalize distinguished name strings so that values that differ only in capitalization are treated as the same value.

Added Option to Clear All Search Filters

Version 5.3.0 adds a Clear All Filters button to the search panel in the Anzo and Administration applications so that users can easily clear their filter selections.

Added Support for Listing Multiple Load File Paths in a Single Query

Version 5.3.0 adds support for parallel loading of multiple Dataset Editions in a single LOAD statement. Specify multiple directory location URIs in a space-separated list. For example, the following LOAD query loads the files from two different directories in parallel:

LOAD <dir:/global/nfs/sales/completed_sales.ttl> <dir:/global/nfs/marketing/prospects.ttl> INTO GRAPH <http://anzograph.com/finance>

This functionality is supported with AnzoGraph Version 2.4.0 or later. When including multiple URIs in a single statement, each URI must have the same scheme, either file: or dir:. All URIs must also have the same file type extension, such as .ttl or.ttl.gz.

Added the Ability to Set a Limit on the Number of Historical Pipeline Runs to Store in the Journal

Version 5.3.0 adds a maxJobExecRecordsToKeep property to the Anzo Server SDI Service. This property is set to unlimited by default but can be modified to limit the number of historical pipeline runs to retain in the system journal.

Added the Ability to Exclude a Data Layer's Data from Graphmart Queries

In Version 5.3.0, the Data Layer Advanced Data Access setting By default, the data outputted by the layer is visible applies to all Graphmart queries, such as from the Query Builder and SPARQL endpoint, as well as Hi-Res Analytics dashboards. When the option is enabled (the default setting) the layer's data is included by default in queries against the Graphmart. When the option is disabled, the layer's data is excluded from Graphmart queries unless the layer URI is specifically mentioned in the query.

Added the Ability to Use Kerberos Authentication with the Sparkler ETL Engine

Version 5.3.0 adds support for configuring the Sparkler ETL Engine to use Kerberos authentication.

Added Environment and Version Tag to the Anzo Application Banner

To help users distinguish between Anzo applications in different environments, Version 5.3.0 adds an <environment> | <version> tag to the top of the Anzo application. For example:

The <environment> tag is the Versioning Environment value that is configured in Server Settings. For more information, see Configure the Versioning Environment in the Anzo Deployment & User Guide.

Improved Usability when Finding Connections in Network Navigator Dashboards

Version 5.3.0 improves the usability of the Find Connections feature in Network Navigator Dashboards by limiting the results to load 100 nodes at a time. After the first 100 nodes are added to the Network View, Anzo presents a dialog box that lists the total number of connections that were found and asks the user if they want load the next 100. This change prevents the dashboard from becoming unresponsive when finding connections for large datasets.

Improved Usability of Query Contexts for Data Layers and Steps

Version 5.3.0 enhances the user interface that is used to configure Query Contexts, which abstract sensitive connection information in queries. Previously, if a user selected an Anzo-generated Context Provider, it was not possible to easily review the variables to use as placeholders for that Context. In Version 5.3.0, selecting an Anzo-generated Context displays the variables onscreen. For more information, see Configuring Query Contexts for Data Source Connections in the Anzo Deployment & User Guide.

Enhanced Data Profile Metrics and Separated Them into Categories

In order to provide a more robust yet streamlined option for generating Data Source, Dataset, and Graphmart Data Profiles, Version 5.3.0 now groups the metrics into Primary and Advanced categories and allows users to choose which category to run.

Added the Ability to Create and Edit Tags on Models

Version 5.3.0 adds the Tags field to the Details tab in the Model editor so that tags can be added to Models.

Added the Ability to Associate a User with Multiple Default Groups

Version 5.3.0 adds the ability to assign multiple Default Groups to a user.

Improved Sorting for Default Access Policy Search Results

Version 5.3.0 improves the sorting of user and group search results for Default Access Policies.

Added Reminder to Re-Log-In when Permissions are Changed

In Version 5.3.0, if an administrator changes permissions, the following alert is displayed to remind the administrator to ask users to log out and back in: "Permissions saved successfully. Users will need to re-login to see changes."

Enabled System Event Logs By Default when Audit Logging is Enabled

Version 5.3.0 logs system events by default when the Audit Logs are enabled. In addition, starting and stopping bundles are now logged as system events. For information about the Audit Logs, see Enabling and Viewing Audit Logs in the Anzo Deployment & User Guide.

Allowed Users with Appropriate Pipeline Permissions to Cancel Running ETL Jobs

Version 5.3.0 enables users to cancel running jobs if they have the appropriate permissions on the Pipeline. Previously only the sysadmin user could cancel running jobs. For more information, see Canceling a Running Job in the Anzo Deployment & User Guide.

Updated the HTTP Client Interface to Support Alternate RDF Serialization Formats

Previously, when Add or Remove client operations were run against the HTTP client servlet, the request needed to be in Anzo's JSON RDF serialization format and headers were not supported. Version 5.3.0 enhances the HTTP client interface to add support for Add and Remove requests in all supported RDF serialization formats. Headers are now supported and specify the format for the request. In addition, a header can be used with Get requests to specify the format of the response.

Enhanced Query Step Types to Suggest URIs Based on Prefixes

In Version 5.3.0, Data Layer Steps that include queries and have Models associated with them have been enhanced to automatically suggest URIs when a user types a prefix.

Improved Performance for Unstructured Pipelines

To improve performance of Unstructured Pipelines, Version 5.3.0 reduces memory usage and increases performance when crawling large document sets. In addition, some of the default values for Pipeline parameters were adjusted to further improve performance.

Added the Option to Increase Logging Verbosity for Unstructured Pipelines

Version 5.3.0 adds a Diagnostic Logging setting for Unstructured Pipelines. Enabling this option runs the Pipeline in diagnostic mode, which captures debug-level logging statements for the duration of the Pipeline.

Improved Performance for Artifact Exports and Imports

Version 5.3.0 includes an improvement that significantly increases the performance of artifact exports and Imports.

Improved Sparkler ETL Engine Performance

Version 5.3.0 includes an enhancement that significantly improves the performance of Sparkler when converting data to RDF format.

Improved Handling of Dataset Upgrades

Version 5.3.0 adds extra validation to ensure that Datasets from earlier Anzo versions are properly upgraded when they are migrated or manually imported to the new version.

Improved the Data on Demand Endpoint Schema Builder

Version 5.3.0 reworks the Data on Demand schema builder to improve how the OData schema is built for the ontologies in the endpoint's Graphmart.

Made the Anzo CLI Register Command Extensible

Version 5.3.0 includes an enhancement that makes the register CLI command more extensible.

Added a Confirmation Message when Server License is Updated

Version 5.3.0 adds a confirmation message that indicates that the server license key was updated and prompts the user to restart the server to apply the change.

Improved the System Monitor Service to Prevent Repetition of Heap Dumps

Version 5.3.0 improves the System Monitor Service so that heap dumps are not repeatedly written for the same error. In Version 5.3.0, when the System Monitor Service is enabled, it will only take a heap dump the first time an error is seen. Subsequent instances of the same error will only produce the stack trace. In addition, to help identify files related to out of memory or connection errors, relevant files are suffixed with _oom or _conn.

Improved Mapping Validation for Classes Split across Models

Version 5.3.0 includes a Mapping enhancement for improved support of classes that are split across Models.

Updated Sparkler to Avoid Uploading Older Library Jar Files

Version 5.3.0 updates Sparkler to use the library .jar files that are set by Sparkler start script. This change avoids uploading older Sparkler library .jar files that are not needed.

Dynamic Resource Log and Preservation of Diagnostic Files for K8s-Based Deployments

Version 5.3.0 adds a Dynamic Resource Log in the Administration application. The log captures details about dynamically deployed applications, including details about all Kubernetes events and diagnostic information, such as AnzoGraph Crashdumps and X-Rays, which are preserved even though the cluster may be de-provisioned.

For new installations, the Dynamic Resource Log is enabled by default. If you are upgrading to Version 5.3.0 from a previous version, the Dynamic Resource Log is not enabled by default. Follow the steps below to enable the log:

  1. In the Administration application, expand the Servers menu and click Advanced Configuration. Click I understand and accept the risk.
  2. Search for the Pax Logging SLF4j Listener bundle and view its details.
  3. Click the Services tab and expand the SLF4j Log Listener service.
  4. Click Add Property. In the Name field, specify pax.logging.logback.public.cloud.logs, and then set the Value to public.cloud.log.
  5. Save the new property and then restart Anzo to apply the configuration change.

Added Support for Configuring Multiple NFS Servers

To support clusters in multiple regions, Version 5.3.0 adds support for multiple NFS mount configurations so that each region can mount an NFS server in the same region.

Added the Ability to Save Launch Configurations

In Version 5.3.0, users have the option of saving the launch configuration that is used when activating a Graphmart (and dynamically deploying an AnzoGraph cluster) and when running a Pipeline (and deploying Anzo Unstructured or Spark clusters). When the launch configuration is saved, Anzo remembers the selections so that the user is not prompted to make the same choices the next time they activate a Graphmart or run a Pipeline. In addition, users can also choose which AnzoGraph settings file to use for the deployment. Saved configurations can be cleared by click Clear Launch Configuration under General information in the Overview tab.

Added the Ability to Deploy a Dynamic AnzoGraph Cluster for Multiple Graphmarts

Version 5.3.0 adds the ability to deploy a dynamic AnzoGraph cluster or AnzoGraph plus Elasticsearch cluster from the AnzoGraph screen in the Administration application. If a dynamic AnzoGraph cluster is deployed in advance, it can be used with multiple Graphmarts rather than provisioning a separate cluster each time a Graphmart is activated.

Released Upgraded Versions of the AnzoGraph, Anzo Agent, and Unstructured K8s Operators

Version 5.3.0 adds the option to upgrade the AnzoGraph, Anzo Agent, and Anzo Unstructured Operators that are used in dynamic, Kubernetes-based deployments. The new version, Version 2.0, updates the schema to introduce greater flexibility for being able to change the configuration of a dynamic application, such as to add a volume or change an environment variable, without having to recreate the node pool.

Upgrading the Operators is optional. You can continue to use Version 1.x Operators after upgrading to Anzo 5.3.0. No action is needed. To learn more, your Cambridge Semantics Customer Success manager can provide details and guide you through the updates to your Kubernetes (K8s) infrastructure. If you have multiple Anzo environments that access the same K8s cluster, all environments that use the cluster must use the same version of the Operators.

Changed Query-Driven Template Steps to RDF-Encode Parameters by Default

In Version 5.3.0, Query-Driven Template Steps use RDF encoding for parameters by default. That means a parameter specified as ${param} is translated as ${rdf.param}. If you do not want to use RDF encoding, you can specify plain text by adding text. before the parameter name, for example, ${text.param}.

Limited Permission to Manage Dictionaries to Anzo Admin

Version 5.3.0 limits the Manage Dictionaries permission to the Anzo Admin role by default.

Added the Manage Models Permission to the Data Scientist Role

To prevent errors when a user with the Anzo Data Scientist role ingests a Data Source, Version 5.3.0 adds the Manage Models permission to that role.

Removed Unused Anzo Distributed Unstructured Dependencies

Version 5.3.0 removes several unused dependencies from the Anzo Unstructured software.

Removed the Masking Tab from Data Layers

Version 5.3.0 removes the Masking tab that was available when creating and configuring Data Layers. The tab was not used.

Removed the Clone Permissions Option when Cloning Graphmarts or Layers

Version 5.3.0 removes the Would you like to clone the permissions? option that was presented when cloning a Graphmart or Data Layer. Disabling the option could cause access issues that were not immediately apparent. In Version 5.3.0, the appropriate permission inheritance is applied automatically.

Corrected Sort Icons in Dashboard Filters

Version 5.3.0 corrects the sort icons that are used to order Dashboard Filter results in ascending or descending order. Previously, the arrows in the icons did not match the text.

Changed Schema Instances to Inherit Instead of Copy ACL Configuration

In Version 5.3.0, Schema instances now inherit their ACL configuration from the related Data Source and Schema artifacts. Previously the configuration was copied from the artifacts to the instance.

Removed System Administrator from Default Access Policy Groups

Version 5.3.0 removes "System Administrator" from the list of values that are displayed when searching for Groups or Users to configure Default Access Policy permissions for.

New Aspose.Note for Java License

Version 5.3.0 includes a new license for Aspose.Note for Java version 3.0.

Open Group List Option Missing from Explore Tab

Version 5.3.0 restores the Open Group List button on the Explore tab for Datasets and Graphmarts. Clicking the Open Group List () icon shows the Models for the Graphmart and enables users to show/hide particular Models.

Volume Creator Always Shown as System Administrator

Version 5.3.0 resolves an issue that caused "System Administrator" to be displayed as the creator of all volumes even if another user created a volume.

Unstructured Pipeline Incorrectly Processed All Editions

Version 5.3.0 corrects an issue that caused Unstructured Pipelines to process all Dataset Editions instead of only the Working Edition.

Metadata was Included in Export without Metadata

When a migrationExport command that included -nometa was issued, the resulting export did include the metadata graph. Version 5.3.0 corrects the issue so that the metadata graph is not exported when the -nometa option is specified.

Imported Properties from a Child Model were Unavailable in the Parent Model

Version 5.3.0 corrects an issue that prevented the properties in an imported child model from being reflected in the parent model.

Graphmart Explore Tab Incorrectly Showed Nodes from System Models

Version 5.3.0 resolves an issue that could cause the Graphmart Explore tab to display nodes from system Models that were not included in the Graphmart.

Pipeline History was Cleared when a Data Source was Re-Ingested

Version 5.3.0 resolves an issue that caused Pipeline history to be cleared when a previously ingested Data Source was re-ingested.

Pipeline with Multiple Incremental Jobs Failed

Version 5.3.0 corrects an issue that caused a Pipeline with multiple incremental jobs to fail even though the jobs were successful when run individually.

User was Unable to View Shared Queries

Version 5.3.0 corrects an issue that caused a user to be unable to view the queries that another user shared via the Query Builder.

Could Not Add Statements via the HTTP Client Interface if the Graph Did Not Exist

Version 5.3.0 fixes an issue that caused an anzoclient/add request to fail with an "Import/Export error" if the graph the statements were added to did not pre-exist in the system journal.

Data Inserted via Query Builder Query Did Not Render in the User Interface

Version 5.3.0 fixes an issue where triples and graphs that were created via INSERT DATA queries in the Query Builder were not returned when they were subsequently queried for in the Query Builder. The new data was returned when searched for on the Query Builder Find tab, however.

Related Topics