IBM Support

Release Notes - IBM Watson Explorer oneWEX, Version 12.0

Release Notes


Abstract

This document describes new features, known issues and workarounds, and installation information for oneWEX of IBM Watson Explorer Version 12.0.

Content






What's New in Version 12.0.3.5

Watson Explorer oneWEX V12.0.3.5 upgrades IBM Java to 8.0.6.20 and includes PSIRT fixes.

Installing Version 12.0.3.5

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.3.5, see Watson Explorer Deep Analytics Edition V12.0.3.5 Readme. For the information about connecting to Fix Central Web site to obtain the software, see  download documents.


What's New in Version 12.0.3.4

Watson Explorer oneWEX V12.0.3.4 upgrades IBM Java to 8.0.6.10 and includes PSIRT fixes.

Installing Version 12.0.3.4

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.3.4, see Watson Explorer Deep Analytics Edition V12.0.3.4 Readme. For the information about connecting to Fix Central Web site to obtain the software, see  download documents.


What's New in Version 12.0.3.3

Watson Explorer oneWEX V12.0.3.3 upgrades IBM Java to 8.0.6.7 and includes PSIRT fixes.

Installing Version 12.0.3.3

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.3.3, see Watson Explorer Deep Analytics Edition V12.0.3.3 Readme. For the information about connecting to Fix Central Web site to obtain the software, see  download documents.


What's New in Version 12.0.3.2

IBM Cloud Private v3.2.1 Support

IBM Cloud Private v3.2.1 is newly supported.

Security Enhancement

Authentication and Transport Layer Security (TLS) of inter-Pod communication are supported and enabled by default.

Annotators from Watson Knowledge Studio

Watson Knowledge Studio machine-learning annotators for Chinese language and Dutch language are supported.

Installing Version 12.0.3.2

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.3.2, see Watson Explorer Deep Analytics Edition V12.0.3.2 Readme. For the information about connecting to Fix Central Web site to obtain the software, see  download documents.


What's New in Version 12.0.3.1

12.0.3.1 image is updated after the initial General Availability (GA).

The initial image does not delete documents by clicking the "Remove" button for a CSV file on a CSV importer setting page. A log file, "/wexdata/logs/trace_<server_name>_wlp.json" will have a message which contains "Incompatible Jackson version". The updated image resolves this issue.  If you have encountered the issue and have upgraded to the new updated image, you can refer to the technote to sync up the dataset document count.
The release date and the size of the updated download packages are in the Download document.
If you have already installed the initial GA image, you can follow the same steps as upgrading from Watson Explorer oneWEX Version 12.0.3 in the Readme document to apply the new updated image.

IBM Cloud Private with OpenShift Support

IBM Cloud Private 3.2.0 with OpenShift is newly supported.

NetworkPolicy for IBM Cloud Private deployment

For oneWEX IBM Cloud Private deployment, NetworkPolicy configurations are installed. Only Gateway pods can be accessed from outside of Kubernetes.

Initial administrator password

Initial administrator password can be configured.

Installing Version 12.0.3.1

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.3.1, see Watson Explorer Deep Analytics Edition V12.0.3.1 Readme. For the information about connecting to Fix Central Web site to obtain the software, see  download documents.

Known Issues and Limitations in Version 12.0.3.1

  • When you perform graphical installation of Watson Explorer oneWEX for single container deployment version 12.0.3.1, you may see the following error:
    *** Error in `/tmp/install.dir.24844/Linux/resource/jre/bin/java': munmap_chunk(): invalid pointer: 0x00007f57007a7d80 ***

    In that case, please use console installation. See Installing Watson Explorer oneWEX on Docker for console installation.


What's New in Version 12.0.3

IBM Cloud Private Foundation V3.2 support

oneWEX IBM Cloud Private deployment supports IBM Cloud Private Foundation V3.2.

Content Miner Report Creation

Content Miner user can take snapshots (visualization and data) of a widget. Moreover, a report can be created with the snapshots. Unique URL is assigned to each report so that Content Miner user can share the insights with other people.

Content Miner Dashboard Save

Content Miner user can save the dashboard layout for daily monitoring.

Map Visualization Customization GUI

Content Miner user can import geojson file via GUI to use custom map visualization.

Content Miner Custom Widget SDK

With SDK, developers can integrate their own facet visualization with Content Miner.

Document Classification Federated Mode

Document Classification model uses document metadata to split the classification space so that many labels can be handled in one collection more efficiently.

Application Builder Association Support

Application Builder user can use Association, the connection of two data sources through metadata with common data.

Japanese New Era Support

Reiwa (Japanese: 令和) is recognized as a date expression.

Installing Version 12.0.3

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.3, see Watson Explorer Deep Analytics Edition V12.0.3 Readme. For the information about connecting to Fix Central Web site to obtain the software, see  download documents.

Known Issues and Limitations in Version 12.0.3

  • If there is "Local oneWEX backend" entity created at oneWEX AppBuilder version V12.0.2, it is not migrated correctly during V12.0.3 installation. This issue happens if the local backend entity was created in V12.0.2. Entities created in V12.0.2.1 or V12.0.2.2 are not affected.
    If affected backend entity is remained in the AppBuilder environment, you will see "400 Bad Request" error when you try to create new data sources.
    To fix this error, please execute these steps:
    1. Open oneWEX AppBuilder
    2. Click "Backends" link in the main menu
    3. Click to open "Local oneWEX backend"
    4. Modify the url from "https://localhost:443/" to "https://localhost:60443/"
    5. Click Save

What's New in Version 12.0.2.2

IBM Cloud Private Foundation V3.1.2 support

oneWEX IBM Cloud Private deployment supports IBM Cloud Private Foundation V3.1.2.

CSV Import enhancements

  • User can review an uploaded CSV file with summarized information before ingesting the file.
  • Supported character sets are increased.

Microsoft SharePoint Server 2019 support

SharePoint Crawler can crawl Microsoft SharePoint Server 2019.

Secure Search for SharePoint Online

An end user can search for documents in SharePoint Online and the search results will only include documents the user is authorized to access.

Machine Learning-based Enrichment using a Model developed with Watson Knowledge Studio

From oneWEX admin GUI, user can import a model developed with Watson Knowledge Studio. New machine learning-based enrichment uses the model to enrich documents for search or mining.

Swedish language support

Swedish language (SV) is newly supported for Part-of-speech tagging.

Watson Discovery Service support as an Application Builder backend

A data scientist or an application developer can build a web application utilizing the cognitive search capability by Watson Discovery Service.

Map visualization in Content Miner customization

Map visualization in Content Miner can be customized with user's map file.

Installing Version 12.0.2.2

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.2.2, see  Watson Explorer Deep Analytics Edition V12.0.2.2 Readme  . For the information about connecting to Fix Central Web site to obtain the software, see  download documents .

Known Issues and Limitations in Version 12.0.2.2

  • When upgrading a Helm release which uses NFS for its Persistent Volumes (i.e., the "Dynamic Provisioning" option is disabled) to 12.0.2.2 on IBM Cloud Private, make sure the "Dynamic Provisioning" option remains disabled in the upgrade dialog as the option value may be different in some cases.
  • oneWEX SharePoint Crawler does not support crawling SharePoint Online with Federated Server environment (typically ADFS), so secure search (document-level security) for SharePoint Online with Federated Server is not supported.
  • When secure search (document-level security) is enabled, oneWEX SharePoint Crawler doesn't support proxy settings with userID / password for the proxy server when connecting to SharePoint Online.
  • Microsoft SharePoint Server 2019 has an issue in handling the content MIME types, which prevents correct crawling.  Consult Microsoft to get the latest information about the issue including a workaround.
  • From this version, oneWEX runs as a non-root user (UID 60001) due to security reason.
    File systems mounted on oneWEX for the following purposes must be readable by the user.
    - Crawler plugin
    - Custom converter
    - Target of file system crawler
  • If there is "Local oneWEX backend" entity created at oneWEX AppBuilder version V12.0.2, it is not migrated correctly during V12.0.2.2 installation. This issue affects both migration paths: direct update from V12.0.2 (for single container deployment), or update via V12.0.2.1 fix pack (for single container deployment or ICP) if the local backend entity was created in V12.0.2.  Entities created in V12.0.2.1 are not affected.
    If affected backend entity is remained in the AppBuilder environment, you will see "400 Bad Request" error when you try to create new data sources.
    To fix this error, please execute these steps:
    1. Open oneWEX AppBuilder
    2. Click "Backends" link in the main menu
    3. Click to open "Local oneWEX backend"
    4. Modify the url from "https://localhost:443/" to "https://localhost:60443/"
    5. Click Save


What's New in Version 12.0.2.1

SUSE Linux Enterprise Server support

oneWEX single container deployment and IBM Cloud Private deployment supports SUSE Linux Enterprise Server as underlying operating system.

IBM Cloud Private Foundation V3.1.1 support

oneWEX IBM Cloud Private deployment supports IBM Cloud Private Foundation V3.1.1.

Installing Version 12.0.2.1

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.1, see  Watson Explorer Deep Analytics Edition V12.0.2.1 Readme  . For the information about connecting to Fix Central Web site to obtain the software, see  download documents .

Known Issues and Limitations in Version 12.0.2.1

  • Especially when upgrading oneWEX for ICP from Version 12.0.2 to Version 12.0.2.1 on ICP, you need to follow the special steps to upgrade. See Watson Explorer Deep Analytics Edition V12.0.2.1 Readme for the detailed steps.
  • Query Recommendation widget with Spell correction type on oneWEX Application Builder won't work. To work around the issue, see Making spell correction work on oneWEX Application Builder for detailed steps.
  • When using a classifier with the dataset field which name does not match a certain rule (r'[A-Za-z0-9]+') in V12.0.1 then migrate to V12.0.2 or later, the classifier won't work as same as V12.0.1. To work around the issue, recreate a new classifier in V12.0.2 or later with the data when you used to train the classifier in V12.0.1.
  • Upgrading to ICP Version 3.1.1 from previous ICP Versions are not supported at the moment. If you want to use oneWEX Version 12.0.2.1 on ICP Version 3.1.1, install ICP Version 3.1.1 freshly and install oneWEX Version 12.0.2.1.


What's New in Version 12.0.2

oneWEX Application Builder

A data scientist or an application developer can build a Cognitive Search web application using this assistant-style tool that includes basic building blocks like machine learning-based technologies to recommend information based on user activities. Various out-of-the box widgets and customizable widgets can be combined to visualize the analysis results or recommended information in appropriate formats like a card and graph.

Secure Search for SharePoint Server

An end user can search for documents in SharePoint server and the search results will only include documents the user is authorized to access.

Collaborative activities and machine learning-based document recommendation

An end user can give feedback by clicking, rating and voting on documents in search results to increase collaboration among search users. Machine learning-based technology is then used to recommend documents based on these collaborative activities.

Search relevancy tuning

A search administration user can improve the end user search experience using query expansion and boosting capabilities. Search queries are also recommended based on statistical information from query usage and machine learning-based technology.

Character pattern annotator

An annotator developer can develop an annotator which uses regular expressions to extract keywords from text. Samples are provided to demonstrate the use of this annotator including annotators that extract basic personal identifiable information like social security number and e-mail address.

NLP improvement

Improvements are provided to parts-of-speech annotation for Dutch and Portuguese languages.

IBM Cloud Private Foundation support

IBM Cloud Private Foundation V3.1 is bundled for use as a supporting program.

Installing Version 12.0.2

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.1, see  Watson Explorer Deep Analytics Edition V12.0.2 Readme  . For the information about connecting to IBM Passport Advantage Online Web site to obtain the software, see  download documents .

Known Issues and Limitations in Version 12.0.2

See following notes about known issues of Watson Explorer oneWEX Version 12.0.2.

Note on 20 Dec 2018: Query Recommendation widget with Spell correction type on oneWEX Application Builder won't work. To work around the issue, see Making spell correction work on oneWEX Application Builder for detailed steps.

Document-level security with oneWEX SharePoint crawler

  • oneWEX V12.0.2 SharePoint crawler crawls both SharePoint Server and SharePoint Online. oneWEX V12.0.2 SharePoint crawler supports secure search (document-level security) for SharePoint Server 2016 with SAML enabled configuration. When configuring oneWEX V12.0.2 SharePoint crawler, make sure to enable "Use SAML authentication" option. See  SharePoint crawler - configuration properties  for detailed information about SharePoint crawler. However, note that oneWEX V12.0.2 SharePoint crawler does not support secure search for SharePoint Online.
  • oneWEX V12.0.2 SharePoint crawler dose not support pre-filtering when using Active Directory Group ACL format with SharePoint Server. To perform the secure search with SharePoint Server, use one of the following method:
    1. Use post-filtering
    2. If pre-filtering is needed, use claim based user names (User Principal Names(UPN) or email address) in SharePoint Server configuration, including "Site Collection Administrator" settings.

See  Document-level security  also for more information about Document-level security and its limitation.

Crawling SharePoint Online with oneWEX SharePoint crawler

Crawling SharePoint Online with Federated Server environment (typically ADFS) with oneWEX SharePoint crawler is not supported.


What's New in Version 12.0.1

Connections Analysis

  • You can use the Connections view to see how one facet is related to another facet in the data set.

Sentiment Analysis Enhancement

  • Use the Sentiment view to explore the sentiment behind facets. You can assess overall sentiment and see how positive and negative expressions occur in context. Passages with positive, negative, or ambivalence status are underlined in search result summaries.

Facet Pairs Analysis Enhancement

  • Facet Pair view displays the correlation between values from the two selected facets. You can view the results as a flat table, as a grid with intersecting cells, or from a bird's-eye-view.

Document Preview Enhancement

  • Annotations associated with a document are listed on the Document Preview window.

UI Mode

  • New UI mode to provide user experience similar to Analytical Components Content Miner for advanced users who prefer a "dense" presentation

Document Export

  • You can export analysis results and download in CSV or JSON format from Content Miner.

Document Flags

  • You can configure flags that users can use to classify documents when they query a collection. For example, you might create a flag that is called Confidential so that users can flag all corporate confidential documents when they review query results.

Backup and Restore

  • You can export all data and configurations in a backup.

Date Range Facet

  • By creating facets that are based on a date range, you enable users to explore collections by limiting the view to a range of date values, such as insurance claims that were submitted during a specific time period.

Decimal Range Facet

  • By creating facets that are based on a decimal numeric range, you enable users to explore collections by limiting the view to a range of numeric values, such as hospital records for patients of certain ages or products for sale at various prices.

Domain Adaptation Curator

  • Web-based enrichment resource customization tool. You can edit dictionaries by point-and-clicking keywords on a Content Miner analytic view. Moreover, candidates for dictionary entry are suggested using machine learning techniques.

New Data Source Support

  • IBM FileNet Content Manager v5.5.0
  • IBM Case Manager v5.3.2
  • Microsoft SharePoint online
  • Microsoft Exchange Server 2016

Installing Version 12.0.1

For the information about installing / upgrading to Watson Explorer oneWEX V12.0.1, see  Watson Explorer Deep Analytics Edition V12.0.1 Readme . For the information about connecting to IBM Passport Advantage Online Web site to obtain the software, see download documents .

Known Issues and Limitations in Version 12.0.1

oneWEX for ICP deployment

  • Minimum number of available Discovery pods configuration parameter
    When deploying oneWEX on ICP, set the minimum number of available Discovery pod (discovery.minAvailable) to be the same as the number of replicas of Discovery pod (discovery.replica) for stability.
  • Required CPU cores for Helm Chart
    oneWEX for ICP Helm Chart README shows an incorrect number of CPU cores (“23 CPU”) in the “Resources Required” section. The actual number of CPU cores required is 8.

oneWEX for single container deployment

  • Domain Adaptation Curator
    If there are no suggestions from the "Related words" suggestion type in the Domain Adaptation Curator, users need to add the environment variable DEPLOYMENT_TYPE=DESKTOP and then restart the container to get the suggestions.
    To add the environment variable, edit the (wex installation directory)/ibm/wex/config/wex.json file to add the DEPLOYMENT_TYPE line in the env section.
    "env": {
               ...
              "DEPLOYMENT_TYPE": "DESKTOP"
    }

Box Crawler

  • When using Box Crawler, server times should be synchronized between Watson Explorer and the BOX server. If the server times are not synchronized, "HTTP/1.1 400 Bad Request" errors can occur.

Content Miner

  • Searched Document Export
    The “Number of Results” field in the “Searched Document Export Options" dialog does not work. The system always exports all the current searched documents.
  • Topic/Trends Analysis Export
    In the analysis view of the Topic or Trends, users can change the time scale and date range dynamically. However, the change will not be applied to the export result. The export result will contain the data specified by the time scale and data range from the Facet Analysis view.
  • Document Flags
    When some document flags are defined for a collection and the user clicks some document flag value row(s) in a Facet Analysis view for Document Flags, Documents view doesn't show correct documents narrowed down by the selected document flag value.
    To work around this issue, construct a query text manually. For example,
    • *:* AND {!flags labels="flag1"}: when the user wants to show the documents with the "flag1"
    • flag.
*:* AND ( {!flags labels="flag1"} OR {!flags labels="flag2"} ): when user wants to show documents with the "flag1" AND "flag2" flags
  • Collection Data Range Zoom in
    In the collection “Data Range” tab, when the Zoom in icon is clicked without any date selected, a blank page will be shown. Users can refresh the browser to return to the previous page.

Creating a Collection

  • There are some reserved words that cannot be used as field names when creating a new collection. Such reserved words include "tags", "flags", "id", and words starting with "_".

Exporting PEAR files from Content Analytics Studio to oneWEX

  • When exporting PEAR files from Content Analytics Studio to oneWEX, the "Fielded Search" option should not be enabled. The "Fielded Search" option is used by Watson Explorer Analytical Components, and is disabled when exporting to oneWEX by default. If the option is changed to enabled via the “Export to Content Analytics” dialog, a new enrichment from the exported PEAR files can be used by a collection but facets with the "Fielded Search" option enabled will not show any values when mining the collection in Content Miner. 
  • Exporting PEAR files from Content Analytics Studio to oneWEX fails. Apply IBM Watson Explorer oneWEX, Version 12.0.1 HotFix 1 to fix the problem. For more information, see  IBM Watson Explorer oneWEX, Version 12.0.1 HotFix 1 Readme . You must log in to IBM Support: Fix Central and download the fix ( 12.0.1.0-WS-WatsonExplorer-DAEoneWEX-Linux-HF001  or  12.0.1.0-WS-WatsonExplorer-DAEoneWEX-Windows-HF001 depending on your platform).
    • Note: This issue is fixed in IBM Watson Explorer oneWEX, Version 12.0.2.


What's New in Version 12.0.0.1

Single container deployment is supported

  • Installer to deploy oneWEX as a single container is provided for Linux and Windows.

Important Notes:

  • Single container deployment does not support high availability (HA) and scale-out by adding additional containers. If HA or scale-out is required, deploy oneWEX on IBM Cloud Private (ICP).
  • If oneWEX is deployed without IBM Cloud Private, it has a strong limitation on resource utilization. This configuration is for processing small amount of data such as in a Proof of Concept use case. A server equipped with 4 to 8 CPU cores is appropriate. Allocating more CPU cores to the system does not improve performance.

Capability to integrate with DSX Local is implemented

  • Python client sample package can be downloaded from API document page. You can use WEX API quickly and easily by using the package in a Jupyter notebook.
  • This feature is supported as-is, and is not officially supported yet.

Installing v12.0.0.1

Known Issues in Version 12.0.0.1

oneWEX for ICP deployment

  • Installation of oneWEX for ICP Version 12.0.0 GA into ICP Version 2.1.0.2 succeeds, but deploying oneWEX fails when configuring access to a local file system. This issue has been fixed in oneWEX for ICP Version 12.0.0.1. For the information about how to upgrade to oneWEX for ICP Version 12.0.0.1 on ICP Version 2.1.0.2, see IBM Watson Explorer oneWEX Readme .
  • You might see the situation that machine learning capability (classifier/ranker) is not processing after you upgrade from oneWEX Version 12.0.0 GA on ICP Version 2.1.0.1 to oneWEX Version 12.0.0.1 on ICP Version 2.1.0.2. If you see such symptom, confirm the pod status and follow these recovery steps:
    1. Perform step 3 in Upgrading to Watson Explorer oneWEX Version 12.0.0.1 section in IBM Watson Explorer oneWEX Readme to delete the definition.
    2. Type "kubectl get pod | grep management" command to check management pod's name. (In this example, the management pod's name is referred as MGMT_POD.)

    3. Type "kubectl delete pod MGMT_POD" command to recreate the management pod.
    4. Wait until the management pod is recreated, and then try the machine learning capability operation again.

Admin console

  • oneWEX Admin UI is now translated into Arabic/Hebrew/Greek but you may still see some labels in English when using Admin UI for oneWEX in Arabic/Hebrew/Greek.

Salesforce crawler on oneWEX

  • Salesforce crawler does not delete the crawled documents in Dataset by recrawling even when the corresponding documents on Salesforce are deleted. There's no work around of the issue. Contact IBM Support for the solution.


What's New in Version 12.0.0

IBM Watson Explorer oneWEX component brings text analytics with machine learning-based technologies to the enterprise.

  • Intuitive GUI for finding trends and patterns in unstructured data with the ability to interact in natural language
  • Natural language query, simulative document clustering and enhanced type-ahead improve search experience
  • Machine learning-based document classification capability
  • Machine learning-based search relevancy tuning for Similarity Search
  • REST API to integrate with the other systems
  • Out-of-the-box annotators including Named Entity Recognition and Sentiment Analysis
  • Out-of-the-box crawlers to ingest data from enterprise datasource and cloud services
  • Highly customizable natural language processing pipeline that can be used to build cognitive solutions
  • Scalable deployment using Docker containers

Known Issues in Version 12.0.0

Admin Console and Content Miner

  • oneWEX Admin UI is not available in Arabic/Hebrew/Greek. Some text appears in English when using Miner UI for oneWEX in Arabic/Hebrew/Greek.

Admin Console

  • Status of a model after cancelling its creation may not be accurate in some cases. It sometimes stays in "Creating datasets", "Preparing training resource", or "Training" state even after cancelling the model creation process in that state. However, the process itself is successfully cancelled in the backend.

[{"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SS8NLW","label":"Watson Explorer"},"ARM Category":[{"code":"a8m0z000000cvvoAAA","label":"Documentation"}],"ARM Case Number":"","Platform":[{"code":"PF016","label":"Linux"},{"code":"PF033","label":"Windows"}],"Version":"12.0.0;12.0.1;12.0.2;12.0.3","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Document Information

Modified date:
24 March 2021

UID

swg27050305