IBM Support

Available patches for Watson Discovery for IBM Cloud Pak for Data

Preventive Service Planning


Abstract

This document lists the available patches for the Watson Discovery service on IBM Cloud Pak for Data. Not all versions include patches. Only versions that include patches are shown.

Content

cpd-watson-discovery-2.2.1-patch-10

Patch name cpd-watson-discovery-2.2.1-patch-10
Released on 27 April 2022
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
 

cpd-watson-discovery-2.2.1-patch-9

Patch name cpd-watson-discovery-2.2.1-patch-9
Released on 30 March 2022
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
 
Instructions

cpd-watson-discovery-2.2.1-patch-8

Patch name cpd-watson-discovery-2.2.1-patch-8
Released on 23 February 2022
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
 
This patch also resolves the following issue:
 
  • Error:  Discovery might stop working properly if there are a large number of requests being submitted continuously. The API requests fail with HTTP status 50X errors and indicate that there is a Java heap space error in the pod.
    Cause: There is a potential memory leak in Java if IBMJCEPlusFIPS is used with some encryption algorithms. 
    Solution: Switched to use IBMJCEFIPS as the FIPS provider and disabled the problematic algorithms.
Instructions

cpd-watson-discovery-2.2.1-patch-7

Patch name cpd-watson-discovery-2.2.1-patch-7
Released on 16 January 2022
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following vulnerability:
Instructions

cpd-watson-discovery-2.2.1-patch-6

Patch name cpd-watson-discovery-2.2.1-patch-6
Released on 16 December 2021
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
Instructions
Special instructions for applying the patch to Elasticsearch to remediate the Log4j vulnerability:
  1. Apply the ibm-cloudpakopen-elasticsearch-operator-patch-1 patch for the ibm-cloudpakopen-elasticsearch-operator assembly.
  2. Follow the standard instructions to apply the patch.
    The patch adds the opencontent-elasticsearch-7.8.0:1.1.880 Elasticsearch image.
  3. Run the following command to update the Elasticsearch image version that is used in the Watson Discovery instance:

    oc patch wd wd --namespace Project --type=merge --patch \
    '{"spec": {"elasticsearch": {"image": {"tag": "1.1.880" } } } }'
  4. Verify that the image tag that is used in the Elasticsearch pods is updated to the new tag (1.1.880) by running the following commands.
    This action might take a few minutes.

    oc get sts --namespace Project \
    wd-ibm-elasticsearch-es-server-client -o \
    jsonpath='{.spec.template.spec.containers[0].image}{"\n"}'

    oc get sts --namespace Project \
    wd-ibm-elasticsearch-es-server-master -o \
    jsonpath='{.spec.template.spec.containers[0].image}{"\n"}'

    oc get sts --namespace Project \
    wd-ibm-elasticsearch-es-server-data -o \
    jsonpath='{.spec.template.spec.containers[0].image}{"\n"}'

cpd-watson-discovery-2.2.1-patch-5

Patch name cpd-watson-discovery-2.2.1-patch-5
Released on 30 November 2021
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
This patch also resolves the following issues:
 
  • Error: Connection pool receives too many J2CA0045E errors in the API container in the discovery-gateway pod under heavy load.
    Cause: When the Discovery upload API is called repeatedly, the discovery-gateway pods use up the connections to the underlying data store. As a result, requests fail with an internal timeout error.
    Solution: Changed the connection pool settings so it can handle the traffic more efficiently.
Instructions

cpd-watson-discovery-2.2.1-patch-4

Patch name cpd-watson-discovery-2.2.1-patch-4
Released on 30 September 2021
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
This patch includes many updates to dependent software.
This patch also resolves the following issues:
  • Error: Enrichment job logs that are stored in log storage are never cleaned up.
    Cause: The cleaner job is unable to connect to log storage.
    Solution: Fixed the cleaner job. Now the logs are cleaned up periodically.
  • Error: The Machine Learning enrichment cannot process documents and slows down the ingestion job.
    Cause: The HTTP connection cannot be maintained in the enrichment job.
    Solution: Increased connection pool size and fixed the inconsistent connection issue.
  • Error: An error notification is displayed in the product user interface that indicates that the Machine Learning enrichment failed to process some documents. The notification is displayed repeatedly.
    Cause: Often, the error is displayed when the Machine Learning pod is running out of memory as it processes a large document. The likelihood of the problem occurring depends on the size and complexity of a Machine Learning model that is being applied. However, the message is displayed repeatedly because the Machine Learning pod cannot recover on its own.
    Solution: Improved the durability of the enrichment so that only large documents that require more memory than is available can cause the error. And when the problem occurs, the Machine Learning runtime can recover from the error.
  • Error: When you use the Google Chrome web browser and select a facet from the Improve and customize page, the user interface shifts upwards sometimes and requires a page reload.
    Cause: Web browser bug.
    Solution: Added an attribute to the checkbox elements (position: relative) to prevent the shift from occurring in Chrome.
Instructions

cpd-watson-discovery-2.2.1-patch-3

Patch name cpd-watson-discovery-2.2.1-patch-3
Released on 25 June 2021
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
 
This patch resolves the following issues:
  • Error: Salesforce Connector cannot crawl Knowledge Articles.
    Cause: Crawling Knowledge Articles requires a special query condition that is not supported.
    Solution: Added support for the query condition that is required for Knowledge Articles to be crawled.
  • Error: Custom connector does not detect documents that were deleted and then added back with the same name.
    Cause: Internal crawl history is not handled properly when a document is deleted at the data source.
    Solution: Improved crawl history handling.
  • Error: When the deployed instance has a heavy work load, the crawl status is not shared between the microservice crawler pods properly.
    Cause: Crawler pods can abandon the microservice connection when the status information times out.
    Solution: Added retry logic to the microservice connections.
Instructions

cpd-watson-discovery-2.2.1-patch-2

Patch name cpd-watson-discovery-2.2.1-patch-2
Released on 14 May 2021
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
This patch resolves the following issue:
  • Error: If you apply the Contracts enrichment or the Table Understanding enrichment to a collection, you might receive the following error message when that collection is ingesting documents: "The number of nested documents has exceeded the allowed limit of [X]."
  • Cause: The Elastic search component had a limit on nested documents.
  • Solution: This patch reduces the number of nested documents that are generated during ingestion to prevent the error from occurring.
Instructions

cpd-watson-discovery-2.2.1-patch-1

Patch name cpd-watson-discovery-2.2.1-patch-1
Released on 2 April 2021
Service assembly watson-discovery
Applies to service version Watson Discovery 2.2.1
Applies to platform version
  • Cloud Pak for Data 3.5.0 
  • Cloud Pak for Data 3.0.1
Patch type
Cumulative
Description
This patch resolves the following common vulnerabilities and exposures:
This patch resolves the following issue that impacts deployments on Cloud Pak for Data 3.5.0:
  • Error: If you try to provision the Discovery service on a cluster where Planning Analytics is running, some of the Discovery pods don't start and installation fails. The logs for the pod show messages such as,java.lang.NumberFormatException: For input string.
  • Cause: An environment variable named COUCHDB_PORT is added to the Kubernetes cluster by the Apache CouchDB service that is installed with Planning Analytics. Discovery does not use CouchDB, and so does not specify a value for the environment variable. However, some pods attempt to parse the variable, which results in the error.
  • Solution: This patch makes the pods ignore the COUCHDB_PORT environment variable.
Instructions

cpd-watson-discovery-2.1.4-patch-2

 
Patch name cpd-watson-discovery-2.1.4-patch-2
Released on 4 January 2021
Service assembly watson-discovery
Applies to service version Watson Discovery 2.1.4
Applies to platform version
  • Cloud Pak for Data 3.0.1 
  • Cloud Pak for Data 2.5.0
Patch type
Cumulative
Description
This patch resolves the following issue:
  • This patch fixes an issue where Watson Discovery failed to ingest PDF files when a Microsoft PowerPoint file was included in a dataset.
Instructions

cpd-watson-discovery-2.1.4-patch-1

 
Patch name cpd-watson-discovery-2.1.4-patch-1
Released on 28 September 2020
Service assembly watson-discovery
Applies to service version Watson Discovery 2.1.4
Applies to platform version
  • Cloud Pak for Data 3.0.1 
  • Cloud Pak for Data 2.5.0
Patch type
Additive
Description
This patch resolves the following issues:
  • This patch fixes an issue where Watson Discovery was not properly parsing filter aggregation phrase queries that contained escaped characters.
Instructions



cpd-watson-discovery-2.1.3-patch-1

 
Patch name cpd-watson-discovery-2.1.3-patch-1
Released on 08 August 2020
Service assembly watson-discovery
Applies to service version Watson Discovery 2.1.3
Applies to platform version
  • Cloud Pak for Data 3.0.1 
  • Cloud Pak for Data 2.5.0
Patch type
Additive
Description
This patch resolves the following issues:
  • Fixes an issue where Watson Discovery returns string values for start_offset and end_offset when doing a search with passages.per_document enabled.
Instructions

[{"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SSHGYS","label":"IBM Cloud Pak for Data"},"ARM Category":[{"code":"a8m0z000000GpDZAA0","label":"Services and Integrations"}],"ARM Case Number":"","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Version(s)","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Document Information

Modified date:
27 April 2022

UID

ibm16240184