Release Notes for Mindbreeze InSpire
Version 22.2

Innovations and new features

Mindbreeze InSpire Browser Extension

With the Mindbreeze Extension, you can additionally retrieve internal company information when searching the Internet. Mindbreeze automatically stores the link of the internal search environment you are using as soon as you pull it up for the first time.
This plugin works in Google Chrome and Microsoft Edge.

Link to the documentation

Download in Chrome Web Store

Download for Microsoft Edge

Saved searches can be shared with others

This extension creates the possibility for you to share your personal saved search queries including settings filters with other people. A simple click on the new "Copy" icon will generate a corresponding link that you can share with other people as you wish.

Link to the documentation

Revised Mindbreeze integration in Outlook

Another technical enhancement is the improvement of our Outlook Add-In.

Besides visual adjustments, you now also have an improved integration with your favourite default browser.

Link to the documentation

Microsoft Graph based indexing of Microsoft 365 OneNote notebooks

OneNote notebooks are very diverse and have complex content. For indexing, it is important to process this content correctly, which is why we have further improved the quality of indexing OneNote files as part of the 22.2 release. This means that OneNote notebooks can now be used more optimally.

Link to the documentation

Optimization of Snapshot processes

In order to further optimize the interactivity options in the MMC (Management Center Menu) and to ensure transparency, Snapshot processes are now processed in the background as a backend task as part of this new release. This means that you can leave the "Snapshots" menu in the MMC at any time when creating and importing a Snapshot and devote yourself to other activities in the meantime. After the Snapshot process is finished, the Snapshot is automatically saved in your "Snapshots" table.

Link to the documentation

New and advanced connectors

Jira Connector supports indexing of Jira Cloud

With the new Atlassian Jira Connector, Jira Cloud instances can now be indexed in addition to on-premises instances. Likewise, with the associated Atlassian Jira Caching Principal Resolution Service, it is now also possible to resolve groups in the Jira Cloud.

Link to the documentation

Microsoft Dynamics CRM supports Kerberos authentication

The Microsoft Dynamics CRM Connector now supports Kerberos authentication. However, you are still free to choose your preferred authentication method.

Link to the documentation

Microsoft File Connector supports indexing of Azure file shares

With the extension of the Microsoft File Connector, we can now offer you the possibility to index Azure file shares. In addition, as part of the new release, this Connector supports both Kerberos authentication and SMBv3 (Server Message Block Version 3), which essentially supports encrypted data transfer over the network.

Link to the documentation

Microsoft Graph Connector indexes further additional metadata

With the new Microsoft Graph Connector, additional metadata can now be indexed.

• Awards
• Certifications
• Languages
• Projects
• Skills
• …

Link to the documentation

Microsoft SharePoint Online Connector indexes additionally last editor of a document

With the new version of the Microsoft SharePoint Online Connector, you are now able to index the name of the last editor of a document.

Link to the documentation

Technical extensions

Maintenance updates

We have updated key software components, which allows us to provide future-oriented development and new, innovative features.

Our 22.2 release offers numerous updates that additionally contribute to performance and security optimizations, among other things.
Furthermore, it is our top priority to always ensure the security of your data, so the Keycloak and Protocol Buffers versions have now been updated as part of the release.

Mindbreeze Java SDK supports Java JDK 11

In order to guarantee developers both the free choice of their JDK version and to simplify their development processes and application possibilities, we now additionally offer you the option of seamlessly using Java JDK 11 for the Mindbreeze Java SDK. No further settings are required for the use of Java JDK 11.

Link to the documentation

Security relevant changes

22.2.2.811

• CoreOS Security Update (OpenSSL CVE-2022-3786, CVE-2022-3602)

22.2.1.734

• PrecomputeACL rejects results from other categories

22.2.0.729

• Keycloak Update 18.0.0
• Update Apache Tomcat to >=8.5.79 (CVE-2021-42340)
• Update from jmxterm to 1.0.2
• Update Java to 8u342 (CVE-2022-34169 CVE-2022-25647 CVE-2022-21541 CVE-2022-21540 CVE-2022-21549 CVE-2022-21496)
• Update Tika to 1.28.4 (MEDIUM CVE-2022-25169, CVE-2022-30126, CVE-2022-30973)

Additional changes

22.2.4.830

• Optionally wait for sync to finish instead of returning HTTP 503
• Fix for: possible deadlock while indexing with a lot of metadata

22.2.3.817

• Fix for: StringTable _cache log is held while trying to aquire the ExclusiveLock
• Fix for: Bug in runc 1.1.3 prevents updates from the Management Center

22.2.2.811

• Re-validate DocInfo after merge
• Improved instrumentation browser plugin
• Fix for: Windows installer: "Failed to Login" when using a service account
• Fix for: Alma 8 mes.rpm is not installable on RHEL8
• Fix for: Sync during deleteBuckets can cause incomplete syncs
• Fix for: _maxParticipantFileSizeInBytes is not effective for fileInfoVector in DocumentInfoVectorMergeService
• Deactivating SemText models should lead to disabling sentence segmentation.
• Fix for: Windows installer does not include mesindex.exe
• Improved instrumentation for ACL evaluation
• Fix for: Microsoft Teams Cache: 22.1 Update: principals are not resolved anymore
• Principal Cache: support match_all flag to handle container membership criteria
• ServiceNow Support for "Match All" User Criteria
• Fix for: ClientService PersistedResources performance issue

22.2.1.734

• Fix for: Precompute ACL reject results from other categories

22.2.0.729

• Add async/blocking parameters to mescontrol for [sync|syncdelta|syncfull]
• Blacklist internal metadata in inverter
• client-base.js contains the 'profilesettings' component
• Document limit per node: mesadminapi processes global document limit in Cluster Management Operations
• Document limit per node: MMC UI shows "Disabled" nodes and displays error messages from mesadminapi Cluster Management Operations
• Disabling Index Core Dumps
• Improvement of the log message in case of accidental configuration of a launched service as (normal) sandbox transformer and improvement of the CSVMetadataMappingItemTransformer
• Increase default value for sync_delta_max_buckets to 20
• Insight App Designer: possibility of non-editable templates
• mesindexclient: Parallel update metadata and configurable indexable flags
• Microsoft Graph Connector uses additional beta API endpoint to get more metadata
• Microsoft Teams Connector: Ensure crawl threads are closed
• Move include of conf.d-vhost-8443 to allow overwriting of /realm/REALM/api/service/SERVICE-ID
• Problem with updating the index document (document cannot be found): Deletion is faster than re-inversion
• Properties that were dynamically marked as aggregatable are apparently not sortable
• Reference resolution in the inverter via ForwardIndex
• SharePoint Online Cache: Optimizing Group Resolution Queries for Microsoft Azure Groups in SharePoint Online Cache
• SharePoint Online: Crawl name of person who last edited a document
• Stemmer Transformer does not apply transliteration rule to query with Terms and SubQuery
• Training POS-Tagger and NER-Tagger for German and English by UD, WikiNer
• Update app.telemetry to version 22.2
• Update Google Protocol Buffers (C++, Java, JavaScript, C#, Python)
• Update to use devtoolset-11 instead of devtoolset-4
• Fix for: BuildIndexTool without -F does not delete existing index
• Fix for: Did you mean for people returns wrong suggestions
• Fix for: Highlighting for terms containing numbers and letters incorrect
• Fix for: Highlighting not correct when combining phrase search and other search terms
• Fix for: Highlighting with composite decomposition for content incorrect term queries
• Fix for: Jericho HTML Filter removes important word separators from HTML
• Fix for: Jira Cache throws a NullPointerException when not using a proxy
• Fix for: Jive Crawler crashes after the 100th run because the heritrix job.log file is locked
• Fix for: Language Detector ("Semantic Text Extraction") does not work correctly when Language Detector Item Transformer is configured
• Fix for: LDAP PrincipalCache: "Include Principals" rules do not work as expected
• Fix for: LDAP PrincipalCache with user filter - missing principals (e.g. everyone, system)
• Fix for: Microsoft Stream error when parsing duration if no milliseconds are included in the duration
• Fix for: Microsoft Teams Crawler: Crawler indexing of channels containing only messages older than 8 months fails
• Fix for: Mindbreeze language change facet values
• Fix for: MS File Connector: Get preview content from data source option is not effective for ListerPlugins
• Fix for: Overflow check missing in score calculation in Similar Terms search
• Fix for: Principal Resolution Cache: resolvePrincipals throws IllegalStateException
• Fix for: Query log dashboard coding issues result in skipped applications
• Fix for: Search by referencing with use_precomputed_acls_for_docinfo_access option enabled can become incomplete
• Fix for: ServiceNow principal cache is not complete: Failed to parse JSON document
• Fix for: SSDs are not remounted after inspire container restart
• Fix for: Structural query: "Restrict results" function in designer
• Fix for: Style tag should not be inside div
• Fix for: Update bucket "compactification" error causes Index Service to crash
• Fix for: Yammer index does not contain SharePoint Online document attachments
• Fix for: ZIP export does not contain binaries because call to findItems fails on queryservice with HTTP status 401
• Fix for: ZIP filter should delete the defined referencetarget from ZIP content