axcelerate 5.11.0 release notesaxcelerate-docs.opentext.com/help/axc-main/5.15/en_us... ·...

Axcelerate 5.11.0Release Notes

Published: 2017-Mar-03

Contents1 Introduction 11

2 New Feature Descriptions for Axcelerate 5.11.0 122.1 Internet Explorer 10 No Longer Supported (AXC-38536) 122.2 Daeja No Longer Available in Axcelerate Classic User Modules (AXC-29736) 122.3 Axcelerate 5 Application Cloning (AXC-36962) 132.4 Application-Level User Security (AXC-37590) 132.5 Client Agent Supported onWindows 10 (FOUND-9958) 142.6 64-bit Crawls configurable (FOUND-10089) 142.7 Case-wide Sharing for Amazon S3 Storages (CORE-16541) 152.8 User Interface Changes 15

3 Issues Resolved in Axcelerate 5.11.0 173.1 AXC-27093 (Duplicate Status Category Display Name) 173.2 AXC-27834 (Results List View) 173.3 AXC-34909 (Results List) 173.4 AXC-35531 (Efficiency Dashboard) 173.5 AXC-35704 (Adaptive Batching) 183.6 AXC-36127 (User Preferences) 183.7 AXC-36675 (Tagging) 183.8 AXC-37003 (Apache Commons BeanUtils) 183.9 AXC-37116 (Associated Results View) 183.10 AXC-37410 (Jobs) 193.11 AXC-37595 (Mission Control > Review Workflows Page) 193.12 AXC-37667 (Navigation Page) 193.13 AXC-37702 (Arrangements) 193.14 AXC-37785 (Conversion) 203.15 AXC-37790 (Recent Searches) 203.16 AXC-38148 (User Preferences) 203.17 AXC-38305 (Tomcat Update) 203.18 AXC-38426 (User Logout Events) 203.19 AXC-38437 (Search) 213.20 AXC-38440 (Page Navigation) 213.21 AXC-38448 (404 Responses) 213.22 AXC-38650 (Production Export) 213.23 CORE-9061 (CSV Crawls) 213.24 CORE-9390 (Publish) 223.25 CORE-12697 (Conversion) 223.26 CORE-14923 (SSL for SQL connections) 223.27 CORE-15615 (Crawls) 223.28 CORE-17045 (Publish) 22

2 | © Recommind, Inc. 2017.

3.29 CORE-17085 (CORE Administration) 233.30 CORE-17561 (OCR Jobs) 233.31 CORE-17729 (CSV Export) 233.32 CORE-17759 (Production Storage Handler) 233.33 CORE-17772 (OCR Jobs) 243.34 CORE-17845 (Conversion) 243.35 CORE-17891 (CSV Load) 243.36 CORE-17984 (Redaction Jobs) 243.37 CORE-18022 (TIFF conversion, OCR) 243.38 CORE-18030 (API Function) 253.39 CORE-18147 (Crawls) 253.40 CORE-18163 (Sortable Fields) 253.41 CORE-18164 (StorageMigration) 253.42 CORE-18231 (Processing Resources) 263.43 CORE-18246 (Taggers) 263.44 CORE-18257 (StorageMigration) 263.45 CORE-18304 (File Storage Initialization) 263.46 CORE-18313 (Daylight Saving TimeOffset) 273.47 CORE-18424 (Republishing) 273.48 CORE-18467 (WordMap) 273.49 CORE-18489 (CSV Merge) 273.50 CORE-18530 (Search) 283.51 CORE-18552 (Exception Handling) 283.52 CORE-18587 (Production) 283.53 CORE-18629 (Index Configuration Update) 283.54 CORE-18670 (Viewer Panel) 293.55 CORE-18705 (Search Query Editor) 293.56 CORE-18858 (Session Expiration) 293.57 FOUND-3331 (Crawls) 293.58 FOUND-6207 (Datamodel Node Expansion) 303.59 FOUND-9321 (Crawls) 303.60 FOUND-10594 (Crawls) 303.61 FOUND-10682 (Crawls) 303.62 FOUND-10729 (Crawls) 303.63 FOUND-10744 (Publish) 313.64 FOUND-10764 (Tomcat Update) 313.65 FOUND-10892 (Crawls) 31

4 New Feature Descriptions for Axcelerate 5.10.0 324.1 Ad-Hoc Batching (AXC-19919) 324.2 Application-Level User Security (AXC-12625) 324.3 Axcelerate 5 Application Cloning (AXC-36956) 334.4 Default Arrangements Enabled for Corresponding Review States (AXC-36238) 334.5 Links Added to Login Page (AXC-36238) 344.6 Oracle Outside In Upgrade to 8.5.3 (FOUND-9845) 35


5 Issues Resolved in Axcelerate 5.10.0 365.1 FOUND-7953 (Error Messaging) 365.2 AXC-7895 (Conversion) 365.3 AXC-7898 (Conversion) 365.4 AXC-21532 (Conversion) 365.5 AXC-27882 (ProductionWorkflow Deletion) 375.6 AXC-34830 (Conversion) 375.7 AXC-34831 (Production) 375.8 AXC-36149 (Conversion) 375.9 AXC-36649 (SQE Search) 375.10 AXC-36884 (Engine Load) 385.11 AXC-36937 (Conversion) 385.12 AXC-37237 (Mouse Pointer) 385.13 AXC-37294 (Batch Display Options) 385.14 AXC-37311 (Processing Page) 395.15 AXC-37313 (Two-ScreenMode) 395.16 AXC-37327 (Quick Tags) 395.17 AXC-37351 (Login Page) 395.18 AXC-37458 (Production) 395.19 AXC-37464 (Production) 405.20 AXC-37635 (Smart Filters) 405.21 AXC-37789 (Production) 405.22 AXC-37955 (Date Values) 405.23 CORE-16015 (Crawls) 415.24 CORE-16646 (OCR) 415.25 CORE-17317 (Field Value Creation) 415.26 CORE-17394 (JDBC Crawls) 415.27 CORE-17406 (DXL parser) 415.28 CORE-17474 (Field-based Search) 425.29 CORE-17481 (Storage Handlers) 425.30 CORE-17498 (Engine Saves) 425.31 CORE-17574 (CSV Load) 425.32 CORE-17833 (Box Native Files) 435.33 CORE-17933 (Native Conversion) 435.34 CORE-17944 (Metadata Display) 435.35 CORE-18010 (Application Deletion) 435.36 CORE-18028 (Index Engine) 445.37 FOUND-6877 (Crawls) 445.38 FOUND-7526 (Conversion) 445.39 FOUND-7953 (Error Messaging) 445.40 FOUND-8803 (Crawls) 445.41 FOUND-8809 (Crawls) 455.42 FOUND-8811 (Crawls) 455.43 FOUND-9554 (Exception Handling) 455.44 FOUND-9562 (Crawls) 45


5.45 FOUND-9975 (Crawls) 455.46 FOUND-10208 (Crawls) 465.47 FOUND-10217 (Crawls) 465.48 FOUND-10288 (Crawls) 465.49 FOUND-10336 (Reporting) 465.50 FOUND-10563 (Index Engine Selection) 465.51 FOUND-10588 (Lotus Notes NSF File Crawls) 475.52 FOUND-10593 (Crawls) 475.53 FOUND-10623 (Crawls) 47

6 New Feature Descriptions for Axcelerate 5.9.2 486.1 Microsoft SharePoint 2013/Online Connector (Reference ID: CORE-14693) 486.2 Microsoft Exchange 2013/Online Connector (Reference ID: CORE-13085) 48

7 Issues Resolved in Axcelerate 5.9.2 507.1 AXC-34830 (Conversion) 507.2 AXC-36149 (TIFF production) 507.3 AXC-36649 (Search Query Editor) 507.4 AXC-38148 (User preferences) 507.5 AXC-38305 (Tomcat security) 517.6 AXC-38437 (Associated Results Searches) 517.7 CORE-9061 (CSV load) 517.8 CORE-9390 (XML document structure) 517.9 CORE-14923 (SSL for SQL connections) 517.10 CORE-17561 (OCR) 527.11 CORE-17729 (Export) 527.12 CORE-17845 (Conversion) 527.13 CORE-18022 (TIFF conversion, OCR) 527.14 CORE-18231 (Batch servers) 537.15 FOUND-10217 (Embeddings) 537.16 FOUND-10593 (Box connector) 537.17 FOUND-10623 (Performance) 537.18 FOUND-10682 (Datasource) 537.19 FOUND-10764 (Tomcat security) 54

8 New Feature Descriptions for Axcelerate 5.9.1 558.1 Business Intelligence Data Reload (Reference ID: AXC-36598) 558.2 Current Criteria Panel Enhancements (Reference ID: AXC-37114) 558.3 "Responsive" Field Renamed (Reference ID: AXC-37119) 568.4 U Quick Tag (Reference ID: AXC-32949) 568.5 New (NAS) Storage Handling (Reference ID: CORE-16209) 568.6 PerformanceMeasures (Reference ID: CORE-16714) 578.7 Transym OCR Support (Reference ID: CORE-15448) 58


9 Issues Resolved in Axcelerate 5.9.1 599.1 AXC-34630 (Foldering panel) 599.2 AXC-36584 (Review Workflows page) 599.3 AXC-36619 (Printing) 599.4 AXC-36706 (Permission synchronization) 599.5 AXC-36811 (Bloomberg chat) 609.6 AXC-37033 (Production workflows) 609.7 AXC-37061 (Business Intelligence) 609.8 AXC-37090 (Pods with high number of applications) 609.9 AXC-37264 (Metadata panel) 619.10 AXC-37270 (Performance) 619.11 AXC-37397 (Tagging rules) 619.12 AXC-37398 (Pre-Conversion) 619.13 AXC-37582 (Production) 629.14 AXC-37597 (Performance) 629.15 CORE-15304 (Redaction file references) 629.16 CORE-15992 (Property Post Processors) 639.17 CORE-16389 (Documentum crawls) 639.18 CORE-16439 (Wildcard expansion) 639.19 CORE-16868 (Storage utilization statistics) 639.20 CORE-16975 (Box connector) 649.21 CORE-17050 (OCR jobs) 649.22 CORE-17106 (Bloomberg data) 649.23 CORE-17124 (Documentum crawls) 649.24 CORE-17133 (Performance) 659.25 CORE-17284 (Performance) 659.26 CORE-17312 (OCR jobs) 659.27 CORE-17316 (Stored page count) 659.28 CORE-17382 (Conversion) 659.29 CORE-17458 (Wordmap) 669.30 CORE-17567 (Fetching files) 669.31 CORE-17592 (Bloomberg) 669.32 CORE-17597 (Bloomberg) 679.33 CORE-17598 (Bloomberg) 679.34 CORE-17639 (Tagging) 679.35 CORE-17683 (Memory) 689.36 CORE-17757 (Batch server status detection) 689.37 CORE-17781 (Index engine write lock) 689.38 CORE-17788 (Categorization state display) 689.39 CORE-18013 (Document history) 699.40 FOUND-7160 (Logmessages) 699.41 FOUND-9536 (Application removal) 699.42 FOUND-9588 (0 byte file handling) 699.43 FOUND-9644 (PDF date parsing) 699.44 FOUND-9712 (Performance) 70


9.45 FOUND-9951 (Performance) 709.46 FOUND-10039 (Performance) 709.47 FOUND-10073 (Kerberos and Active Directory login) 709.48 FOUND-10099 (Data source start) 719.49 FOUND-10104 (Application case name) 719.50 FOUND-10177 (Configuration access) 719.51 FOUND-10214 (SSO login) 719.52 FOUND-10291 (Exchange connector) 719.53 FOUND-10308 (Monitoring) 729.54 FOUND-10466 (Box data sources) 72

10 New Feature Descriptions for Axcelerate 5.9.0 7310.1 Business Intelligence Data Reload (Reference ID: AXC-36598) 7310.2 Current Criteria Panel Enhancements (Reference ID: AXC-37114) 7310.3 "Responsive" Field Renamed (Reference ID: AXC-37119) 7410.4 U Quick Tag (Reference ID: AXC-32949) 7410.5 PerformanceMeasures (Reference ID: CORE-16714) 7410.6 Transym OCR Support (Reference ID: CORE-15448) 7510.7 SAML 1.1 AuthN in Axcelerate 5 (Reference ID: Found-9480) 75

11 Issues Resolved in Axcelerate 5.9.0 7711.1 AXC-30193 (Production workflows) 7711.2 AXC-34630 (Foldering panel) 7711.3 AXC-35769 (Review workflows) 7711.4 AXC-36156 (Navigation page) 7711.5 AXC-36439 (Highlighting) 7811.6 AXC-36584 (Review workflows) 7811.7 AXC-36619 (Printing) 7811.8 AXC-36681 (Document access) 7811.9 AXC-36786 (Jobs page performance) 7811.10 AXC-36806 (Navigation page) 7911.11 AXC-36811 (Bloomberg chat) 7911.12 AXC-36899 (Security vulnerability in 7zip 9.20) 7911.13 AXC-37033 (Production workflows) 7911.14 AXC-37061 (Business Intelligence) 8011.15 AXC-37095 (Review batches) 8011.16 AXC-37264 (Metadata panel) 8011.17 AXC-37397 (Tagging rules) 8011.18 CORE-15070 (Engine starts) 8111.19 CORE-15749 (Bloomberg email thread detection) 8111.20 CORE-15992 (Property Post Processors) 8111.21 CORE-16008 (Engine save) 8111.22 CORE-16439 (Wildcard expansion) 8211.23 CORE-16460 (Engine save) 8211.24 CORE-16576 (Performance) 82


11.25 CORE-16613 (Field value deletion) 8211.26 CORE-16620 (Search) 8211.27 CORE-16681 (Junk detection) 8311.28 CORE-16754 (Training data jobs) 8311.29 CORE-16755 (Coding queue performance) 8311.30 CORE-16765 (CSV merge) 8311.31 CORE-16777 (Performance) 8411.32 CORE-16868 (Storage utilization statistics) 8411.33 CORE-16932 (Performance) 8411.34 CORE-16975 (Box connector) 8411.35 CORE-17050 (OCR jobs) 8411.36 CORE-17106 (Bloomberg data) 8511.37 CORE-17133 (Performance) 8511.38 CORE-17187 (Storage size calculation) 8511.39 CORE-17220 (Native files) 8511.40 CORE-17284 (Performance) 8611.41 CORE-17312 (OCR jobs) 8611.42 CORE-17316 (Stored page count) 8611.43 CORE-17449 (Storage SQL feature) 8611.44 FOUND-7160 (Logmessages) 8611.45 FOUND-9588 (0 byte file handling) 8711.46 FOUND-9644 (PDF date parsing) 8711.47 FOUND-9712 (Performance) 8711.48 FOUND-9951 (Performance) 8711.49 FOUND-10039 (Performance) 8711.50 FOUND-10102 (Performance) 8811.51 FOUND-10104 (Application case name) 88

12 New Feature Descriptions in version 5.7.0 Update 1 8912.1 JAVA update to JDK 8Update 92 (Reference ID: Found-9744) 8912.2 SAML 1.1 AuthN in Axcelerate 5 (Reference ID: Found-9480) 8912.3 Unlimited Number of Fields per Type (Reference ID: CORE-16491) 9012.4 Box Connector v2 (Reference ID: CORE-14333) 90

13 Issues Resolved in 5.7.0 Update 1 9113.1 AXC-34980 (Viewer) 9113.2 AXC-35230 (Printing) 9113.3 AXC-36357 (Production rules) 9113.4 AXC-36379 (Business Intelligence) 9113.5 AXC-36464 (Report preview) 9213.6 AXC-36557 (Comments field) 9213.7 AXC-36590 (Production) 9213.8 AXC-36648 (Associated Results) 9213.9 AXC-36676 (Tagging panel) 9313.10 AXC-36682 (Filters on Assignments page) 93


13.11 AXC-36688 (Matter access) 9313.12 AXC-36899 (Security vulnerability in 7zip 9.20) 9313.13 CORE-16577 (Tagging processing) 9413.14 CORE-16579 (Junk detection) 9413.15 CORE-16712, AXC-36811 (Bloomberg invitations) 9413.16 FOUND-9171 (Inconsistent configurations) 9513.17 FOUND-9568 (Service tier) 9513.18 FOUND-9633 (Large cache) 9513.19 FOUND-9646 (Cleanup after crawl) 9513.20 FOUND-9670 (Templates) 9513.21 FOUND-9682 (Oracle OutsideIn operations) 9613.22 FOUND-9750 (Stored search) 9613.23 FOUND-9763 (Engine user session) 96

14 New Feature Descriptions for Axcelerate 5.8.0 9714.1 Production Option "Extracted text vs. Slip-sheet text (Reference ID: AXC-10527)9714.2 UI Support to Re-convert Documents (Reference ID: AXC-20497) 9814.3 Document Flagging for Jobs (Reference ID: CORE-15701) 9914.4 Axcelerate - Continuous Improvements (Reference ID: AXC-35731) 9914.5 Unlimited Number of Fields per Type (Reference ID: CORE-16491) 10014.6 Classic user interfaces run inside Tomcat 8 application (Reference ID: FOUND-983) 10014.7 Retry mini job execution in case of batch server failure (Reference ID: CORE-14508) 101

15 Issues Resolved in Axcelerate 5.8.0 10215.1 AXC-27381 (Bulk OCR/Production timeOCR) 10215.2 AXC-32122 (Robustness and stability) 10215.3 AXC-34980 (Viewer) 10215.4 AXC-36295 (Metadata fetching) 10215.5 AXC-36357 (Production rules) 10315.6 AXC-36379 (Business Intelligence) 10315.7 AXC-36461 (Highlighting) 10315.8 AXC-36535 (Robustness and stability) 10315.9 AXC-36557 (Comments field) 10415.10 AXC-36567 (Viewer) 10415.11 AXC-36590 (Production) 10415.12 AXC-36648 (Associated Results) 10415.13 AXC-36676 (Tagging panel) 10415.14 AXC-36682 (Filters on Assignments page) 10515.15 AXC-36688 (Matter access) 10515.16 AXC-36712 (Review performance) 10515.17 AXC-36732 (Associated results display) 10515.18 AXC-36733 (Review performance) 10515.19 CORE-14974 (Text field used for enrichment during publish) 106


15.20 CORE-15862 (LiveLink connector) 10615.21 CORE-16000 (Jobs) 10615.22 CORE-16009 (Performance) 10615.23 CORE-16071 (Large searches) 10715.24 CORE-16101 (Query exception handling) 10715.25 CORE-16218 (Field value sorting) 10715.26 CORE-16303 (SharePoint XML) 10715.27 CORE-16371 (Performance) 10715.28 CORE-16419 (Write lock removal logging) 10815.29 CORE-16447 (Stack trace dumping) 10815.30 CORE-16492 (Robustness and stability) 10815.31 CORE-16577 (Tagging processing) 10915.32 CORE-16579 (Junk detection) 10915.33 CORE-16614 (Asynchronous saves) 10915.34 CORE-16621 (Conversion performance) 11015.35 CORE-16641 (CertificateUtilities) 11015.36 CORE-16662 (Native conversion) 11015.37 Found-9087 (MSG/PST files with TNEF attachments) 11015.38 FOUND-9171 (Inconsistent configurations) 11015.39 FOUND-9513 (Process control robustness) 11115.40 FOUND-9582 (Adobe Indesign Interchange documents) 11115.41 FOUND-9612 (Application import) 11115.42 FOUND-9633 (Large cache) 11115.43 FOUND-9646 (Cleanup after crawl) 11115.44 FOUND-9670 (Templates) 11215.45 FOUND-9728 (Loading performance) 11215.46 FOUND-9740 (CSV LoadWizard) 11215.47 FOUND-9750 (Stored search) 11215.48 FOUND-9763 (Engine user session) 112

16 Issues Resolved in Axcelerate 5.7.2 11316.1 AXC-35556 (Search Query Editor) 11316.2 AXC-36156 (Login) 11316.3 AXC-36183 (Storage size audit log) 11316.4 AXC-36187 (Storage size collection) 113

17 Issues Resolved in Axcelerate 5.7.1 11417.1 AXC-35230 (Printing) 114

18 Contact Us 115

19 Terms of Use 116


1 IntroductionThis document lists the new features and resolved issues of the current and previousversions.

Features and resolved issuesmay bementionedmore than one time, as some of theprevious releases refer only to Axcelerate Cloud or only to Axcelerate 5 on premise.


1 Introduction

2 New Feature Descriptions for Axcelerate5.11.0

2.1 Internet Explorer 10 No Longer Supported(AXC-38536)Behavior before change

Internet Explorer 10 and 11 were supported.

Behavior after changeInternet Explorer 10 is no longer supported.Internet Explorer 11 is still supported.

2.2 Daeja No Longer Available in Axcelerate Clas-sic User Modules (AXC-29736)Behavior before change

It was possible to conduct the following actions in the classic Axcelerate user inter-face with the Daeja ViewONE applet:

l Create productionsl View documents using Image viewl View produced documents using Production viewl Convert documents to TIFFl Create conversion slip-sheetsl Apply redactions and annotationsl Mark and unmark documentsRedaction required

Behavior after changeTheDaeja ViewONE applet is no longer available in the Axcelerate classic usermodules. As a result, the following actions are no longer possible in the classic Axcel-erate user interface:

l Creating productionsl Viewing documents using Image viewl Viewing produced documents using Production viewl Converting documents to TIFF


2 New Feature Descriptions for Axcelerate 5.11.0

l Creating conversion slip-sheetsl Applying redactions and annotationsl Marking and unmarking documentsRedaction required

Note: These actions remain available in the Axcelerate 5 user interface.

2.3 Axcelerate 5 Application Cloning (AXC-36962)Behavior before change

For a new Axcelerate 5matter cloned from an existing one, the following applic-ation-sided settingswere copied into the new matter:

l Fields and values, includingReview State field valuesl Tagging arrangements, including the corresponding fields and values, taggingrules, quick tags, and themapping to review states

Behavior after changeFor a new Axcelerate 5matter cloned from an existing one, the following applic-ation-sided settings are also copied into the new matter:

l Universes, review workflows and their settings, including document independentsearches (scopes, samples and iterations are not copied).

l Batching Templatesl Production Templates

Changes to Default ConfigurationThe axcelerate.standaloneSystemTemplate-Axc5-v2.config.xml has a new prop-erty enabled by default (litigation_ngTemplatable).

Known LimitationsThematter to clonemust be in the same pod andmust be both running and enabledfor cloning (new internal application configuration/enabled by default for axcel-erate.standaloneSystemTemplate-Axc5-v2.config). Cloning an offlinematter or amatter existing in a different pod is not possible.All settings are cloned, i.e., you cannot pick and choose which settings are copied tothe new matter.Schedules are not supported by templates.

2.4 Application-Level User Security (AXC-37590)Behavior before change

Access to Axcelerate 5 featureswas controlled by assigning user groups to one ormultiple of the following roles:



l Reviewerl Power Reviewerl Power Userl CaseManager

In addition to the existing role-based security, access to the following feature couldbe added or removed individually for individual user groups:

l Download native files

Behavior after changeAccess to the following features can also now be added or removed individually forindividual user groups:

l Manage batchesl Bulk tag documentsl View associated documents outside of batchl Manage rights of users to pull batchesl Add andmodify field values on the tagging and foldering panelsl Manage fields via the Fields & Values pagel Manage arrangementsl Access document historyl Manage batching templates

2.5 Client Agent Supported on Windows 10(FOUND-9958)Behavior before change

Agent supported to run onWindows 7.

Behavior after changeAgent supported to run onWindows 10.

2.6 64-bit Crawls configurable (FOUND-10089)Behavior before change

Data source crawls were run in 32-bit mode. Crawls were always limited to a 2GBmemory. If large files were loaded in parallel, there was a risk of running out ofmemory.



Behavior after changeFor file share and for publishing data sources, administrators can now configure thedata source to use Java 64-bit runtime environment, provided that Microsoft Outlook64-bit is installed on the crawler server. 64-bit crawls are not limited to the 2GBmemory limit, are faster, and they reduce the risk of running out of memory due tolarge files.The default crawl mode is still 32-bit mode, which requiresMicrosoft Outlook 32-biton crawler servers.

Known LimitationsThe Aid4Mail parser is not compatible with the new 64-bit mode.

2.7 Case-wide Sharing for Amazon S3 Storages(CORE-16541)Behavior before change

Axcelerate Cloud used SQL-managed storage with single-instance storage, i.e., de-duplication per storage handler.

Behavior after changeNew system templates are available that allow case-wide sharing of image files andnative files, i.e., file sharing between Axcelerate Ingestion andmatters created fromAxcelerate Ingestion, thus reducing storage space. If case-wide sharing is correctlyconfigured, deleting Axcelerate Ingestion has no effect on publishedmatters.

2.8 User Interface ChangesBehavior before change

TheStatus icon in the following locationswas red:

l Navigation and Processing pages, to indicate “Not running” applicationsl Fields and Values page, to indicate inactive fieldsl Templates page, to indicate inactive templatesl ProductionWorkflows>Production Rules page, to indicate inactive productionrules

l Preconversion Rules page, to indicate inactive preconversion rules

The icon used to add a new value to a field on theAdministration > Fields andValues page was represented by the icon.In the Tagging panel, theEdit button was used to select documents.On theAnalysis page, if you had rights to add or edit field values in the Taggingpanel, you could take these actions only after first selecting a document to activatethe panel.



Behavior after changeTheStatus icon is now gray .The icon used to add a new value to a field on theAdministration > Fields andValues page is represented by the icon.In the Tagging panel, theSelect button replaces theEdit button for documentselection.On theAnalysis page, if you have rights to add or edit field values in the Taggingpanel, you can now take these actionswithout first selecting a document to activatethe panel.



3 Issues Resolved in Axcelerate 5.11.0

3.1 AXC-27093 (Duplicate Status Category DisplayName)

Summary New Behavior

A globalDuplicate Status category hadan incorrect display name, i.e.,Remove.

The display name of the globalDuplicateStatus category is again namedDuplicate.

3.2 AXC-27834 (Results List View)Summary New Behavior

Changing theResults list grid columnswhileviewing documents in QAMmode, e.g., doc-uments queued for production or documents ina production set, with theApplied rule orNewrule column active, caused a JavaScript errorand broke the Results list grid.

Resorting columnswhen theApplied rule orNew rule columnis active no longer causes aJavaScript error.

3.3 AXC-34909 (Results List)Summary New Behavior

Values for multi-value fields were not sor-ted correctly when viewed in theResultslist.

Values for multi-value fields are now sor-ted correctly when viewed in theResultslist.

3.4 AXC-35531 (Efficiency Dashboard)Summary New Behavior

TheEfficiency dashboard was not using thestandard resizing behavior for all text, whichresulted in text of different sizes.

TheEfficiency dashboard now usesthe same resizing behavior for all text.


3 IssuesResolved in Axcelerate 5.11.0

3.5 AXC-35704 (Adaptive Batching)Summary New Behavior

Adaptive batching was skipped forsamples.

Adaptive batching no longer skipped forsamples.

3.6 AXC-36127 (User Preferences)Summary New Behavior

Saving the user preferences showed aSave Successmessage.

Saving the user preferences now shows aPreferences saved successfullymes-sage.

3.7 AXC-36675 (Tagging)Summary New Behavior

Save tagging and foldering requests werenot validated correctly, which allowedmanipulation of tagging and text fields.

Save tagging and foldering requests arenow validated correctly and tagging andtext fields cannot bemanipulated.

3.8 AXC-37003 (Apache Commons BeanUtils)Summary New Behavior

Older version of Apache CommonsBeanUtilsdo not suppress the class property, which couldallow remote attackers tomanipulate theClassLoader and execute arbitrary code viathe class parameter.

The security issue is fixed by usingthe newest version of Apache Com-monsBeanUtils.

3.9 AXC-37116 (Associated Results View)Summary New Behavior

Navigation to a previous document inAssociated Results view caused anerror when there weremore than 500documents.

Navigation to a previous document inAssociated Results view no longercauses an error when there aremore than500 documents.



3.10 AXC-37410 (Jobs)Summary New Behavior

A JavaScript error occured when apaused job was restarted on the Jobspage.

A JavaScript error no longer occurs whena paused job is restarted on the Jobspage.

3.11 AXC-37595 (Mission Control > Review Work-flows Page)


Loading a review workflow page withmany target review states and/or manytagging fields could run into a timeout forlarge universes.

Loading a review workflow page withmany target review states and/or manytagging fields no longer causes a timeoutfor large universes.

3.12 AXC-37667 (Navigation Page)Summary New Behavior

Application sync changeswere synchronizedwith one large database transaction, whichcould cause a timeout and prevent somemat-ters from displaying on the Navigation page.

The Application sync changes arenow done with small transactionsand will not cause a timeout.

3.13 AXC-37702 (Arrangements)Summary New Behavior

When an arrangement with idaxcelerate.tagging.folderingwaschanged from type Folder to Taggingand then a fields-and-values synchronizewas triggered, the arrangement wasremoved from the user interface so itcould not be used, edited or restored.

When an arrangement with idaxcelerate.tagging.foldering ischanged to a Tagging type, the arrange-ment no longer disappears from the userinterface after a fields-and-values syn-chronize.



3.14 AXC-37785 (Conversion)Summary New Behavior

BravaServer generated a job descriptionthat pointed to a non-existant XDL sourcefolder, which resulted in failed conversions.

This issue has been resolved inBravaServer, thereby avoiding failedconversions.

3.15 AXC-37790 (Recent Searches)Summary New Behavior

Deletion of recent searches causedthe Tagging panel to not load.

Deletion of recent searches is now performedusing the database instead of application logic,thereby ensuring the Tagging panel will load.

3.16 AXC-38148 (User Preferences)Summary New Behavior

Saving user preferences could lead to anerror if some unknown parameters weresent to the server.

Unknown parameters are ignored sosaving user preferences does not fail.

3.17 AXC-38305 (Tomcat Update)Summary New Behavior

The current Tomcat version for Axcel-erate 5 was 7.0.63, which containedknown security issues.

The Tomcat version for Axcelerate 5 hasbeen updated to 7.0.73, which containsfixes for these security issues.

3.18 AXC-38426 (User Logout Events)Summary New Behavior

A deadlock BPMprocess caused theworkflow engine to not restart the BPMprocess, resulting in user logout events.

The workflow engine now tries to restartthe BPMprocess up to 10 times, therebyalleviating user logout events.



3.19 AXC-38437 (Search)Summary New Behavior

Search requests for associated results wereexecuted twice in parallel when navigatingthrough an associated results list.

Search requests for associated res-ults are always executed one at atime.

3.20 AXC-38440 (Page Navigation)Summary New Behavior

Persisting themost recently accessed page fora user could lead to an exception in the data-base and cause a red error boxwhen switchingto another page in the application.

Switching to another page in theapplication no longer causes a rederror box.

3.21 AXC-38448 (404 Responses)Summary New Behavior

Thewrong internal usage of a REST end-point led to 404 responses from the ser-vice-tier, polluting the service-tier Tomcataccess log.

The REST endpoint is now correctlyused and 404 responses no longer pol-lute the service-tier Tomcat access log.

3.22 AXC-38650 (Production Export)Summary New Behavior

Exporting of a subset of sub-productions couldcause the export job to fail when the sub-pro-ductionswere not contiguous (for instance:export of Vol1 and Vol3 instead of Vol1, Vol2and Vol3).

Exporting arbitrary subsets of sub-productions no longer causes theexport job to fail.

3.23 CORE-9061 (CSV Crawls)Summary New Behavior

CSV crawls could produce nested tagsinside a rm_numeric_identifier field.

The XML parser is fixed and producescorrect data.



3.24 CORE-9390 (Publish)Summary New Behavior

Document XML structure couldbecome invalid during publishing.

The internal <rm_id.profile> elements arenow always added to the end of the doc-ument's XML structure.

3.25 CORE-12697 (Conversion)Summary New Behavior

Running a "convert for display" job on a doc-ument that had been viewed previously resultedin unnecessary andmisleading stack traces andunnecessary zip operations on batch servers.

The offending logging and zipoperation has been disabled.

3.26 CORE-14923 (SSL for SQL connections)Summary New Behavior

SSL could only be configured if the systemhad the non-standard configuration "Useno SQL connection pool". Thismeant thatSSLwas not possible for SQL-managedstorages.

The tool used to configure the SQL con-nection pool accepts a parameter forSSL and configures the connection pooland its consumers accordingly.

3.27 CORE-15615 (Crawls)Summary New Behavior

Content Server connector did not supportlarge collections.

Content Server connector now supportslarge collections.

3.28 CORE-17045 (Publish)Summary New Behavior

AWS S3 connectivity slowness during largepublish.

Log error message improved.



3.29 CORE-17085 (CORE Administration)Summary New Behavior

Opening the Jobs tab for a project with ameta-engine could result in aNullPointerExceptionwhen one of theunderlying single-engineswas unavail-able and thus did not send its job status.

The Jobs tab can now be opened for aproject with meta-engine even if an under-lying single-engine is unavailable. Inform-ation on such engine is thenmissing fromthe list and the event is logged as an error.

3.30 CORE-17561 (OCR Jobs)Summary New Behavior

Due to the resourcemanagement in distributedcomputing environments (OCR resource != con-version resource), the engine jobmanager wouldoccasionally start all document jobs for a bulkOCR operation at once. Most of those jobswouldwait in a blocked state within the engine for theirrespective resource. Pausing a bulk operationwas not possible, because all document jobswere already started and had to be finished first.

The engine prevents starting newdocument jobs for amacro jobwhen a certain threshold ofblocked jobs for that macro job idis exceeded.

3.31 CORE-17729 (CSV Export)Summary New Behavior

For meta engine projects, it could happenthat the CSV export of some sub-enginesfailed.

We reduced the likelihood of failure andimproved the error handling.

3.32 CORE-17759 (Production Storage Handler)Summary New Behavior

It was possible to configure a storage handlerof an engine forBates Number File Storageand at the same time provide options forStructured Storage, which are neitherrequired nor wanted forBates Number FileStorage.

Added a sanity check, which causesan engine (re-)start to fail whenBates Number File Storage is mis-configured.



3.33 CORE-17772 (OCR Jobs)Summary New Behavior

A canceledOCR job led to orphaned tem-porary files on batch processing servers.

Temporary files are cleaned up after ajob is canceled.


On-the-fly conversions had a long timeoutwhich caused batch servers to receive andhold toomany jobs.

Timeout for on-the-fly jobs is back totwominutes.

3.35 CORE-17891 (CSV Load)Summary New Behavior

Bloomberg documents with many recipients,loaded byCSV load, produced illegal valuesfor the email thread detection and led tocrawler out-of-memorywhile email threadswere computed.

Validity is ensured bymeans of addi-tional sanity checks and out-of-memory has been avoided by suitablelimits.

3.36 CORE-17984 (Redaction Jobs)Summary New Behavior

Redaction jobs left unused temporary fileson batch processing servers.

Temporary files of redaction jobs arecleaned up properly.

3.37 CORE-18022 (TIFF conversion, OCR)Summary New Behavior

Axcelerate Classic conversionand BulkOCR did not cover excep-tion documents.

Axcelerate Classic conversion and BulkOCRcover exception documents. Only documentswith these exception classes are excluded fromAxcelerate Classic conversion and BulkOCR:File format, Archive, Password pro-tected/encrypted.



3.38 CORE-18030 (API Function)Summary New Behavior

The changeTextType() API function couldbemisused to change taxonomy fields, by-passing folder-level security.

The changeTextType() API functioncan no longer bemisused to change tax-onomy fields.


Incremental crawl was slow due to the rootresource for deletions being processedmultipletimes, resulting in redundant deletionresources.

The root resource for deletions isprocessed only once.

3.40 CORE-18163 (Sortable Fields)Summary New Behavior

The rm_attachmentcount and rm_embeddingcount fields were notmarked as sortable in the datamodel con-figuration. As a consequence theywerenot visible in the Axcelerate 5 user inter-face.

A new datamodel template is now activewith rm_attachmentcount and rm_embeddingcount beingmarked as sort-able. This change affects only new pro-jects created with Axcelerate 5.11.

3.41 CORE-18164 (Storage Migration)Summary New Behavior

When removing entries from the storage loc-ations persistency, the smallest key of theunderlyingmapwas sometimes not updatedcorrectly. This led to an exception during stor-age location exports.

Themin andmax keys of the storagelocations persistency are nowupdated correctly when removingobjects.



3.42 CORE-18231 (Processing Resources)Summary New Behavior

A bug in the job grid caused the engineto ignore offline states of bat servers.This caused jobs to be sent to batchservers that were in offline state.

The engine now properly checks the avail-ability of processing resources. In addition, awide range of Exceptions regarding the con-nection to the batch processor will now res-ult in a retry for the job affected.

3.43 CORE-18246 (Taggers)Summary New Behavior

Uninstalling taggers blocked other oper-ations unnecessarily.

The index can be used during theremoval of taggers.

3.44 CORE-18257 (Storage Migration)Summary New Behavior

We introduced a tagging into the storagemigration, which tags all documents that weresuccessfully copied to AWS S3; however, thistagging was not respected when cancelingand restarting the initial copy-job.

When restarting the storagemigra-tion copy job, potential tagging of pre-vious runs are now respected.

3.45 CORE-18304 (File Storage Initialization)Summary New Behavior

For read-only storage handlers, existenceof directories was checked upon start upof an engine, even though an enginenever needs to access these directories.

Existence of directories for file-based stor-age handlers is only performed for thosedirectories which are actively needed,and lazily for read-only handlers.



3.46 CORE-18313 (Daylight Saving Time Offset)Summary New Behavior

When theUse Unified Time Zone andDate Pattern for Conversion andExport feature was active, dates in thenear native view did not respect daylightsaving time.

For each document to be rendered in thenear native view, the document's sen-t/delivery date is retrieved in order to com-pute the time zone offset for daylightsaving time, if appropriate.

3.47 CORE-18424 (Republishing)Summary New Behavior

On republish, conversions and redac-tionswere invalidated due to the pos-sibility of changes in natives.

On republish, we now try to identify whether anative or image has been changed. Ifchanged, or if the state cannot be easilydetermined, we still invalidate; otherwise, con-versions and redactions are kept.

3.48 CORE-18467 (Word Map)Summary New Behavior

Manual interventions temporarily setcheckpoints aside and this could leadto a situation where replaying thewords of the checkpoints overwroteexisting wordmap entries withoutbeing noticed.

Additional sanity checkwas added to pre-vent overwriting existing words in a wordmapwith other words from replaying acheckpoint. Note that this situation can onlyhappen bymanual intervention of the trans-action recovery and in such cases theengine start up will be aborted.

3.49 CORE-18489 (CSV Merge)Summary New Behavior

A recently introduced storage type caused theCSVmerge wizard to block, warning the tar-get had no writable storage.

The CSV wizard now handles alltypes of writable storage.



3.50 CORE-18530 (Search)Summary New Behavior

Queries with proximity operators where oneof the arguments contained an apostrophe(for instanceO'Reilly /3William) caused aparse error in Axcelerate 5, but worked finein other search interfaces.

Queries with proximity operatorswhere one of the arguments containsan apostrophe now work fine in allsearch interfaces.

3.51 CORE-18552 (Exception Handling)Summary New Behavior

Erroneous exception handling could causeretrainer service to become unusable, blockingengine save operations.

Exception handling has beenfixed.

3.52 CORE-18587 (Production)Summary New Behavior

A batch server was turned off while a productionjob was running. The production job failed and per-sisted the exception. The engine - as intended -rescheduled the job on a different batch server.Although the job was restarted it still rememberedthe initial exception and would throw it over andover again.

The production job does notmemorize its exception stateupon restart anymore.

3.53 CORE-18629 (Index Configuration Update)Summary New Behavior

When updating the index configurationwhile an engine was running, an incon-sistent state could be created whenthere were checkpoints present thatmodified the wordmap upon replay.

When updating the index configuration whilean engine is running, and there are check-points present that would lead to an incon-sistent configuration state, the operation isaborted. The user is asked to save theengine and then perform the update of theindex configuration again.



3.54 CORE-18670 (Viewer Panel)Summary New Behavior

The system handled occasional connectionissues in a bad way; which caused latency ofabout 2minutes in theViewer panel. In theworst case, this state became permanent untilthe APP servers were restarted.

Failure recovery is robust withrespect to occasional connectionfailures and repairs itself.

3.55 CORE-18705 (Search Query Editor)Summary New Behavior

Searching for multiple numeric identifiers couldcause an engine crash if all documents of thecorresponding numeric identifier prefix hadbeen removed.

In this special case, the enginedoes not crash anymore.

3.56 CORE-18858 (Session Expiration)Summary New Behavior

Sessions could not expire if theKey PerformanceIndicator feature was active. This feature is activeby default.

Sessions can expire.

3.57 FOUND-3331 (Crawls)Summary New Behavior

Notes parser configuration containedwrong expression forReply Noticeforms.

Notes parser configuration contains cor-rect expression forReply Notice forms.



3.58 FOUND-6207 (Datamodel Node Expansion)Summary New Behavior

Sometimes opening the datamodelnode in NAT took several seconds. Thishappened after a restart of Masterser-vice and when the internal cache wasoutdated.

The slow database query loading the rel-evant information was replaced by amoreefficient one. The node will expand fasternow even if data is loaded from the data-base instead of the cache.


HTML documents were filtered if a MIME typefilter existed for text/xml. Additionally, thesedocuments created two filter events: one for"unsplit attachment" and one for the doc-ument itself.

HTML documents are not filtered if aMIME type filter exists for text/xml. Fil-ter events were also fixed.


meta_rfc5322_thread_path could con-tain wrong entries if elements contain $ characters.

meta_rfc5322_thread_path is correctlybuilt even if elements contain $ char-acters.


Data source could crash if the recipienttable of anOutlookmail is corrupt.

Data source does not crash if the recipienttable of anOutlookmail is corrupt.


Microsoft XPS documents got file exten-sion PPT in native storage.

Microsoft XPS documents get file exten-sion XPS in native storage.



3.63 FOUND-10744 (Publish)Summary New Behavior

DXL-files (or NSF-entries) containingcolon characters in item names producedinvalid XMLwhich could not be published.

Colon characters in item names in DXL-files (or NSF-entries) are encoded in theXML and can be published.

3.64 FOUND-10764 (Tomcat Update)Summary New Behavior

Tomcat version used in CORE was8.0.33.

Tomcat version used in CORE is now 8.0.39which fixes (among others) CVE-2016-3092.


Corrupt PST folder structures could causeJobExceptionswhich could ultimately leadto aborted crawls.

Corrupt PST folder structures aretracked as Archive errors and do notabort the crawl.




4.1 Ad-Hoc Batching (AXC-19919)Behavior before change

TheReview Workflow panel was used to batch documents from theAnalysispage. This required prior project setup where at least one review workflow wasadded to the universe and often the default batching templates needed to be cus-tomized tomatch project specifications.

Behavior after changeDocuments can now be batched from theAll Documents universe using the newBatch Documentswizard, located on theAnalysis page in theActionsmenu fortheResults list. The wizard allows users to add review workflows and configure cus-tom batch settings on-the-fly. If prior project setup was conducted, users also havethe option to select existing review workflows and preconfigured batching tem-plates.If documents aremissing from the universe, the user is warned those documentswill be skipped from the batching.

Known LimitationsAd hoc batching can only be used with theAll Documents universe. The originalbatching functionality remains and can be used to batch documents from custom uni-verses.

4.2 Application-Level User Security (AXC-12625)Behavior before change

Access to Axcelerate 5 featureswas controlled by assigning user groups to one ormultiple of the following roles:

l Reviewerl Power Reviewerl Power Userl CaseManager

Behavior after changeIn addition to the existing role-based security, the following feature can be added orremoved individually for individual user groups:Download native files.



4.3 Axcelerate 5 Application Cloning (AXC-36956)Behavior before change

For a new Axcelerate 5matter, no application-sided settingsweremade.

Behavior after changeFor a new Axcelerate 5matter cloned from an existing one, the following applic-ation-sided settings are copied into the new matter:

l Fields and values, includingReview State field valuesl Tagging arrangements, including the corresponding fields and values, taggingrules, quick tags, and themapping to review states

Caution: The security for arrangements is not copied, i.e., it remains set toEveryone to make sure all users in the new matter have access. If anarrangement in the new application requires security, adjust the user groupaccess on theAdministration > Arrangements page.

Changes to Default ConfigurationThe axcelerate.standaloneSystemTemplate-Axc5-v2.config.xml has a new prop-erty enabled by default (litigation_ngTemplatable).

Known LimitationsThematter to clonemust be in the same pod andmust be both running and enabledfor cloning (new internal application configuration/enabled by default for axcel-erate.standaloneSystemTemplate-Axc5-v2.config). Cloning an offlinematter or amatter existing in a different pod is not possible.All settings are cloned, i.e., you cannot pick and choose which settings are copied tothe new matter.

4.4 Default Arrangements Enabled for Cor-responding Review States (AXC-36238)Behavior before change

The default tagging arrangements were not enabled for any review state.

Behavior after changeTwo default associationswill exist in new Axcelerate Review & Analysismatters:

l The 1st Level tagging arrangement will be enabled for the First Pass ReviewComplete review state.

l The 2nd Level tagging arrangement will be enabled for theSecond PassReview Complete review state.



Changes to Default ConfigurationThe default 1st Level and 2nd Level tagging arrangement are enabled for the cor-responding First / Second Pass Review Complete review state.Existingmatters are not affected, this change is only used for newly created Axcel-erate 5matters.

4.5 Links Added to Login Page (AXC-36238)Behavior before change

The Axcelerate 5 login page contained two links:Reset password andSupportterms.

Behavior after changeThe Axcelerate 5 login page contains up to five links: Reset password, Supportterms, Training & Certification, Mind over Matters blog, andOnline Help. Via con-figuration these links can be changed.

Changes to Default ConfigurationNew settings in the webtier.user.properties can be used to control the label and tar-get for up to five links on the login page of Axcelerate 5:

l axcng.navigation.root.branding.section1.label=Passwort zurücksetzenl axcng.navigation.root.branding.section1.url=http://zurueck.recomind.del axcng.navigation.root.branding.section2.label=Nutzungsbebedingungenl axcng.navigation.root.branding.section2.url=https://terms.recommind.del axcng.navigation.root.branding.section3.label=AXCTRAX@YouTubel axcng.navigation.root.branding.section3.url=https://www.youtube.com/channel/UC6NHYGswQju66MOgRrbMiXA

l axcng.navigation.root.branding.section4.label=l axcng.navigation.root.branding.section4.url=/l axcng.navigation.root.branding.section5.label=LaunchPadl axcng.navigation.root.branding.section5.url=https://launchpad.rmcloud.recommind.com/

Per default, the three new links reference the Training & Certification center, Mindover Matters blog, and Axcelerate 5 online help portal.

Known LimitationsThe total width for all five links is limited. Exceeding the width limit will cause the linksto wrap and not display well.



4.6 Oracle Outside In Upgrade to 8.5.3 (FOUND-9845)Behavior before change

Outside In 8.5.1 was used.

Behavior after changeOutside In 8.5.3 will be used on new crawls for new Axcelerate 5 projects. This ver-sion supports new formats for data loading, including: Microsoft Word, Excel andPowerPoint 2016; Microsoft Outlook 2011 for Mac (OLMand EML); CorelWordPer-fect; Corel Quattro Pro; Corel Presentations; Corel Draw X7; AutoCAD 2015.Outside In 8.5.1 will be used for new and old data sources of existing projects as adefault, to avoid unexpected changes in hash values.Administrators have the ability to change the data source configuration of updatedprojects to support the new data formats.




5.1 FOUND-7953 (Error Messaging)Summary New Behavior

Last Crawl Details dialog in CORE Admin-istration showed an error message withoutany reference to the error reason. Thisoccurred when the index engine was not run-ning.

The Last Crawl Details dialog nowshowsmore information on the errorreason when the index engine is notrunning.


Vertical axis titles not properly rotated dur-ing conversion to PDF.

Fixed by upgrade to Oracle Outside In8.5.3 (see FOUND-9845).


Shifted gridlines during conversion toPDF.

Fixed by upgrade to Oracle Outside In 8.5.3(see FOUND-9845).


HugeHTML tables could not be con-verted.




5.5 AXC-27882 (Production Workflow Deletion)Summary New Behavior

A production workflow could not bedeleted after a production was expor-ted through CORE Administration.

The deletion of a production workflow doesnot lead to an exception when productionsets related to the workflow were exportedthrough CORE Administration.


The conversion to TIFF performance wasslow for certain documents.

The conversion to TIFF performancehas been improved.

5.7 AXC-34831 (Production)Summary New Behavior

Produced email shows incorrect timeoffset.



Some documents could not be produced toTIFF successfully due to "IPC pipe closed" con-version issues.

These documents are now pro-duced successfully.

5.9 AXC-36649 (SQE Search)Summary New Behavior

SomeSQE searches time outafter 60 seconds.

The performance of the SQE search has beenimproved by reducing the number of requestsfrom the browser and by combining searches.



5.10 AXC-36884 (Engine Load)Summary New Behavior

Opening theReview Workflows pagegenerated substantial load on theengine.

Opening theReview Workflows page nolonger generates a substantial load on theengine.


PDF export misses some parts for ahtml file.


5.12 AXC-37237 (Mouse Pointer)Summary New Behavior

When hovering over the disabled icon forprevious or next document in theViewerpanel, themouse pointer turned into thehand pointer. This indicated the button wasclickable, but it was actually not.

When hovering over the disabled iconfor previous or next document in theViewer panel, themouse pointer nowstays in the default state.

5.13 AXC-37294 (Batch Display Options)Summary New Behavior

Check box for Show AssociatedResults outside of batchwas shown asdisabled in theBatch Display fly-in,although the function itself was enabledby default for batch review.

Check box for Show AssociatedResults outside of batch is nowshown as enabled in theBatch Displayfly-in, as the function is enabled bydefault for batch review.



5.14 AXC-37311 (Processing Page)Summary New Behavior

When using the # Users filter in theSystemPropertiesSmart Filter group on the Axcelerate 5Processing page, it was not possible to filter formatters with zero users because the input formonly accepted values greater than zero.

It is now possible to filter to mat-ters that have zero users.

5.15 AXC-37313 (Two-Screen Mode)Summary New Behavior

In FireFox, when users opened theAnalysis page in two-screenmode, thesecond screen was locked.

In FireFox, the second screen is no longerlocked when users open theAnalysispage in two-screenmode.

5.16 AXC-37327 (Quick Tags)Summary New Behavior

When users tried to bulk tag documentsusing the Last Applied quick tag, therewas no error message shown to indicate theaction was not allowed.

An error message is now shownwhenusers try to bulk tag documents usingthe Last Applied quick tag.

5.17 AXC-37351 (Login Page)Summary New Behavior

In the case of a fatal error, the Axcelerate 5error page was grayed out and users werenot shown a link to log in again.

In the case of a fatal error, the Axcel-erate 5 error page displays correctly.


Production snapshot was incorrectlyflagged as "Stale."

Stale icon is now only rendered if last snap-shot date is a valid date and older than 7days.




Projects in S3 storage deduced file exten-sions of exported production native filesbased on the file'sMIME type, which is notalways unique and is inconsistent with howfile-based projects deduce file extensions.

The exported file extensions are cor-rect: new records stored in S3 explicitlyprovide fast access to the extension,and old ones fetch the correct extensionfrom the index.

5.20 AXC-37635 (Smart Filters)Summary New Behavior

If a filter value contained aNOT prefix, e.g.,Not Privileged, and the filter was set toExclude that value, theCurrent Criteriapanel showed only the value's display name,e.g.,Privileged, and did not show theNOTprefix of the value's display name.

If a filter is set toExclude a value con-taining aNOT prefix, the value's fulldisplay name, e.g.,Not Privileged, iscorrectly shown in theCurrentCriteria panel.


When a production workflow was exportedvia theProductions tab of CORE Admin-istration, the corresponding default pro-duction rules were not deleted as expected.

Default production rules are alwaysdeleted when their corresponding pro-duction workflow is deleted.

5.22 AXC-37955 (Date Values)Summary New Behavior

Date values presented with a time zone,e.g., on theAnalysis page, reports and doc-ument history always showed the daylightsaving time (DST) variant of the user's pref-erence time zone. The date values, how-ever, were converted to the user's time zonerespecting DST. This resulted in wrong timevalues (shifted byDST offset) to be dis-played for dates not within the DST intervalof the user's time zone.

The displayed time zone now matchesthe DST variant of the date to be dis-played, i.e., dates within the DST inter-val of the user's preference time zonewill show the DST variant and datesoutside of the DST interval will displaythe regular time zone variant.




Incorrect Oracle Outside In PDF con-version.


5.24 CORE-16646 (OCR)Summary New Behavior

If OCR was executed for an incompleteattachment family and then new doc-uments were indexed and the engine wasflushed, the documents for which OCRwas executed could have been flagged asMISSING_PARENT and/or MISSING_ROOT by attachment processing.

When inserting new documents andflushing the engine, the attachment pro-cessing no longer affects documents forwhich OCR had been executed pre-viously, in particular these are not incor-rectly flagged asMISSING_PARENTand/or MISSING_ROOT.

5.25 CORE-17317 (Field Value Creation)Summary New Behavior

Creating a field value with properties viathe CORE API happened in two stepsand a downtime of the engine betweenthem could leave the field value createdwithout its properties.

Creating a field value with properties viathe CORE API is now one transaction,i.e., a downtime of the engine during thatoperation does not leave the field valuecreated without its properties.

5.26 CORE-17394 (JDBC Crawls)Summary New Behavior

Incremental JDBC crawls failed if allchunk entries (document IDs) hadoccurred in previous chunks.

Incremental JDBC crawls continue afterchunks that contain only entries that haveoccurred in previous chunks.

5.27 CORE-17406 (DXL parser)Summary New Behavior

DXL parser sometimes caused HTMLentities to be shown as document content.

DXL parser does not cause HTML entit-ies to be shown as document content.



5.28 CORE-17474 (Field-based Search)Summary New Behavior

Field-based searcheswere extended by auto-matically detected phrases, resulting inmatches containing these phrases anywhere(not necessarily in the specified field).

Field-based searches are no longerextended by automatically detectedphrases.

5.29 CORE-17481 (Storage Handlers)Summary New Behavior

The storage verification did not work prop-erly after a read-only storage handler hadbeen erroneously upgraded to SQL.

The storage verification ignores anyread-only handler that has beenupgraded to SQL and works properly.

5.30 CORE-17498 (Engine Saves)Summary New Behavior

High server peakmemory usagecould cause engines to become unre-sponsive.

Internal exception handling has beenimproved such that low memory conditionscannot cause engines to become unre-sponsive anymore.

5.31 CORE-17574 (CSV Load)Summary New Behavior

If the content of the field rm_attachmentprimary sent by aCSV load was invalid or con-tradicted the value of rm_attachmentroot, then this brokefeatures such as family sorting.

If the content of the field rm_attachmentprimary sent by a CSV load isinvalid or contradicts the value of rm_attachmentprimary, then it is automatically cor-rected at insertion time. Any correction otherthan adding amissing value leads to the doc-ument being tagged asContent modificationin theException ClassSmart Filter and asResolved inconsistent or invalidattachment primary information in theException TypeSmart Filter.



5.32 CORE-17833 (Box Native Files)Summary New Behavior

Box native files were always downloadedduring re-crawls despite their unchangedmodification date.

Box native files are only downloadedduring re-crawls if necessary.

5.33 CORE-17933 (Native Conversion)Summary New Behavior

The conversion of some legacyMicrosoftWord and PowerPoint documents failed withthe error message, "The document-file is cor-rupt because it contains no storage object."

The affected legacyMicrosoft Wordand PowerPoint documents can beconverted correctly.

5.34 CORE-17944 (Metadata Display)Summary New Behavior

The date 1.1.1970 00:00 was displayed as anempty string in themetadata section of theHTML-View that is shown, e.g., in COREAdministration. All other dates including31.12.1969 23:59 and 1.1.1970 00:01 were dis-played correctly.

The date 1.1.1970 00:00 is now dis-played correctly in themetadatasection of theHTML-View.

5.35 CORE-18010 (Application Deletion)Summary New Behavior

Amissing database connection could res-ult in inconsistent project relations inCORE Administration when deleting anAxcelerate Ingestion or AxcelerateReview & Analysis application: theinvolved entities, e.g., engines, weredeleted, but still shown in CORE Admin-istration.

A failing database connection when clear-ing a case in CORE Administration willnot lead to broken relations. Theremoved entities are deleted and theuser is informed to delete the databaseentriesmanually. Leftover entries do notharm further operations.



5.36 CORE-18028 (Index Engine)Summary New Behavior

Write-locking an index engine fol-lowed by a save, blocked field valuemodifications.

Field valuemodifications are processed evenif the index engine is write-locked. Addi-tionally, save jobs can be canceled while wait-ing for the "write unlock."


CrawlingMSG files resulted in OracleOutside In error.


5.38 FOUND-7526 (Conversion)Summary New Behavior

Oracle Outside In failed to convert cer-tain .pdf and .doc files.


5.39 FOUND-7953 (Error Messaging)Summary New Behavior

Last Crawl Details dialog in CORE Admin-istration showed an error message withoutany reference to the error reason. Thisoccurred when the index engine was not run-ning.

The Last Crawl Details dialog nowshowsmore information on the errorreason when the index engine is notrunning.


Documents created when loadingMicrosoft Project files could contain tem-porary file names.

Documents created when loadingMicrosoft Project files contain original filenames.




Oracle Outside In added white space intext strings for ICS files.



Oracle Outside In added \n characters intext strings for VCF files.


5.43 FOUND-9554 (Exception Handling)Summary New Behavior

Documents with multiple paragraphscaused postprocessing exceptions if anold (not property-) postprocessor wasconfigured.

Documents with multiple paragraphs donot cause postprocessing exceptions ifold (not property-) postprocessors areconfigured.


Oracle Outside In erroneously reportedPDF file as corrupt during crawl.

Fixed by upgrade to Oracle Outside In 8.5.3 (see FOUND-9845).


Large binaryOutlookHTMLmail bodiescould causeOutOfMemory errors if themaildid not contain an RTF body.

Large binaryOutlookHTMLmail bod-ies do not causeOutOfMemory errors.




WithMicrosoft Outlook 2013 the crawlercrashed if a PST contained a corrupt entrythat could not be exported as a nativeMSG file.

PSTs that contain an entry that cannotbe exported as a nativeMSG file are notprocessed and aremarked as corrupt.


Embedding names that contained illegalfile system characters following the lastdot caused exceptions.

Embedding names containing illegal filesystem characters following the last dotdo not cause exceptions.


MBOX files with mixed newline characterswere not processed correctly.

MBOX files with mixed newline char-acters are processed correctly.

5.49 FOUND-10336 (Reporting)Summary New Behavior

A fix user name and password were usedto generate reports in Axcelerate Inges-tion.

Information from the current user's ses-sion is used to generate the reports.

5.50 FOUND-10563 (Index Engine Selection)Summary New Behavior

LaunchPad automatically selected indexengine hosts based on CPU load.

LaunchPad automatically selects indexengine hosts based on free RAM.



5.51 FOUND-10588 (Lotus Notes NSF File Crawls)Summary New Behavior

The default CORE configuration was notcompatible with LotusNotesDomino incase RMI over TLS was enabled.

The default CORE configuration is com-patible with LotusNotesDomino in caseRMI over TLS is enabled.


In certain scenarios the number of wait-ing jobswas very high, leading to highmemory consumption.

An optimization was implemented that not-ably decreases the number of scenarioswhere this can happen.


There was a performance regressionwhile extracting native files for PST entriesand attachedMAPI messages.

Performance for extracting native filesfor PST entries and attachedMAPI mes-sages is restored.




6.1 Microsoft SharePoint 2013/Online Connector(Reference ID: CORE-14693)Behavior before change

Ther was no connector supportingMicrosoft SharePoint 2013 or Online for onpremise installations.

Behavior after changeThe new SharePoint 2013/Online connector supportsMicrosoft SharePoint 2013and SharePoint Office 365 for on premise installations.With this connector, all content types, attachments and librairy documents areindexed. In addition, indexing of containers, user profiles and hidden items, or itemstagged asNoCrawl is possible.The connector offers various data scope filters, such as content type and list tem-plate type inclusion/exclusion. Incremental crawls with this connector support dele-tion.

Known Limitations

l Load balanced URLs are not supported.l SharePoint farm level is only supported for the User Profile crawl.l Native files can only be copied at crawl time.l SharePoint webs and sub webs located on different ports are not supported.

6.2 Microsoft Exchange 2013/Online Connector(Reference ID: CORE-13085)Behavior before change

There was no connector supportingMicrosoft Exchange 2013/Online for on premiseinstallations.

Behavior after changeWith the new Microsoft Exchange 2013/Online connector, Microsoft Exchange 2013andOffice 365 for Exchangemailboxes can be crawled for on premise installations.The connector features date range filtering, mailbox inclusion/exclusion, email folderinclusion/exclusion and item type inclusion/exclusion filtering. In addition, it cancrawl journal mailboxes for Exchange 2013OnPremise.



Known Limitations

l Exchange address resolution is not supported.l Distribution list expansion is not supported.l Exchange 2013OnPremise: only one journal mailbox per data source can becrawled.

l No auto discovery support. User must specify Exchange web service URL.l No public folder support.l No search folder support.

l Calendar entries accepted into the calendar (ICS files) will not contain any attach-ments in the native file itself as this is not supported byMicrosoft. However, attach-ments aremaintained within the CORE index, are displayed in the user interfaceand can be exported. Thismay be an issue for productions that only contain nativefiles. These files should therefore be produced as images.





The conversion to TIFF performance wasslow for certain documents.

The conversion to TIFF performancehas been improved.

7.2 AXC-36149 (TIFF production)Summary New Behavior

Some documents could not be produced toTIFF successfully due to "IPC pipe closed" con-version issues.

These documents are now pro-duced successfully.

7.3 AXC-36649 (Search Query Editor)Summary New Behavior

SomeSearchQuery Editorsearches timed out after 60seconds.

The performance of the SearchQuery Editorsearch has been improved by reducing the num-ber of requests from the browser and by com-bining searches.

7.4 AXC-38148 (User preferences)Summary New Behavior

Saving user preferences could lead to anerror if some unknown parameters weresent to the server.

Unknown parameters will be ignored sothat saving user preferenceswill not fail.



7.5 AXC-38305 (Tomcat security)Summary New Behavior

There were several security issues inTomcat 7.0.63.

Tomcat is updated to 7.0.73 which containsfixes for these security issues.

7.6 AXC-38437 (Associated Results Searches)Summary New Behavior

Search requests for associated results areexecuted twice in parallel when navigatingthrough the Associated Results lists.

Search requests for associated res-ults are always executed once at atime.

7.7 CORE-9061 (CSV load)Summary New Behavior

CSV data load could produce nested tagsinside an "rm_numeric_identifier" field.

The XML parser is fixed and producescorrect data.

7.8 CORE-9390 (XML document structure)Summary New Behavior

Document XML structure could becomeinvalid during publishing, due to wrongplace of <rm_id.profile> elements.

The internal <rm_id.profile> elementsare now always added to the end of thedocument's XML structure.

7.9 CORE-14923 (SSL for SQL connections)Summary New Behavior

SSL could only be configured if the systemhad the non-standard configuration "Useno SQL connection pool". Thismeant thatSSLwas not possible for SQL-managedstorages.

The tool used to configure the SQL con-nection pool accepts a parameter forSSL and configures the connection pooland its consumers accordingly.



7.10 CORE-17561 (OCR)Summary New Behavior

Due to the resourcemanagement in distributedcomputing environments (OCR resource != con-version resource) the engine jobmanager occa-sionally started all document jobs for a bulkOCRoperation at once. Most of those jobs had to waitin a blocked state within the engine and wait fortheir respective resource. So, pausing a bulkoperation was not possible, since all documentjobswere already started and had to be finishedfirst.

The engine prevents starting newdocument jobs for amacro jobwhen a certain threshold ofblocked jobs for that macro job isexceeded.

7.11 CORE-17729 (Export)Summary New Behavior

In meta engine projects, the CSV exportfrom some sub-engines could fail.

Likelihood of failure is reduced and theerror handling is improved.


On-the-fly conversions had a long timeoutwhich caused batch servers to receive andhold toomany jobs.

Timeout for on-the-fly jobs is back totwominutes.

7.13 CORE-18022 (TIFF conversion, OCR)Summary New Behavior

Axcelerate Classic conversionand BulkOCR did not cover excep-tion documents.

Axcelerate Classic conversion and BulkOCRcover exception documents. Only documentswith these exception classes are excluded fromAxcelerate Classic conversion and BulkOCR:File format, Archive, Password pro-tected/encrypted.



7.14 CORE-18231 (Batch servers)Summary New Behavior

A bug in the job grid caused the engineto ignore offline states of batch servers.This caused jobs to be sent to batchservers that were in offline state.

The engine now properly checks the avail-ability of processing resources. In addition,a wide range of exceptions regarding theconnection to the batch processor will nowresult in a retry for the job affected.

7.15 FOUND-10217 (Embeddings)Summary New Behavior

Embedding names containing illegal filesystem characters following the last dotcaused exceptions.

Embedding names containing illegal filesystem characters following the last dotdon't cause exceptions.

7.16 FOUND-10593 (Box connector)Summary New Behavior

In certain scenarios the number of wait-ing jobswas very high, leading to highmemory consumption.

An optimization is implemented that not-ably decreases the number of scenarioswhere this can happen.

7.17 FOUND-10623 (Performance)Summary New Behavior

There was a performance regressionwhile extracting native files for PST entriesand attachedMAPI messages.

Performance for extracting native filesfor PST entries and attachedMAPI mes-sages is restored.

7.18 FOUND-10682 (Datasource)Summary New Behavior

Datasource could crash if the recipienttable of anOutlookmail was corrupt.

Datasource doesn't crash if the recipienttable of anOutlookmail is corrupt.



7.19 FOUND-10764 (Tomcat security)Summary New Behavior

Tomcat version used in CORE was8.0.33.

Tomcat version used in CORE is now 8.0.39which fixes (among others) CVE-2016-3092.




8.1 Business Intelligence Data Reload (ReferenceID: AXC-36598)Behavior before change

Data for the Business Intelligence dashboard was loaded every 4 hours. A manualreload was not available.

Behavior after changeIn addition to the scheduled reload every 4 hours, the new Reload DashboardData button in the toolbar menu for anyBusiness Intelligence dashboard can beused to load the latest data into the dashboard.

Known LimitationsNone.

8.2 Current Criteria Panel Enhancements (Refer-ence ID: AXC-37114)Behavior before change

The applied criteria section in theSearch panel only allowed to lock and deleteapplied criteria. To display the specific values, and to edit, the panel needed to bemaximized. Themaximized height took less than 50% of theSearch panel, andallowed to display <~10 different criteria without scrolling.

Behavior after changeThe applied criteria section in theSearch panel allows to display, lock, edit anddelete applied criteria, without having tomaximize the panel height. Themaximizedheight has been increased to 50% of theSearch panel, allowing to display >~10 dif-ferent criteria without scrolling.




8.3 "Responsive" Field Renamed (ReferenceID: AXC-37119)Behavior before change

For new matters, the default field to track responsiveness of a document wasnamedResponsive with yes and no values.

Behavior after changeFor new matters, the default field to track responsiveness of a document is namedResponsiveness withResponsive andNot responsive values.

Changes to Default ConfigurationThe default system template is now using the new display nameResponsivenessfor the ax_responsive field, and loads the default valuesResponsive andNotresponsive from a new OWL file.TheOWL file with yes and no values still exists, but is no longer used by the currenttemplate.


8.4 U Quick Tag (Reference ID: AXC-32949)Behavior before change

The L button applied the last complete tagging.

Behavior after changeThe L button continues to apply the last complete tagging.A new U button can be used to apply only the last changes to tagging.

Changes to Default ConfigurationNone.


8.5 New (NAS) Storage Handling (ReferenceID: CORE-16209)Behavior before change

NAS based projects used storage handlers, called structured storage handler, tostore and access documents, e.g. natives or images, from directory based locations.



Behavior after changeThere is now a new (NAS-) storage handler, called structured storage V2, enablingcompression and case-wide deduplication and thus enhancing the existing struc-tured storage handler. It truly uses the SQL-managed storage locations, includingbilling support. The handler also allows for multiple locations, e.g. drives or NASs, inorder to add storage space in case of full disks. This handler is also the default fornewly created projects.Furthermore, a workflow is supplied that allows the user to convert an existing NASstorage into a new NAS storage with different configuration parameters for storagedatabase, compression, and deduplication.

Changes to Default ConfigurationThe structured storage V2 is the default storage for new projects.The V2-handler comeswith its own set of configuration options, most of them areequal to other handlers (i.e. structured storage), except for the storage roots. Thesestorage roots are used to configure the directories where the objects are saved(known as storage location prefix for the structured storage), and it is possible togivemore than one directory where one of these has to be designated as "master".In order to enable case-wide deduplication, the handler will make use of the casename and client id as configured in the application configuration (or master serviceas fallback).

Known LimitationsThe new setup does not support "matter storage".

8.6 Performance Measures (Reference ID: CORE-16714)Behavior before change

Aggregated performance figureswere not collected or reported.

Behavior after changePerformancemeasures are collected, aggregated and periodically reported to logfile, a separate csv file, and via event notification to the Axcelerate 5 reporting data-base. This provides the ability to pro-actively detect high load situations andprovides insight into root causes for high load.

Changes to Default ConfigurationFour parameters have been added to the datamodel configuration underCommon> Event Notification and KPI Reporting to control the behavior of the collectionand generated reports. The added parameters are:

l Active-checkbox (default: on)l Interval for aggregation and reporting (default: 5 minutes)



l CSV-filename (default: "kpi.csv" in engine-directory)l Checkbox for reporting database (default: on)


8.7 Transym OCR Support (Reference ID: CORE-15448)Behavior before change

Latest supported Transym version was 4.0.0.9.

Behavior after changeLatest supported Transym version is 4.0.0.14.





9.1 AXC-34630 (Foldering panel)Summary New Behavior

Partially corrupted field valuescaused the retrieval of folders in theFoldering panel to fail, rendering thecomplete foldering feature unusable.

The corrupted part of the field values isignored and the field value is listed in theFoldering panel. Functionality is notimpacted by themissing part. The incident,including the ID and name of the corruptedfield value, is logged in the backend.

9.2 AXC-36584 (Review Workflows page)Summary New Behavior

Review Workflows page was notaccessible if internal field value prop-erties weremissing.

Review Workflows page is accessible asthe internal field value and its properties areinserted in the same transaction.

9.3 AXC-36619 (Printing)Summary New Behavior

Printing with theDocument Overlayoption could not be started, despite havinga valid text overlay.

Printing with theDocument Overlayoption and a valid text overlay nowworks.

9.4 AXC-36706 (Permission synchronization)Summary New Behavior

Group permissions could not be syn-chronized when there was a locked applic-ation.

Group permissions can be synchronizedif there is a locked application.



9.5 AXC-36811 (Bloomberg chat)Summary New Behavior

For Bloomberg chat documents con-taining invitationswith messages, the invit-ationswith messageswere not shown inText view.

Text view for Bloomberg chat doc-uments now displays both regular mes-sages aswell as invitationswithmessages.

9.6 AXC-37033 (Production workflows)Summary New Behavior

Valueswere not appearing in theAppliedRuleSmart Filter in production workflow pre-view mode, even though theApplied Rulecolumn of theResults list displayed the val-ues.

Values appear as expected in theApplied RuleSmart Filter while inproduction workflow preview mode.

9.7 AXC-37061 (Business Intelligence)Summary New Behavior

Billingmetrics were not updated and newBusiness Intelligence dashboards couldnot be created when there was ameta-engine project in the pod.

Billingmetrics are updated and BusinessIntelligence dashboards can be createdeven when there is ameta-engine pro-ject in the pod.

9.8 AXC-37090 (Pods with high number of applic-ations)


Reviewer efficiency, creation of new Busi-ness Intelligence dashboards, adaptivebatching, automatic set creation, auto-matic iteration creation and pre-conversionrules did not work for some projects whenthere weremore than 50 applications inthe pod.

Reviewer efficiency, creation of newBusiness Intelligence dashboards,adaptive batching, automatic set cre-ation, automatic iteration creation andpre-conversion rules should work withany number of applications.



9.9 AXC-37264 (Metadata panel)Summary New Behavior

Fields with a hyphen in the ID (e.g.Message-Id)were not parsed correctlyand could not be used in theMetadatapanel.

Fields with a hyphen in the ID (e.g.Message-Id) are parsed correctly andcan be used in theMetadata panel.

9.10 AXC-37270 (Performance)Summary New Behavior

The system tried to get display names for all fields thatwere listed in theView - modifiable fields table of thedocument model in CORE Administration, even forfields that were not needed for Axcelerate 5, therebydecreasing performance.

Now only fields that areused in either field viewgroup named axcel-erate.tagging or axcelerate.folderingare synchronized (plus ser-vice tier extra list of otherimportant Axcelerate 5fields). The performance ofdelete calls and updatecalls is increased.

9.11 AXC-37397 (Tagging rules)Summary New Behavior

The backend serversmaintain a local copyof all tagging rules. When one serverdeleted a rule from the database, the otherservers did not adapt their local copies andcontinued to use the outdated rule for dis-play aswell as validation.

The backend servers still maintain alocal copy of all tagging rules. The syn-chronization process from the data-base to these local copies nowsupports deletion.

9.12 AXC-37398 (Pre-Conversion)Summary New Behavior

A pre-conversion rule was running in con-tinuous loop for a corrupted document andlead to a newly created job every 5minutes.

A pre-conversion rule is no longer run-ning in continuous loop for a corrupteddocument.




Documents were not tagged as pro-duction exceptions in case of a productionerror (e.g. Conversion pageswas notset).

Now documents are tagged as pro-duction exceptions in case of a productionerror (e.g. Conversion pageswas notset).

9.14 AXC-37597 (Performance)Summary New Behavior

Concurrent tagging requests frommultiplereviewers caused the reload of temporarilycached tagging rules to be triggeredmul-tiple times (once for each tagging request).As these processeswere queued and notallowed to be run in parallel, some review-ers had to wait very long (up to timeouts inthe browser) until their individual processwas run. Because this process is tied to thetagging action itself, their review per-formance decreased.

The reload processes are still queued ina way, but subsequent processeswillnow detect that a previous processalready reloaded the tagging rules andthen stop immediately. Thus, while allreviewers will still have to wait a shorttime for one (i.e. the first) process tocomplete, they don't have to wait formultiple, redundant reloads anymore.

9.15 CORE-15304 (Redaction file references)Summary New Behavior

When re-publishing a redacted doc-ument with images, the redaction fileswere deleted but still referenced in thexml.

When re-publishing a redacted documentwith images, not only the redaction filesare deleted but also the references in thexml.



9.16 CORE-15992 (Property Post Processors)Summary New Behavior

When the recipient_counter was notlast on the Property Post Processors (PPP)list for the data source, the recipient counterfield was not populated with the correctcount.

The recipient_counter is nowlast on the Property Post Processors(PPP) list for the data source.Whenthe script updateLegacyAp-plication.bat runs as part of theupgrade procedure, it makes surerecipient_counter is last on thePPP list.

9.17 CORE-16389 (Documentum crawls)Summary New Behavior

Documentum crawls with custom querybut without folder or cabinet in URI werenot supported.

Documentum crawls with custom querybut without folder or cabinet in URI aresupported.

9.18 CORE-16439 (Wildcard expansion)Summary New Behavior

The data structure used for wildcardexpansion was not always updatedproperly causingmissing entries.

The sorted set data structure is now updatedproperly in all cases. Affected engines areautomatically repaired during engine startup.An E2message is logged if missing entriesare found.

9.19 CORE-16868 (Storage utilization statistics)Summary New Behavior

CORE storage collection for PDF andEXPORT used hard-coded directoriesand thus reported incorrect storage util-ization statistics.

COREore storage collection for PDF andEXPORT now use configured directoriesand storage utilization statistics are cor-rectly reported.



9.20 CORE-16975 (Box connector)Summary New Behavior

When a document in Box had versions and noexisting creator, no documents were indexedfor the versions of that document.

Versionswith no existing creator inBox are now indexed.

9.21 CORE-17050 (OCR jobs)Summary New Behavior

Canceling OCR jobs resulted in lots oftemp files (.tif) in the temp folder on theengine host.

Temp files produced by canceledOCRjobs are now cleaned up properly.

9.22 CORE-17106 (Bloomberg data)Summary New Behavior

Whenmultiple Bloomberg TXT fileswere found underneath a TXT dir-ectory only the first one was used. Inaddition, TXT files with an infix afterthe .[I]B.startdate-enddate sectionand between the TXT file suffix wereskipped.

Bloomberg TXT files with an arbitrary infixafter the .[I]B.startdate-enddate section arenow considered. In addition, support hasbeen added for subdirectories underneath aTXT parent directory with more than a singleTXT file. Subfolders containingmixedBloombergmessages and Bloomberginstant messageswith separate TXT files arealso considered.

9.23 CORE-17124 (Documentum crawls)Summary New Behavior

When crawling a Documentum source,the native storage failed for documentswithout content description.

When crawling a Documentum source,native storage succeeds for documentswithout content description.



9.24 CORE-17133 (Performance)Summary New Behavior

An internal warningmessage was dis-played too often causing the engine tobecome unresponsive.

The logging frequency of the internal warn-ingmessage has been reduced, therebyfixing the performance issue.


Ameta engine unnecessarily computedregex based highlighting expressions, lead-ing to unnecessarymemory consumptionthat could lead to out of memory situations.

A meta engine does not computeregex based highlighting expressionsas they are not needed in this case.


Results of an OCR job were stored locally ina processing unit specific temp directorythat was cleared when the respective pro-cessor crashed, resulting in a FileNotFoundException for the respective OCR doc-ument.

Results of an OCR job are now storedin a temp directory of the respectiveengine such that a crash of the OCRprocessor does not wipe out existingresults.

9.27 CORE-17316 (Stored page count)Summary New Behavior

Changing a textual tagging field (either in theuser interface or using a CSVmerge) causedthe stored page count to be removed.

The stored page count is not affectedanymore when other fields arechanged.


Documents with failed native conversionwere not always tagged as an exception andnot shown in the Conversion Smart Filter.

Documents with failed native con-version are now always tagged as anexception.



9.29 CORE-17458 (Wordmap)Summary New Behavior

Very long CJK field values could causea buffer overflow in theWordMap poten-tially causing the engine process tocrash.

The internal buffer in theWordMap hasbeen increased to prevent it from over-flowing. In addition, themaximum length ofa field value has been reduced.

9.30 CORE-17567 (Fetching files)Summary New Behavior

The system couldn't perform all file fetchoperations on faulty projects which containcompressed and uncompressed data.

The system can fetch both compressedand uncompressed files within a faultyproject.

9.31 CORE-17592 (Bloomberg)Summary New Behavior

Attachments in Bloomberg chat text files thatcontained any of the file ID delimiters "[" or "]"in their display namewere not processed cor-rectly and caused the document to be taggedwith anArchive entry processing error

Attachments in Bloomberg chat textfiles that contain any of the file IDdelimiters "[" or "]" in their displayname are now properly processed.




Duplicate user information in the headerof Bloomberg text was causing doc-uments to be tagged with theMiscellaneous parser error - alreadyseen mail address error.

A duplicate user information (with partiallyincomplete data) will now bemapped to itsmore complete counterpart during parsingof a Bloomberg text header.

The errorMiscellaneous parser error -already seen mail addresswill still betagged to documents which are associatedto Bloomberg text files that contain morethan one user for the same email address.


When the pipe ("|") character was part of a userdata attribute in a Bloomberg text file the cor-responding chat actions could not be resolvedand caused the document to be tagged withMiscellaneous parser error - unable toresolve user error.

Bloomberg chat with users whichcontain the pipe ("|") character intheir name is properly parsed now.

9.34 CORE-17639 (Tagging)Summary New Behavior

Executing a bulk tagging while a crawlerwas running could cause a deadlock in theindex engine.

There is no deadlockwhen executing abulk tagging while a crawler is running.



9.35 CORE-17683 (Memory)Summary New Behavior

There was amemory leakwhen execut-ing the same searchmultiple timeswiththe same session.

Thememory leakwhen executing thesame searchmultiple timeswith the samesession is fixed.

9.36 CORE-17757 (Batch server status detection)Summary New Behavior

Jobswere sent to batch servers that werein some undefined offline state. The res-ulting exception caused the job to fail butdid not refresh the online/offline status.

The corresponding exception causes thehost to be treated as offline. Furthermore,logging is improved to enable better track-ing of similar issues in the future.

9.37 CORE-17781 (Index engine write lock)Summary New Behavior

An exception thrown during an indexengine save could cause subsequentwrite lock attempts to fail.

Internal save counter is reduced in all casesnow, even if an exception occurs, andshould not cause write lock attempts to fail.

9.38 CORE-17788 (Categorization state display)Summary New Behavior

For hierarchical fields the displayed cat-egorization state (Agreed,Disagreed,Suggested) waswrong.

The categorization state is correctlyshown for both flat and hierarchicalfields.



9.39 CORE-18013 (Document history)Summary New Behavior

If text field valueswere changed using cer-tain scripts, the respective entry in theDocument history caused an exception inAxcelerate 5.

If text field values are changed usingscripts, the respective entry in theDocument history is displayed.

9.40 FOUND-7160 (Log messages)Summary New Behavior

Some logmessageswere formatted withlog4j (WARN/FATAL) priority.

Logmessages are formatted withoutlog4j priority.

9.41 FOUND-9536 (Application removal)Summary New Behavior

The removal of an application tree with ameta engine lead to error messages intheMasterService log (without any fur-ther impact).

The removal of an application tree con-taining ameta engine has been revised.Error messageswith no impact are not cre-ated in theMaster Service log.

9.42 FOUND-9588 (0 byte file handling)Summary New Behavior

NIST filter was applied for empty files. NIST filter is not applied for empty files.

9.43 FOUND-9644 (PDF date parsing)Summary New Behavior

Some formats of themodification date ofcomments in PDF documents could not beparsed.

Modification dates of comments in PDFdocuments get parsed correctly.




For the ECA/Ingestion application, theMatters tab loading speed was lowwith huge numbers of matters.

For the ECA/Ingestion application it is nowpossible to filter for last updatedmatters inorder to increase the loading speed of theMatters tab.


A large number (> 250k) of temp files from aterminated process caused an out ofmemory error in the launcher service whiledeleting those files.

The launcher service is able to clean upan arbitrary number of temp files after aprocess has terminated.


Oracle OutsideIn parser service cre-ated and left temp files in user temp dir-ectory.

Oracle OutsideIn parser service createstemp files in data source temp directorywhich is deleted automatically.

9.47 FOUND-10073 (Kerberos and Active Directorylogin)


The Kerberos and ActiveDirectory loginmodules used different formats for serverlists, leading to errors when copying andpasting from onemodule to another.

The Kerberos and ActiveDirectory loginmodules now use the same server listformat so that a value can be copied andpasted from onemodule to another.



9.48 FOUND-10099 (Data source start)Summary New Behavior

Data sources could no longer be started ifthey had been automatically resumed acouple of times before.

Data sources can be started inde-pendently of the number of previousautomatic resumes.

9.49 FOUND-10104 (Application case name)Summary New Behavior

New Applicationwizard truncatedCase Name property at the displayname's forward slash.

New Applicationwizard no longer trun-catesCase Name property at the displayname's forward slash.

9.50 FOUND-10177 (Configuration access)Summary New Behavior

In CORE Administration concurrentaccess by different processes to defaultconfiguration could fail.

In CORE Administration concurrentaccess by different processes to defaultconfigurations is now supported.

9.51 FOUND-10214 (SSO login)Summary New Behavior

Login via SSOcould cause the user sessionto terminate.

Login via SSOworks as expected.

9.52 FOUND-10291 (Exchange connector)Summary New Behavior

When using theMicrosoft Exchange con-nector, invalid characters in MicrosoftExchange webservice responses couldcause a crawl failure.

Invalid characters in MicrosoftExchange webservice responses arefiltered and do not cause a crawl failure.



9.53 FOUND-10308 (Monitoring)Summary New Behavior

In theMonitoring section of CORE Admin-istration, the HOST ID column on theDataSources tab showed the host's display name.

TheHOST ID column on theDataSources tab shows the host ID.

9.54 FOUND-10466 (Box data sources)Summary New Behavior

During Box data source crawlsmemory consumption for con-nections to Boxwas proportional tothe overall number of users to becrawled. For high number of usersthis could lead to Out-Of-Memoryissues.

Memory consumption for connections is nowproportional to the number of users that arecurrently being processed. This numbershould be smaller or equal to the number ofthreads used for crawling. For crawls in whichthe overall number of users is high, this resultsin a considerably smaller memory footprint,strongly decreasing the chance for a Out-Of-Memory issues.




10.1 Business Intelligence Data Reload (ReferenceID: AXC-36598)Behavior before change

Data for the Business Intelligence dashboard was loaded every 4 hours. A manualreload was not available.

Behavior after changeIn addition to the scheduled reload every 4 hours, the new Reload DashboardData button in the toolbar menu for anyBusiness Intelligence dashboard can beused to load the latest data into the dashboard.


10.2 Current Criteria Panel Enhancements (Refer-ence ID: AXC-37114)Behavior before change

The applied criteria section in theSearch panel only allowed to lock and deleteapplied criteria. To display the specific values, and to edit, the panel needed to bemaximized. Themaximized height took less than 50% of theSearch panel, andallowed to display <~10 different criteria without scrolling.

Behavior after changeThe applied criteria section in theSearch panel allows to display, lock, edit anddelete applied criteria, without having tomaximize the panel height. Themaximizedheight has been increased to 50% of theSearch panel, allowing to display >~10 dif-ferent criteria without scrolling.




10.3 "Responsive" Field Renamed (ReferenceID: AXC-37119)Behavior before change

For new matters, the default field to track responsiveness of a document wasnamedResponsive with yes and no values.

Behavior after changeFor new matters, the default field to track responsiveness of a document is namedResponsiveness withResponsive andNot responsive values.

Changes to Default ConfigurationThe default system template is now using the new display nameResponsivenessfor the ax_responsive field, and loads the default valuesResponsive andNotresponsive from a new OWL file.TheOWL file with yes and no values still exists, but is no longer used by the currenttemplate.


10.4 U Quick Tag (Reference ID: AXC-32949)Behavior before change

The L button applied the last complete tagging.

Behavior after changeThe L button continues to apply the last complete tagging.A new U button can be used to apply only the last changes to tagging.



10.5 Performance Measures (Reference ID: CORE-16714)Behavior before change

Aggregated performance figureswere not collected or reported.



Behavior after changePerformancemeasures are collected, aggregated and periodically reported to logfile, a separate csv file, and via event notification to the Axcelerate 5 reporting data-base. This provides the ability to pro-actively detect high load situations andprovides insight into root causes for high load.

Changes to Default ConfigurationFour parameters have been added to the datamodel configuration underCommon> Event Notification and KPI Reporting to control the behavior of the collectionand generated reports. The added parameters are:

l Active-checkbox (default: on)l Interval for aggregation and reporting (default: 5 minutes)l CSV-filename (default: "kpi.csv" in engine-directory)l Checkbox for reporting database (default: on)


10.6 Transym OCR Support (Reference ID: CORE-15448)Behavior before change

Latest supported Transym version was 4.0.0.9.

Behavior after changeLatest supported Transym version is 4.0.0.14.


10.7 SAML 1.1 AuthN in Axcelerate 5(Reference ID: Found-9480)Behavior before change

SAML authentication was not supported for Axcelerate 5.

Behavior after changeSAML 1.1 authentication is now supported for Axcelerate 5.






When adding documents to a production work-flow while many taggings are still pending(long coding queue), the user receives atimeout error though the actual operation is stillrunning in the background and typically com-pletes successfully at a later time.

The "add to production workflow"action finishes immediately and theactual operation continues to run inthe background.

11.2 AXC-34630 (Foldering panel)Summary New Behavior

Partially corrupted field valuescaused the retrieval of folders in theFoldering panel to fail, rendering thecomplete foldering feature unusable.

The corrupted part of the field values isignored and the field value is listed in theFoldering panel. Functionality is notimpacted by themissing part. The incident,including the ID and name of the corruptedfield value, is logged in the backend.

11.3 AXC-35769 (Review workflows)Summary New Behavior

Calls to retrieve document counts onReview Workflow pages for large reviewworkflows caused timeouts and heavyserver loads.

Sequential calls to retrieve documentscounts onReview Workflow pages donot cause timeouts and heavy serverloads.

11.4 AXC-36156 (Navigation page)Summary New Behavior

When paging to a page other than thefirst one of theNavigation page (mat-ter list), the user was not able to log in toamatter.

When paging to another page than the firstone of theNavigation page (matter list),the user will see all availablematters and isable to log in to eachmatter.



11.5 AXC-36439 (Highlighting)Summary New Behavior

When reviewing documents in two-screenmode, the highlight summarybutton wasmissing.

The highlight summary button is shown inboth single-screen and two-screenmodeon theAnalysis andReview pages.

11.6 AXC-36584 (Review workflows)Summary New Behavior

Review Workflow page was notaccessible if internal field value prop-erties weremissing.

Review Workflow page is accessible asthe internal field value and its properties areinserted in the same transaction.


Printing with theDocument Overlayoption could not be started, despite havinga valid text overlay.

Printing with theDocument Overlayoption and a valid text overlay nowworks.

11.8 AXC-36681 (Document access)Summary New Behavior

When a filter was applied which invalidated abatch restriction, a reviewer could potentiallyaccess restricted documents.

A reviewer can only see documentsassigned to their currently selectedbatch.

11.9 AXC-36786 (Jobs page performance)Summary New Behavior

Whenmany jobswere present on theJobs page, it could potentially take toolong to load and become inaccessible tothe user.

Jobs page loads approximately 5-10times faster and does not become inac-cessible whenmany jobs are present.



11.10 AXC-36806 (Navigation page)Summary New Behavior

With large amounts of groups and/orapplications and with multiple servicetiers, timeouts and optimistic lockingexceptions could occur. As a con-sequence, updates to theNavigationpagematter list were not executed.

Even with large amount of groups andapplications, the syncwill succeed andupdate theNavigation pagematter list.Multiple service tiers instanceswill notsync in parallel and thus no optimistic lock-ing exceptions are logged.

11.11 AXC-36811 (Bloomberg chat)Summary New Behavior

For Bloomberg chat documents con-taining invitationswith messages, the invit-ationswith messageswere not shown inText view.

Text view for Bloomberg chat doc-uments now displays both regular mes-sages aswell as invitationswithmessages.

11.12 AXC-36899 (Security vulnerability in 7zip 9.20)Summary New Behavior

7zip version shipped with Axcelerate 5 was9.20, which had known security vul-nerabilities.

7zip version is 16.0, which fixes theknown security vulnerabilities.


Valueswere not appearing in theAppliedRuleSmart Filter in production workflow pre-view mode, even though theApplied Rulecolumn of theResults list displayed the val-ues.

Values appear as expected in theApplied RuleSmart Filter while inproduction workflow preview mode.




Billingmetrics were not updated and newBusiness Intelligence dashboards couldnot be created when there was ameta-engine project in the pod.

Billingmetrics are updated and BusinessIntelligence dashboards can be createdeven when there is ameta-engine pro-ject in the pod.

11.15 AXC-37095 (Review batches)Summary New Behavior

Marking batches as complete calculatedthe review count for all batches in the pro-ject instead of only the selected ones,thereby reducing the execution speed ofthe operation and resulting in browsertimeouts.

Review count is now calculated only forthe selected batches rather than allbatches, significantly improving the exe-cution speed of the operation and avoid-ing unnecessary browser timeouts.

11.16 AXC-37264 (Metadata panel)Summary New Behavior

Fields with a hyphen in the ID (e.g.Message-Id)were not parsed correctlyand could not be used in theMetadatapanel.

Fields with a hyphen in the ID (e.g.Message-Id) are parsed correctly andcan be used in theMetadata panel.

11.17 AXC-37397 (Tagging rules)Summary New Behavior

The backend serversmaintain a local copyof all tagging rules. When one serverdeleted a rule from the database, the otherservers did not adapt their local copies andcontinued to use the outdated rule for dis-play aswell as validation.

The backend servers still maintain alocal copy of all tagging rules. The syn-chronization process from the data-base to these local copies nowsupports deletion.



11.18 CORE-15070 (Engine starts)Summary New Behavior

A defect in the 3rd party software Liquibasecould result in a failure to release SQLlockswhich blocked all engines during thestartup procedure.

The schemas are now updated duringthe software update, avoiding blockedengines during the startup procedure.

11.19 CORE-15749 (Bloomberg email thread detec-tion)


Thread detection of larger Bloombergemails could cause data sources andengines to run out of memory.

Thread detection of larger Bloombergemails does not cause data sources andengines to run out of memory.

11.20 CORE-15992 (Property Post Processors)Summary New Behavior

When the recipient_counter was notlast on the Property Post Processors (PPP)list for the data source, the recipient counterfield was not populated with the correctcount.

The recipient_counter is nowlast on the Property Post Processors(PPP) list for the data source.Whenthe script updateLegacyAp-plication.bat runs as part of theupgrade procedure, it makes surerecipient_counter is last on thePPP list.

11.21 CORE-16008 (Engine save)Summary New Behavior

A rules.xml containing rules ref-erencing non-existing field values couldcause a deadlock during an engine save.

A rules.xml containing rules ref-erencing non-existing field values nolonger causes a deadlock during anengine save.



11.22 CORE-16439 (Wildcard expansion)Summary New Behavior

The data structure used for wildcardexpansion was not always updatedproperly causingmissing entries.

The sorted set data structure is now updatedproperly in all cases. Affected engines areautomatically repaired during engine startup.An E2message is logged if missing entriesare found.

11.23 CORE-16460 (Engine save)Summary New Behavior

Triggering both amanual shutdown andan emergency shutdown could lead to afailing save job.

Triggering both amanual shutdown andan emergency shutdown no longer leadsto a failing save job.


Update of engine statistics could becomestuck if engine was very busy.

The updating of engine statistics per-formance under load has been improved.

11.25 CORE-16613 (Field value deletion)Summary New Behavior

Deleting a huge amount of field values inbulk caused a delay in engine startup.

Deleting a huge amount of field values inbulk no longer causes a delay in enginestartup.

11.26 CORE-16620 (Search)Summary New Behavior

Serializing huge search requests to indexpartitions resulted in out-of-memory situ-ations since the requests did not fit intothe smaller java heap.

Search requests are only sent to theindex partitions if there is enough javaheap left; otherwise, they are held backuntil another search returns.



11.27 CORE-16681 (Junk detection)Summary New Behavior

Junk detection occasionally producedinvalid document XMLwhile replacingjunkwith replacement text.

New documents always have valid XML,and old documents with invalid XML arerepaired during publish and document dis-play.

11.28 CORE-16754 (Training data jobs)Summary New Behavior

The flush post processing job could beblocked by potentially long running trainingdata job. As a consequence, concurrentread operations (i.e., document fetchingand searches) could be blocked aswell.

The flush post processing job is notblocked anymore by training data jobs.In addition, the training data jobs are run-ningmuch faster now if the small flushfeature is active.

11.29 CORE-16755 (Coding queue performance)Summary New Behavior

Rules processing had a large RMI over-head, causing a negative impact on thecoding queue performance.

The RMI overhead of rules processing isimproved, reducing the negative impacton the coding queue performance.

11.30 CORE-16765 (CSV merge)Summary New Behavior

When using CSVmerge on a project that is pre-pared for storagemigration but not yet migrated,if themerge replaced the existing imageswithfewer images, some old images remained in thestorage.

CSVmerge no longer leads toorphaned images in storage.




Some index calls involving the field valueexistence checks inmergingmeta engineswere slow.

Performance of the field value existencechecks inmergingmeta engines isimproved.

11.32 CORE-16868 (Storage utilization statistics)Summary New Behavior

CORE storage collection for PDF andEXPORT used hard-coded directoriesand thus reported incorrect storage util-ization statistics.

COREore storage collection for PDF andEXPORT now use configured directoriesand storage utilization statistics are cor-rectly reported.


Conversionswith multiple binary records(especially image views) had long latency ifbinary recordswere served fromS3.

Multiple streams are now downloadedconcurrently, thereby reducing thelatency.

11.34 CORE-16975 (Box connector)Summary New Behavior

When a document in Box had versions and noexisting creator, no documents were indexedfor the versions of that document.

Versionswith no existing creator inBox are now indexed.


Canceling OCR jobs resulted in lots oftemp files (.tif) in the temp folder on theengine host.

Temp files produced by canceledOCRjobs are now cleaned up properly.



11.36 CORE-17106 (Bloomberg data)Summary New Behavior

Whenmultiple Bloomberg TXT fileswere found underneath a TXT dir-ectory only the first one was used. Inaddition, TXT files with an infix afterthe .[I]B.startdate-enddate sectionand between the TXT file suffix wereskipped.

Bloomberg TXT files with an arbitrary infixafter the .[I]B.startdate-enddate section arenow considered. In addition, support hasbeen added for subdirectories underneath aTXT parent directory with more than a singleTXT file. Subfolders containingmixedBloombergmessages and Bloomberginstant messageswith separate TXT files arealso considered.


An internal warningmessage was dis-played too often causing the engine tobecome unresponsive.

The logging frequency of the internal warn-ingmessage has been reduced, therebyfixing the performance issue.

11.38 CORE-17187 (Storage size calculation)Summary New Behavior

Projects with storage compression pop-ulated the rm_nativestoragesizefield with the compressed sizes instead ofthe file size. This happened during crawloperations.

rm_nativestoragesize is pop-ulated with the "managed size," i.e., nor-mal file size. For existing projects, this isautomatically corrected at the nextengine restart.

11.39 CORE-17220 (Native files)Summary New Behavior

The publish wizard configured new publishengines incorrectly for SaaS-based S3 pro-jects, resulting in inaccessible published nat-ives.

The publish engines are configuredcorrectly and the published natives areaccessible.




Ameta engine unnecessarily computedregex based highlighting expressions, lead-ing to unnecessarymemory consumptionthat could lead to out of memory situations.

A meta engine does not computeregex based highlighting expressionsas they are not needed in this case.


Results of an OCR job were stored locally ina processing unit specific temp directorythat was cleared when the respective pro-cessor crashed, resulting in a FileNotFoundException for the respective OCR doc-ument.

Results of an OCR job are now storedin a temp directory of the respectiveengine such that a crash of the OCRprocessor does not wipe out existingresults.

11.42 CORE-17316 (Stored page count)Summary New Behavior

Changing a textual tagging field (either in theuser interface or using a CSVmerge) causedthe stored page count to be removed.

The stored page count is not affectedanymore when other fields arechanged.

11.43 CORE-17449 (Storage SQL feature)Summary New Behavior

The storage SQL feature could be dis-abled by themaster switch even if the pro-ject was not empty.

The storage SQL feature cannot be dis-abled by themaster switch if the projectis not empty.

11.44 FOUND-7160 (Log messages)Summary New Behavior

Some logmessageswere formatted withlog4j (WARN/FATAL) priority.

Logmessages are formatted withoutlog4j priority.



11.45 FOUND-9588 (0 byte file handling)Summary New Behavior

NIST filter was applied for empty files. NIST filter is not applied for empty files.

11.46 FOUND-9644 (PDF date parsing)Summary New Behavior

Some formats of themodification date ofcomments in PDF documents could not beparsed.

Modification dates of comments in PDFdocuments get parsed correctly.


For the ECA/Ingestion application, theMatters tab loading speed was lowwith huge numbers of matters.

For the ECA/Ingestion application it is nowpossible to filter for last updatedmatters inorder to increase the loading speed of theMatters tab.


A large number (> 250k) of temp files from aterminated process caused an out ofmemory error in the launcher service whiledeleting those files.

The launcher service is able to clean upan arbitrary number of temp files after aprocess has terminated.


Oracle OutsideIn parser service cre-ated and left temp files in user temp dir-ectory.

Oracle OutsideIn parser service createstemp files in data source temp directorywhich is deleted automatically.




Large patch files deployed for a large num-ber of processes caused themaster serviceto run out of memorywhen a patch reportwas created.

Memory utilization in themaster ser-vice is optimized while generating alarge patch report.

11.51 FOUND-10104 (Application case name)Summary New Behavior

New Applicationwizard truncatedCase Name property at the displayname's forward slash.

New Applicationwizard no longer trun-catesCase Name property at the displayname's forward slash.



12 New Feature Descriptions in version 5.7.0Update 1

12.1 JAVA update to JDK 8 Update 92(Reference ID: Found-9744)Behavior before change

JDK bundled with the CORE package was a JAVA previous version.

Behavior after changeJDK version 8 Update 92 is bundled with the CORE package. Please reference theOracle release notes for further information:[http://www.or-acle.com/technetwork/java/javase/8u92-relnotes-2949471.html].



12.2 SAML 1.1 AuthN in Axcelerate 5(Reference ID: Found-9480)Behavior before change

SAML authentication was not supported for Axcelerate 5.

Behavior after changeSAML 1.1 authentication is now supported for Axcelerate 5.




12 New Feature Descriptions in version 5.7.0 Update 1

http://www.oracle.com/technetwork/java/javase/8u92-relnotes-2949471.html

http://www.oracle.com/technetwork/java/javase/8u92-relnotes-2949471.html

12.3 Unlimited Number of Fields per Type(Reference ID: CORE-16491)Behavior before change

Number of fields in any given engine was limited to 256.

Behavior after changeAn index engine where the number of fields is enlarged is now automaticallyupgraded at engine start. If the number of fields changes across a power of 2, theupgrade processmay take some time, since a number of data structures need to beconverted.

Changes to Default ConfigurationNone, existing automatic update to data structure now works up to 65,000+ insteadof up to 256.


12.4 Box Connector v2(Reference ID: CORE-14333)Behavior before change

Box connector was running on deprecated Box API.

Behavior after changeIBox connector was completely refactored and is now running on themost currentversion of the Box API (as of May 2016).


Known LimitationsKnown limitations are described in the Box connector documentation.


12 New Feature Descriptions in version 5.7.0 Update 1

13 Issues Resolved in 5.7.0 Update 1

13.1 AXC-34980 (Viewer)Summary New Behavior

Viewer sometimes did not load with theerror: "Unable to verify the integrity of theconfiguration data", due to an encryptionissue when checking the signature of theconfig xml.

Retry logic has been implemented whichgenerates a new signed configuration inthe case of a decryption failure, therebyavoiding that the error occurs a secondtime.


Printing in Internet Explorer 11 did not workbecause the image download responses hadno content type header.

The content type header is now set inthe BravaServerProxy.

13.3 AXC-36357 (Production rules)Summary New Behavior

When accessing the the rules tab on theproduction workflows page for workflowswith a high number of documents, arequest timeout could occur.

Only documents where the rule tagginghas changed are updated when accessingthe rules tab, thereby reducing taggingoperations and the risk of timeouts.


Deleting a user from the database and logginginto the application again as that user leads tothe generation of a new UUID which pre-vented the user from accessing their Qliksheets.

In addition, the entire user directory was syn-chronized rather than a specific user.

A user will get the sameUUID evenif the user attributes are deleted fromthe database.

In addition, only the current userwho is a casemanager will be syn-chronized in Qlik.


13 IssuesResolved in 5.7.0 Update 1

13.5 AXC-36464 (Report preview)Summary New Behavior

Downloading complex reports couldcause anOut Of Memory Error on theApp Server.

The downloading of complex reports is nowstreamed and the preview is limited to 100values per axis to prevent Out Of MemoryErrors.

13.6 AXC-36557 (Comments field)Summary New Behavior

The comments text box in the taggingpanel did not wrap in newer Chrome ver-sions.

The comments text box in the taggingpanel now wraps in newer Chrome ver-sions.


Production folder could not be retrieved if aproduction workflow was renamed and aproduction for this workflow was startedfrom the Classic UI.

Production groups belonging to pro-duction workflows in AXC 5 are not dis-played anymore in the Classic UI.

13.8 AXC-36648 (Associated Results)Summary New Behavior

The associated results view could high-light the parent document of a selecteddocument, rather than the selected doc-ument, if the associated results listexceeded the 500 documents limit.

If the associated results list exceeds the 500documents limit, and the selected doc-ument is outside the viewable range, theselected document is not highlighted andthe user sees the error message: "Selecteddocument is outside of viewable range."



13.9 AXC-36676 (Tagging panel)Summary New Behavior

UI does not indicate to the user if theTagging panel is loading.

UI blocks the review panel and indicatesstatus to the user if the Tagging panel is load-ing.

13.10 AXC-36682 (Filters on Assignments page)Summary New Behavior

The Target Review StateSmart Filteron theAssignments page shows thetotal number of documents in the projectto all users.

The Target Review StateSmart fFlterand theAssignee Smart Filter on theAssignments page now show anApplylink rather than document counts.

13.11 AXC-36688 (Matter access)Summary New Behavior

If no pod base URL could be retrieved orwas not set by the sys admin, the userwas not able to access her matter.

If no pod base URL could be retrieved orwas not set by the sys admin, the user willaccess thematter in the same tab.

13.12 AXC-36899 (Security vulnerability in 7zip 9.20)Summary New Behavior

7zip version shipped with Axcelerate 5 was9.20, which had known security vul-nerabilities.

7zip version is 16.0, which fixes theknown security vulnerabilities.



13.13 CORE-16577 (Tagging processing)Summary New Behavior

The coding queue growth was caused by slowprocessing of change queue entries. This wascaused by lock contention on the index enginedue to amixture of *many* read requests and*long* read requests. Furthermore, the dur-ation of the periodically executed asyn-chronous Save - where no coding can beprocessed - was quite long. Additionally weobserved reviewer impact of the long queuewhich wasmainly caused by conversion tag-ging.

The processing speed of the save issignificantly improved, minimizingthe pause in potential processing oftagging (coding). In order to reducethe impact of the coding queue onthe reviewers several changes arenow in effect:

l The data structures for access-ing pending coding informationare reconfigured to allow forfaster access.

l The conversion tagging are pro-cessed without blocking writelocks.

l The locking of the last result vec-tors is optimized.

l The usage of in-memory struc-tures is broadened.


When base64 junkwas distributed overmultiple <text><p> tag sequences, theterm based junk detection did not engage.

XML tags between potential junk termsare now skipped during junk detection sojunk is recognized across tags.

13.15 CORE-16712, AXC-36811 (Bloomberg invit-ations)


Chats containing invites were alwaystagged with a single value even if theycontainedmessages.

Two separate values, "InviteWithMessageSent" and "InviteWithout Message Sent",are now used for documents with invites.



13.16 FOUND-9171 (Inconsistent configurations)Summary New Behavior

Due to a template relation con-dition, it was possible to createinconsistent configurations.

The template relation condition has been fixedand it is no longer possible to create incon-sistent configurations. Additionally a script isprovided tomanually fix broken projects byinheriting specified template values.

13.17 FOUND-9568 (Service tier)Summary New Behavior

The service tier startup blocked forover an hour when the jgroups clusterjoin processwas stuck.

When the jgroups cluster join process isstuck, the service tier startup block is limitedto 10minutes and terminates after this time,freeing the startup synchronizer lock.

13.18 FOUND-9633 (Large cache)Summary New Behavior

Cache of presentation objects was growingover time.

Cache is now self-cleaning.

13.19 FOUND-9646 (Cleanup after crawl)Summary New Behavior

Data sources alwayswrote checkpointinformation which could lead to time con-suming cleanup at end of crawl.

Data sourceswrite checkpoint inform-ation only if started with enabled "Auto-matic resume" option.

13.20 FOUND-9670 (Templates)Summary New Behavior

Extended template was not used in datasource creation via web service.

Extended template is correctly used indata source creation via web service.



13.21 FOUND-9682 (Oracle OutsideIn operations)Summary New Behavior

Oracle OutsideIn operations (parsing, MIMEtype detection, conversion) failed if runningasWindows service.

Oracle OutsideIn operations runningasWindows service do not fail.

13.22 FOUND-9750 (Stored search)Summary New Behavior

Error message regarding a storedsearch without search parameters didnot reference the stored search.

Error message regarding a stored searchwithout search parameters now includesname of the stored search.

13.23 FOUND-9763 (Engine user session)Summary New Behavior

Engine user session was left untouchedduring long running count operation, caus-ing user to get logged out.

Engine user session is touched duringlong running count operation so userdoes not get logged out.




14.1 Production Option "Extracted text vs. Slip-sheet text(Reference ID: AXC-10527)Behavior before change

The production text for documents produced as a slip-sheet matched the slip-sheettext. When doing a production of documents lacking a native, but matching a rulerequestingNative and Image output, only the image was provided. (While a rulewith onlyNative output would result in the creation of a native placeholder for thesedocuments.)

Behavior after changeThe user has the option to either use the document text when producing with slip-sheet output or to use the slip-sheet text (default, and used for all legacy rules).If a rule with document text output will be used for a redacted document, the cor-responding production cannot be started, and the user is informed via a notification,with direct links tomodify the settings of the rule or production workflow.Whendoing a production of documents lacking a native, but matching a rule requestingNative and Image output, the image is provided along with a native placeholderfile. (Same aswhen using a rule with onlyNative output which also results in the cre-ation of a native placeholder for these documents.)

Changes to Default ConfigurationThere are no changes to the project-level configuration (in CORE Administration).In the application, the default for new rules is to use the Text matches slip-sheetoption. This is also used for existing production format rules as thismatches the leg-acy behavior.




14.2 UI Support to Re-convert Documents(Reference ID: AXC-20497)Behavior before change

Documents could not be re-converted by end users through the Axcelerate 5 userinterface. Command line scripts were needed for this task, which was time con-suming and potentially error prone.The twomain use caseswhen a re-conversion is needed are:1. Re-try failed conversions: To fix conversion errors, the users needed to fix thecause - e.g. extend a timeout, repair the document's native, or apply a software fix,and then run a commandline script to re-convert the relevant documents, (e.g., iden-tified through aWorkspace filter value).2. Apply updated conversion settings: To apply updated conversion settings - suchas the display of tracked changes for Microsoft Word documents - the user neededto apply the change of the setting, restart the engine, and then run a commandlinescript to re-convert for the relevant documents, (e.g., identified through aWorkspace filter value), making sure that no redacted documents were included asthismight cause shifted or lost redactions. The commandline script had dangerousoptions such as dropping redactions, or applying the old redactions to new rep-resentations of the document (whichmight have changed).

Behavior after changeDocuments can be re-converted by end users through the Axcelerate 5 user inter-face, allowing them to fix temporary errors of the environment or software, or toapply updated conversion settings.The twomain use caseswhen a re-conversion is needed are:1. Re-try failed conversions: To fix conversion errors, the users needs to fix thecause - e.g., extend a timeout, repair the document's native, or apply a software fix,and after running a search / selecting the documents, use theConvert wizard to re-convert documents with a failed conversion. Successfully converted documents areskipped, and so are oneswith a redaction on the native representation.2. Apply updated conversion settings: To apply updated conversion settings - suchas the display of tracked changes for Microsoft Word documents - the user needs toapply the change of the setting, restart the engine, and after running a search /selecting the documents, use theConvert wizard to re-convert documents with afailed conversion. Documents with a redaction on the native representation are auto-matically skipped. Documents with redactions cannot be re-converted through theUI, thus avoiding the risk of loosing redactions or applying redactions to shifted rendi-tions of the document.


Known LimitationsImages created via the classic Axcelerate Analysismodule cannot be reconverted.



14.3 Document Flagging for Jobs(Reference ID: CORE-15701)Behavior before change

Documents processed by a job were not flagged and these documents could hencenot be retrieved effectively fromwithin the Axcelerate 5 user interface. For somejobs, the effect of a job execution was not visible at all (e.g. when bulk printing doc-uments); for others the effect could be seen (e.g. using theConversion Smart Fil-ter when converting a document). But even in these cases a repeated executionwas invisible aswell (e.g. converting a documents 1 or 2 times resulted in the sameflagging in theConversion Smart Filter).

Behavior after changeDocuments processed by specific jobs are tagged into a new Job Processing fieldand can be retrieved fromwithin the Axcelerate 5 using a new Job ProcessingSmart Filter in theWork Product Smart Filter group.Tagging is available for these job types:

l Conversion (only for Bulk, not for on-the-fly jobs)l Global Redactionl Redaction rRemovall Productionl Printingl OCR

Changes to Default ConfigurationAn additional default field is used for the Job Processing field, with the internal idrmJobStatus.


14.4 Axcelerate - Continuous Improvements(Reference ID: AXC-35731)Behavior before change

The indication of the boundaries between families was sometimes hard to see, espe-cially in caseswhen only parts of families are shown in the result set (i.e., when theroot doc ismissing)In an Associated Results list (e.g. when using theReview page in 2-screenmode),the buttons to navigate to the next main result are only shown onmouse-over.Unless the user is aware of this, the existence of this functionality is usuallymissed.



Behavior after changeThe indication of the boundaries between families has been improved bymaking theboundary within a family lighter, and by replacing the gradient boundary betweenfamily by a dashed line.In an Associated Result list (e.g. when using theReview page in 2-screenmode),the buttons to navigate to the next main result are always shown, andmaking theuser aware of this functionality. Onmouse-over, the opacity for these buttons isincreased to highlight them further.

14.5 Unlimited Number of Fields per Type(Reference ID: CORE-16491)Behavior before change

Number of fields in any given engine was limited to 256.

Behavior after changeAn index engine where the number of fields is enlarged is now automaticallyupgraded at engine start. If the number of fields changes across a power of 2, theupgrade processmay take some time, since a number of data structures need to beconverted.

Changes to Default ConfigurationNone, existing automatic update to data structure now works up to 65,000+ insteadof up to 256.


14.6 Classic user interfaces run inside Tomcat 8application(Reference ID: FOUND-983)Behavior before change

TheClassic user interfaces (Classic Review & Analysis, Axcelerate Ingestion, ECA,CORE Administration, CaseManagement Tool, Perceptiv) run in Tomcat 6.

Behavior after changeTheClassic user interfaces (Classic Review & Analysis, Axcelerate Ingestion, ECA,CORE Administration, CaseManagement Tool, Perceptiv) run in Tomcat 8. TheAxcelerate 5 user interface continues to run in Tomcat 7.



Changes to Default ConfigurationIncreasedmetaspace default is set to 512MB as Tomcat 8 has a higher memory con-sumption.

Known LimitationsSupport for the classic Axcelerate Analysis and Axcelerate Review modules is lim-ited to 10 concurrent reviewers., These interfaces are only used during themigra-tion of cases to the Axcelerate 5 user intercae and not intended for massivedocument review. .

14.7 Retry mini job execution in case of batchserver failure(Reference ID: CORE-14508)Behavior before change

Job execution is usually done as a sequence of steps that are calledmini jobs, typ-ically a single processing job for each document in the executed job. For most jobs,such amini job is sent from aCORE engine server to a batch server for remote exe-cution. A batch server outage resulted in reported exceptions for all mini jobs in theexecuted job.

Behavior after changeMini jobs affected by outages are now automatically retried on a different batchserver. There is no longer any exception reported since themini job is now suc-cessfully executed. The corresponding batch server is tracked as unavailable until itbecomes online again.Affected job types include:

l Conversionl Productionl Printingl OCRl Bulk redaction






15.1 AXC-27381 (Bulk OCR/Production time OCR)Summary New Behavior

When Transym 4 fails to OCR a pagewith error code 8, this caused the wholedocument to be flagged as an exception.

When Transym 4 fails to OCR a page witherror code 8, the corresponding page isreplaced with this text: "OCR of pagefailed".

15.2 AXC-32122 (Robustness and stability)Summary New Behavior

Cache of UserCaseConfiguration causedOOMdue to duplicated strings.

Cache now contains less redund-ancies.


Viewer sometimes did not load with theerror: "Unable to verify the integrity of theconfiguration data", due to an encryptionissue when checking the signature of theconfig xml.

Retry logic has been implemented whichgenerates a new signed configuration inthe case of a decryption failure, therebyavoiding that the error occurs a secondtime.

15.4 AXC-36295 (Metadata fetching)Summary New Behavior

Metadata fields for a document werefetched from the engine.

Metadata fields for a document are nowfetched from the cache, thereby improvingperformance.



15.5 AXC-36357 (Production rules)Summary New Behavior

When accessing the the rules tab on theproduction workflows page for workflowswith a high number of documents, arequest timeout could occur.

Only documents where the rule tagginghas changed are updated when accessingthe rules tab, thereby reducing taggingoperations and the risk of timeouts.


Deleting a user from the database and logginginto the application again as that user leads tothe generation of a new UUID which pre-vented the user from accessing their Qliksheets.

In addition, the entire user directory was syn-chronized rather than a specific user.

A user will get the sameUUID evenif the user attributes are deleted fromthe database.

In addition, only the current userwho is a casemanager will be syn-chronized in Qlik.

15.7 AXC-36461 (Highlighting)Summary New Behavior

Metadata highlighting for documentsearch did not work with Regular Expres-sion.

Metadata highlighting works for doc-ument search with Regular Expression.

15.8 AXC-36535 (Robustness and stability)Summary New Behavior

Arrangement descriptions and associated datawere loaded intomemory and stored spe-cifically for each user and project. Thesedescriptionswere never unloaded frommemory.With a growing number of users andprojects, thememory consumption grewaccordingly, eventually causing out-of-memorysituations.

Arrangement descriptions areunloaded frommemorywhen theyhave been idle (i.e. have not beenaccessed in someway) for 30minutes.



15.9 AXC-36557 (Comments field)Summary New Behavior

The comments text box in the taggingpanel did not wrap in newer Chrome ver-sions.

The comments text box in the taggingpanel now wraps in newer Chrome ver-sions.


Sporadically, view loads broke on loadingthemarkup file.

Improved closing of views before reuseand error handling.


Production folder could not be retrieved if aproduction workflow was renamed and aproduction for this workflow was startedfrom the Classic UI.

Production groups belonging to pro-duction workflows in AXC 5 are not dis-played anymore in the Classic UI.

15.12 AXC-36648 (Associated Results)Summary New Behavior

The associated results view could high-light the parent document of a selecteddocument, rather than the selected doc-ument, if the associated results listexceeded the 500 documents limit.

If the associated results list exceeds the 500documents limit, and the selected doc-ument is outside the viewable range, theselected document is not highlighted andthe user sees the error message: "Selecteddocument is outside of viewable range."

15.13 AXC-36676 (Tagging panel)Summary New Behavior

UI does not indicate to the user if theTagging panel is loading.

UI blocks the review panel and indicatesstatus to the user if the Tagging panel is load-ing.



15.14 AXC-36682 (Filters on Assignments page)Summary New Behavior

The Target Review StateSmart Filteron theAssignments page shows thetotal number of documents in the projectto all users.

The Target Review StateSmart fFlterand theAssignee Smart Filter on theAssignments page now show anApplylink rather than document counts.

15.15 AXC-36688 (Matter access)Summary New Behavior

If no pod base URL could be retrieved orwas not set by the sys admin, the userwas not able to access her matter.

If no pod base URL could be retrieved orwas not set by the sys admin, the user willaccess thematter in the same tab.

15.16 AXC-36712 (Review performance)Summary New Behavior

Each click onSave+Next unreviewedbutton triggered a back end call to determ-ine the next unreviewed document in thelist.

Before using the back end to determinethe next unreviewed document, the list ofdocuments on client side will be used.

15.17 AXC-36732 (Associated results display)Summary New Behavior

The document displayed did not match the high-lighted document in the associated results view ontheAnalysis page.

The display is now correct.

15.18 AXC-36733 (Review performance)Summary New Behavior

Single document tagging resulted in anexpansive call to determine theMIME type of that document.

Single document tagging now sends theMIME type of that document with thesave request.



15.19 CORE-14974 (Text field used for enrichmentduring publish)


When using a text field for enrichment andstarting a publish with more than 1000 doc-uments, the engine ran into an Illeg-alStateException due to a safetymeasurethat blocked enriching text fields for morethan 1000 documents.

When using a text field for enrichmentand starting a publish with more than1000 documents, the engine does notcause an exception.

15.20 CORE-15862 (LiveLink connector)Summary New Behavior

LiveLink connector option Ignore MIMEtypewas not functioning.

LiveLink connector option Ignore MIMEtype functions.

15.21 CORE-16000 (Jobs)Summary New Behavior

The system didn't ensure the proper jobpriority with the following order: 1. "OnThe Fly Conversion", 2. "Bulk Print", 3."Bulk Conversion

The system ensures the proper job pri-ority with the following order: 1. "On TheFly Conversion", 2. "Bulk Print", 3. "BulkConversion"


Clicking on the last document of a searchpage caused the index to load unne-cessarily high amounts of data into the javaheap.

Unnecessarymemory consumption hasbeen eliminated and index routines havebeenmademore robust.



15.23 CORE-16071 (Large searches)Summary New Behavior

Large boolean searches for numeric iden-tifiers of the form "ID=<value1> AND ID=-=<value2>..." were not optimized andresulted in highmemory consumption.

Large boolean searches for numeric iden-tifiers of the form "ID=<value1> AND ID=-=<value2>..." are now optimized andresult in lessmemory consumption.

15.24 CORE-16101 (Query exception handling)Summary New Behavior

Improper exception handling for verycomplex queries could cause the engineto crash

Exceptions caused by very complex quer-ies are now handled properly and do notcrash the engine.

15.25 CORE-16218 (Field value sorting)Summary New Behavior

Sorting field values by property couldcause an exception.

Sorting field values by property no longercauses an exception.

15.26 CORE-16303 (SharePoint XML)Summary New Behavior

Invalid XML created by SharePoint web-serviceswasn't handled gracefully.

Invalid XML created by SharePoint web-services is handledmore gracefully


A high percentage of Bloomberg doc-uments have a large number of recip-ients and storing these recipients inthe corresponding fields took a sig-nificant portion of the crawl time.

A post processor can be activated that trig-gers for documents with recipient countsabove a configurable threshold (default:50).For these documents themeta entries 'rm_to', 'meta_recipientname', and 'meta_toname' are reduced to a single value'LARGE_RECIPIENT_COUNT'.



15.28 CORE-16419 (Write lock removal logging)Summary New Behavior

No logmessage was created when a usermanually removed a write-lock on anengine.

A logmessage is created when a usermanually removes a write-lock on anengine.

15.29 CORE-16447 (Stack trace dumping)Summary New Behavior

Mini dump creation could fail due to anaming conflict.

The process ID is now used tomake themini dump file names unique.

15.30 CORE-16492 (Robustness and stability)Summary New Behavior

An out-of-memory of physicalRAM caused a server down-time.

Memory consumption is lower than before afterdeactivation of expensive and unnecessary prefixsearch indices. Furthermore, thememory con-sumption of searches in an engine with many ongo-ing changes has been reduced significantly. Therisk of system downtime is reduced due toincreased stability of the exception handling.



15.31 CORE-16577 (Tagging processing)Summary New Behavior

The coding queue growth was caused by slowprocessing of change queue entries. This wascaused by lock contention on the index enginedue to amixture of *many* read requests and*long* read requests. Furthermore, the dur-ation of the periodically executed asyn-chronous Save - where no coding can beprocessed - was quite long. Additionally weobserved reviewer impact of the long queuewhich wasmainly caused by conversion tag-ging.

The processing speed of the save issignificantly improved, minimizingthe pause in potential processing oftagging (coding). In order to reducethe impact of the coding queue onthe reviewers several changes arenow in effect:

l The data structures for access-ing pending coding informationare reconfigured to allow forfaster access.

l The conversion tagging are pro-cessed without blocking writelocks.

l The locking of the last result vec-tors is optimized.

l The usage of in-memory struc-tures is broadened.


When base64 junkwas distributed overmultiple <text><p> tag sequences, theterm based junk detection did not engage.

XML tags between potential junk termsare now skipped during junk detection sojunk is recognized across tags.

15.33 CORE-16614 (Asynchronous saves)Summary New Behavior

Asynchronous save job concurring with a cat-egorymodification operation could cause theengine to run into a deadlocked state whenthe save job was stopped (either via the jobstab or due to another thread removing oradding a category).

When an asynchronous save isstopped the internal save gate willnow always be properly opened forother threads again.



15.34 CORE-16621 (Conversion performance)Summary New Behavior

Streaming of conversion results from batchserver to Engine used an unexpectedly smallbuffer size, resulting in poor throughputrates.

Buffer size has been increased andthe streaming operation is about threetimes faster.

15.35 CORE-16641 (CertificateUtilities)Summary New Behavior

CertificateUtilities was not available inToolBox.

CertificateUtilities is available inToolBox.

15.36 CORE-16662 (Native conversion)Summary New Behavior

S3-based projects failed to convertMicrosoft Office documents of recentMicrosoft Office versions.

S3-based projects convert as expected;the required file extension is set cor-rectly.

15.37 Found-9087 (MSG/PST files with TNEF attach-ments)


SMIME encrypted/signedMSG files/PSTentries with TNEF attachments were notsupported.

SMIME encrypted/signedMSG files/PSTentries with TNEF attachments are sup-ported.

15.38 FOUND-9171 (Inconsistent configurations)Summary New Behavior

Due to a template relation con-dition, it was possible to createinconsistent configurations.

The template relation condition has been fixedand it is no longer possible to create incon-sistent configurations. Additionally a script isprovided tomanually fix broken projects byinheriting specified template values.



15.39 FOUND-9513 (Process control robustness)Summary New Behavior

A non-responding DNS resolver couldbreak the process status handling.

The process control ismore robustagainst a non-responding DNS resolver.

15.40 FOUND-9582 (Adobe Indesign Interchange doc-uments)


Adobe Indesign Interchange documentscaused a NullPointerException if Oracle Out-sideInMIME type detection was enabled.

Adobe Indesign Interchange doc-uments no longer cause a NullPoint-erException.

15.41 FOUND-9612 (Application import)Summary New Behavior

Imports failed if any of theMSV files wereempty.

Imports do not fail even if anMSV file isempty.

15.42 FOUND-9633 (Large cache)Summary New Behavior

Cache of presentation objects was growingover time.

Cache is now self-cleaning.

15.43 FOUND-9646 (Cleanup after crawl)Summary New Behavior

Data sources alwayswrote checkpointinformation which could lead to time con-suming cleanup at end of crawl.

Data sourceswrite checkpoint inform-ation only if started with enabled "Auto-matic resume" option.



15.44 FOUND-9670 (Templates)Summary New Behavior

Extended template was not used in datasource creation via web service.

Extended template is correctly used indata source creation via web service.

15.45 FOUND-9728 (Loading performance)Summary New Behavior

Timeouts occur due to unnecessarylong loading timeswhen clicking on doc-uments or saving tags.

Loading timeswhen clicking on documentsor saving tags are faster and do notincrease with the number of availablebatches.

15.46 FOUND-9740 (CSV Load Wizard)Summary New Behavior

ArrayIndexOutOfBoundsException waspossible in CSV LoadWizard.

Array indices are now handled correctlyinside CSV LoadWizard.

15.47 FOUND-9750 (Stored search)Summary New Behavior

Error message regarding a storedsearch without search parameters didnot reference the stored search.

Error message regarding a stored searchwithout search parameters now includesname of the stored search.

15.48 FOUND-9763 (Engine user session)Summary New Behavior

Engine user session was left untouchedduring long running count operation, caus-ing user to get logged out.

Engine user session is touched duringlong running count operation so userdoes not get logged out.




16.1 AXC-35556 (Search Query Editor)Summary New Behavior

When running a report for a complexquery in SQE, the browser couldtimeout.

When running a report for a complex queryin SQE, the timeout will only occur after 5minutes.

16.2 AXC-36156 (Login)Summary New Behavior

When paging to another page than thefirst one, the user was not able to loginto thematter.

When paging to another page than the firstone, the user will see all available casesand is able to login to eachmatter.

16.3 AXC-36183 (Storage size audit log)Summary New Behavior

For billing, in the audit.core_stor-age_sizes table, a storage handler isnot reported if there is no storage alloc-ated yet.

There is now an entry for every storagehandler in the audit.core_storage_sizes table, even if there is no storageallocated yet (with sizes 0 bytes).

16.4 AXC-36187 (Storage size collection)Summary New Behavior

For projects that have not been started orhave been removed from disk only, a mis-leading error was reported when col-lecting their storage sizes (E1).

For projects that have not been started orhave been removed from disk only, awarning is reported together with ames-sage containing possible causes.





In Internet Explorer 11, when pressing the printbutton of the Axcelerate 5 viewer panel, a printjob was run, but the print result could not beshown.Instead, the user had to cancel the printjobmessage displayed.

Pressing the Print button in theAxcelerate 5 viewer results in cor-rect print output.



18 Contact Us

About RecommindRecommind provides themost accurate and automated enterprise search, automaticclassification, and eDiscovery software available, giving organizations and their usersthe information they need when they need it.

Visit us at http://www.recommind.com.

SupportFor support issues on Recommind products, visit the Recommind Ticketing System athttps://rts.recommind.com.

DocumentationFind Recommind product documentation, Knowledge Base articles, andmore inform-ation at the Recommind Customer Portal at https://supportkb.recommind.com. Forlogin access to the site, contact your product support:

l For : [email protected] For : [email protected]

TheRecommind Documentation team is interested in your feedback.

For comments or questions about Recommind product documentation, contact us [email protected].


18 Contact Us

http://www.recommind.com/

https://rts.recommind.com/

https://supportkb.recommind.com/

mailto:[email protected]?subject=Link to Recommind Customer Portal?subject=Link to Recommind Customer Portal

mailto:[email protected]

mailto:[email protected]?subject=Documentation Request?subject=Documentation Request

19 Terms of Use

DisclaimerThis document, as well as the products and services described in it, is furnished underlicense andmay only be used or copied in accordance with the terms of the license. Theinformation in this document is furnished for informational use only, is subject to changewithout notice, and should not be construed as a commitment byRecommind, Inc.,including its affiliates and subsidiaries (collectively, "Recommind"). Recommindassumes no responsibility or liability for any errors or inaccuracies that may appear inthis document or any software or services that may be provided in association with thisdocument.

Except as permitted by such license, no part of this document may be reproduced,stored in a retrieval system, or transmitted in any form or by anymeanswithout theexpresswritten consent of Recommind. Information in this document is provided in con-nection with Recommind's products and services. No license, express or implied, byestoppel or otherwise, to any intellectual property rights is granted by this document.

EXCEPT AS PROVIDED IN RECOMMIND’S SOFTWARE LICENSE AGREEMENTOR SERVICES AGREEMENT FOR SUCH PRODUCTS OR SERVICES,RECOMMIND ASSUMES NOLIABILITYWHATSOEVER, AND RECOMMINDDISCLAIMS ANY EXPRESS OR IMPLIEDWARRANTY, RELATINGTOSALEAND/OR USE OFRECOMMIND PRODUCTS OR SERVICES INCLUDINGLIABILITY ORWARRANTIES RELATINGTOFITNESS FOR A PARTICULARPURPOSE, MERCHANTABILITY, OR INFRINGEMENTOF ANY PATENT,COPYRIGHTOR OTHER INTELLECTUAL PROPERTY RIGHT. RECOMMINDMAKES NOWARRANTIES REGARDINGTHE COMPLETENESS OR ACCURACYOF ANY INFORMATION, NOR THAT THE PRODUCTS OR SERVICESWILL BEERROR FREE, UNINTERRUPTED, OR SECURE. IN NOEVENTWILLRECOMMIND, THEIR DIRECTORS, EMPLOYEES, SHAREHOLDERS ANDLICENSORS, BE LIABLE FOR ANY CONSEQUENTIAL, INCIDENTAL, INDIRECT,SPECIALOR EXEMPLARY DAMAGES INCLUDING, BUT NOT LIMITED TO, LOSSOF ANTICIPATED PROFITS OR BENEFITS.

Recommindmaymake changes to specifications, and product and service descriptionsat any time, without prior notice. Recommind's productsmay contain design defects orerrors known as errata that may cause the product or service to deviate from publishedspecifications. Current characterized errata are available on request. Whilst every efforthas beenmade to ensure that the information and content within this document is accur-ate, up-to-date and reliable, Recommind cannot be held responsible for inaccuracies orerrors. Recommind software, services and documentation have been developed andprepared with the appropriate degree of skill, expertise and care. While every effort hasbeenmade to ensure that this documentation contains themost up-to-date and accur-ate information available, Recommind accepts no responsibility for any damage that


19 Terms of Use

may be claimed by any user whatsoever for the specifications, errors or omissions in theuse of the products, services and documentation.

Trademarks and PatentsRecommind's underlying technology is patented underU.S. Patent Nos. 6,687,696,7,328,216, 7,657,522, 7,747,631, 7,933,859, 8,024,333, 8,103,678, 8,429,159 and8,489,538

Recommind, Inc. is the leader in predictive informationmanagement and analysis soft-ware, delivering business applications that transform the way enterprises, governmententities and law firms conduct eDiscovery, enterprise search, and information gov-ernance. Recommind, Axcelerate, Axcelerate Cloud, Axcelerate OnDemand, andCORE’s name and logo are registered trademarks of Recommind, Inc.

CopyrightCopyright © Recommind, Inc. 2000-2017.


19 Terms of Use