EducationSoftwareStrategy.com
StrategyCommunity

Knowledge Base

Product

Community

Knowledge Base

TopicsBrowse ArticlesDeveloper Zone

Product

Download SoftwareProduct DocumentationSecurity Hub

Education

Tutorial VideosSolution GalleryEducation courses

Community

GuidelinesGrandmastersEvents
x_social-icon_white.svglinkedin_social-icon_white.svg
Strategy logoCommunity

© Strategy Inc. All Rights Reserved.

LegalTerms of UsePrivacy Policy
  1. Home
  2. Topics
  3. KB485919: How to enable the fix for large cube cache synchronization defects

How to enable the fix for large cube cache synchronization defects


Jine Wang

Quality Engineer, Principal • MicroStrategy


This article outlines how to enable the fix for large cube cache synchronization defects.

Starting with the release of Strategy ONE (March 2024), dossiers are also known as dashboards.

Background


In cluster environment, customer might run into the below performance and functionality problems if big cube(s) is republished on one node and meanwhile be-using on remote nodes:

Symptom 1


In cluster nodes if user republish a big cube and the cube cache would be synchronized to remote node. Considering the cube cache file's large size, it would cost sometime for the cube cache to be loaded/ready on the remote node. If during this time window (the cube cache loading is going on), users in the remote node try to run objects such as subset report or dossier against the specific cube cache, they would experience performance issue.

Symptom 2


In cluster nodes if user republish a big cube while it's in-use/occupied by remote node(s). After the republish is complete and synchronization for remote node(s) is also finished, users would experience inconsistent data issue if they execute objects against the specific cube(s): the user on the node where the republish happen would notice the latest data. However, the users on the remote node(s) would still see the old data. And if the node while the publish happen is restarted, the latest data from the cube publish would be lost. 

Action


Starting from Strategy 2020, the fix is provided for above problems by adding 3 registry entries.  Below are instructions to apply the fix.  

Windows platform

  1. Launch the 'Registry Editor'.
  2. Add a DWORD type register entry "HugeCubeSynch" with value 1  under [HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\MicroStrategy\Data Sources\CastorServer]. Sample is below: 
ka0PW0000002UMzYAM_0EM4W000006zJhG.jpeg
  • You can specify the project to apply the fix using the following methods:
    • Add a DWORD type register entry "EnableCubeFileShareDelete" with value 1 also under [HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\MicroStrategy\Data Sources\CastorServer].  Besides, add another entry "ProjectsToUseFileShareDeleteMode" to specify which projects would apply the fix. The value is the GUID(s) of project(s). You can list multiple projects GUID, each is separated by semicolon. Sample is below: 
ka0PW0000002UMzYAM_0EM4W000006zJhQ.jpeg
    • Starting in Strategy ONE (June 2024), add a DWORD register entry "EnableCubeFileShareDelete" with value 2. This entry will apply the fix to all the projects, including newly added projects. You do not need to specify the project GUID. 
ka0PW0000002UMzYAM_0EMPW000007jZbt.jpeg
  • Do the above manipulation for each cluster node and restart each cluster node to pick up the modification. 

Linux platform

  1. Locate the MSIReg.reg file and edit it.
  2. Add a DWORD type register entry "HugeCubeSynch"=dword:00000001  under [HKEY_LOCAL_MACHINE\SOFTWARE\MicroStrategy\Data Sources\CastorServer]. Sample is below:
                    [HKEY_LOCAL_MACHINE\SOFTWARE\MicroStrategy\Data Sources\CastorServer]  

                       "HugeCubeSynch"=dword:00000001


 

  1. You can specify the project to apply the fix using the following methods:
    • Add a DWORD type register entry "EnableCubeFileShareDelete"=dword:00000001 also under [HKEY_LOCAL_MACHINE\SOFTWARE\MicroStrategy\Data Sources\CastorServer].  Besides, add another entry "ProjectsToUseFileShareDeleteMode" to specify which projects would apply the fix. The value is the GUID(s) of project(s). You can list multiple projects GUID, each is separated by semicolon. Sample is below:
                      [HKEY_LOCAL_MACHINE\SOFTWARE\MicroStrategy\Data Sources\CastorServer]

                         "EnableCubeFileShareDelete"=dword:00000001
                         "ProjectsToUseFileShareDeleteMode"="GUID1;GUID2;GUID3;GUID4"
    • Starting in Strategy ONE (June 2024), add a DWORD type register entry "EnableCubeFileShareDelete"=dword:00000002 also under [HKEY_LOCAL_MACHINE\SOFTWARE\MicroStrategy\Data Sources\CastorServer]. This applies the fix to all projects, including newly added projects. You do not need to specify the project GUID. Sample is below:
                      [HKEY_LOCAL_MACHINE\SOFTWARE\MicroStrategy\Data Sources\CastorServer]

                         "EnableCubeFileShareDelete"=dword:00000002
                         
  2. Do the above manipulation for each cluster node and restart each cluster node to pick up the modification. 


NOTE: The registry change need to be applied on all cluster nodes and each IServer need to be restarted after modifying the registry file.
 


Comment

0 comments

Details

Knowledge Article

Published:

February 8, 2023

Last Updated:

June 21, 2024