Skip to main content
Announcements
Qlik Connect 2024! Seize endless possibilities! LEARN MORE

_deletedentitylogs take up disk space on C:\

No ratings
cancel
Showing results for 
Search instead for 
Did you mean: 
Filippo_Nicolussi_P

_deletedentitylogs take up disk space on C:\

Last Update:

Mar 4, 2022 2:46:26 PM

Updated By:

NadiaB

Created date:

Jul 2, 2018 8:59:47 AM

Qlik Sense Enterprise soft deletes objects and stores the entities in the _deletedentytilog table in the QRS database. 

A heavily used system with many tasks and frequent development activity, this table can consume a lot of disk space and could impact performance.

 

Environment:

Qlik Sense Enterprise on Windows 

 

Resolution

Note: Please perform a Repository database backup prior any changes.

As of Qlik Sense February 2019, an automatic clean up process cleans this table, the frequency of this process can be adjusted in the repository.exe.config file.

The following configuration options were added: 

  • DeletedEntityLogMaintenance.Enabled - Defines if maintenance work should be enabled or not. Default true.
  • DeletedEntityLogMaintenance.Interval - Defines how often the maintenance work should run. Default 24 hours.

The following configuration options have been changed from previous versions: 

  • DeletedEntityLogMaintenance.StartAt - Is no longer specified in local time. Instead the time is relative to UTC.
  • DeletedEntityLogMaintenance.CleaningThreshold - Renamed from CleaningThresholdInDays and is no longer an integer value. Instead the value is entered as a TimeSpan in format days.hh:mm:ss. Default 90.00:00:00. (Default value initially was 90 days, later versions was changed to 30, this value can be configured if its necessary to cleanup in shorter period of time).

But if space needs to be reclaimed faster, a workaround can be deployed: 

If the table that is wasting space on C:\ drive is "_deletedentitylog"   it is safe to empty the table manually: 

  1. Schedule a minimum of 30 min of down time and stop all Qlik Sense Services except for the Qlik Sense Repository Database service.
  2. Connect to Postgres DB with PGAdmin Installing and Configuring pgAdmin 4 for Qlik Sense June 2017 Release and up
  3. Select the QSR Database 
  4. Open Query tool 
  5. Issue the query 
    truncate "_deletedentitylog";​
  6. Restart Qlik Sense Services starting from Qlik Sense Repository Service and Qlik Sense Dispatcher

When truncating the table a new empty file will be created for that table.   

In May 2021 a new parameter was introduced that allows to exclude logging entries that are highly contributing to the size of the table and are NOT valuable in terms of potential backtracking.

<add key="DeletedEntityLog.ExcludedEntities" value="TempContent;ExecutionResult;ExecutionResult.Detail;ExecutionSession" />

 

Labels (1)
Comments
Filippo_Nicolussi_P

At today on Qlik Sense February 2023 here below list of options used to keep under control size of the table mentioned: 

<add key="DeletedEntityLog.ExcludedEntities" value="TempContent;ExecutionResult;ExecutionResult.Detail;ExecutionSession" />
<add key="DeletedEntityLogMaintenance.BatchSize" value="5000" />
<!-- Minimum value is 1 -->
<add key="DeletedEntityLogMaintenanceByDate.CleaningThreshold" value="30.00:00:00" />
<!-- Default value is 30 days -->
<add key="DeletedEntityLogMaintenanceByDate.Enabled" value="true" />
<add key="DeletedEntityLogMaintenanceByDate.Interval" value="1.00:00:00" />
<!-- Minimum value is 1 minute -->
<add key="DeletedEntityLogMaintenanceByDate.StartAt" value="02:00:00" />
<!-- Relative to UTC time (hh:mm:ss), minimum value is 00:00:00, maximum value is 23:59:59 -->
<add key="DeletedEntityLogMaintenanceBySize.CleaningThreshold" value="1024" />
<!-- Default value is 1024 mega bytes -->
<add key="DeletedEntityLogMaintenanceBySize.Enabled" value="true" />
<add key="DeletedEntityLogMaintenanceBySize.Interval" value="1.00:00:00" />
<!-- Minimum value is 1 minute -->
<add key="DeletedEntityLogMaintenanceBySize.StartAt" value="04:00:00" />
<!-- Relative to UTC time (hh:mm:ss), minimum value is 00:00:00, maximum value is 23:59:59 -->

 

In multi-node scenario suggestion is to enable the maintenance only on two nodes (for example Central and one Failover Candidate) and to shift the StartAt  for the second node. 

 

Version history
Last update:
‎2022-03-04 02:46 PM
Updated by: