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

Qlik Replicate and Aurora Postgres: null values on varchar data type

100% helpful (1/1)
Showing results for 
Search instead for 
Did you mean: 

Qlik Replicate and Aurora Postgres: null values on varchar data type

Last Update:

Sep 27, 2023 4:27:35 AM

Updated By:


Created date:

Sep 27, 2023 1:20:54 AM

Due to an issue with MySQL ODBC Driver 8.0.027 to 8.0.033, empty TEXT columns may not be replicated correctly during Full Load.

Running the full load on a table, it appears the data alignment is off because the target .csv file has null values when the source field is defined as not nullable. 



Qlik Replicate  
Aurora Postgres


Identifying the root cause:

Verify the source table DDL:

Examining the source table DDL will show a datatype with no precision (varchar NOT NULL,). 


CREATE TABLE tran.category  (   
  id     integer NOT NULL,
  name        varchar NOT NULL,
  description varchar NOT NULL,
  category         varchar NOT NULL,
  is_active            boolean NOT NULL DEFAULT true,
  create_employee      integer NOT NULL,
  create_datetime      timestamp NOT NULL,



This is a documented limitation. See VARCHAR(n).

VARCHAR without a length (n) is not recognized as a valid data type by target endpoints. Consequently, if a source column data type is set to VARCHAR without an explicit length, Replicate will set a default length of 8000 bytes.


You can change the default by deploying an internal parameter, unboundedVarcharMaxSize, and setting a custom value. 

  1. Go to the  Endpoint connection
  2. Switch to the Advanced tab
  3. Click Internal Parameters
  4. Add unboundedVarcharMaxSize

    unboundedvarcharmaxsize internal parameter.png

    For more information about Internal Parameters, see Qlik Replicate: How to set Internal Parameters and what are they for?


Labels (1)
Version history
Last update:
‎2023-09-27 07:27 AM
Updated by: