Data load error using Qlik Replicate - NULL result in a non-nullable column

[satyender]

Greetings!

I am encountering an error while loading the data from DB2. Any code/expression to fix this?

Thanks!

Error:

NULL result in a non-nullable column

[Barb_Fill21]

Hello @satyender 

Not sure if this will help. 

. Open the table from the bottom right of the Designer screen. Select Transform. On the screen that opens, click the Expression field for the column. Enter: CASE $ColumnName WHEN null Then ' ' Else $ColumnName END Actually this might work just as well: ifnull($ColumnName,' ')

Please try and let me know.

Thanks,

Barb

[Heinvandenheuvel]

Well,  it reports 'null result', and source and target are numeric (some encoding for a date like YYMMDD perhaps).

Is it possibly that some IBM 'numerics' are not recognizable as numeric by other like 'all spaces' an 'overpunch' or trailing vs leading sign? in that case the ibm source would not have a null as such, but something which cannot map to a number on an other platform.  Can you 'keep' or copy the CSV and study that instead of the source? It my help you identify the source row triggering the error.

File '452' suggests many have rows been successfully loaded, is that true? 

The same error appears to repeat itself. You did not include the timestamp(s) are they rapid-fire or multiple seconds apart suggesting a forever failing retry?

Low 1 could be a  sloppy error message, or it is the first row in the first file and nothing was ever loaded?

Maybe a transformation such as @Barb_Fill21 suggests can help but considering it is a numeric column instead  of '' I'd suggest a recognizable number like 987654, 123456, 424242 - if you catch my drift. Then after the load (if it now completes) search for those  rows on target and then the corresponding rows on source.

hth,

Hein.

[saikrishna8]

Hi @satyender  We are having similar issue for oracle as source, what was the outcome from the tread you discussed with QLIK PS. Any inputs would help us resolve the issue.

Thank you,

SK Pyata

[john_wang]

Hello @saikrishna8 ,

Many reasons may lead to the same error in Snowflake during Full Load stage. In most cases please try to add an internal parameter ,emptyFieldAsNull , and set it to TRUE, then Reload the task(or table) to see if it works for you.

Hope this helps.

Regards,

John.

[Barb_Fill21]

Hi @satyender 

As this is a DB2 iSeries source, please ask the DB2 DBA to make sure you have *BOTH set for the journaling of the tables.   And depending on the target and version of Replicate it may be an issue that is already fixed.

Sincerely

Barb

[harsh2]

hey hi  @satyender 

I am having the same issue, so could you please tell me what conclusion you get in the end? It would be helpful for me.

[suvbin]

@satyender

Am also facing similar issue. Source is  SAP Hana , Target is Snowflake. How to know exactly in which column the error would be?

[harsh2]

Hi suvbin,

refer to following community's link this might help 

link : https://community.qlik.com/t5/Qlik-Replicate/Data-type-issue-in-Qlik-replicateTask/m-p/2142942#M8304

Regards,

Harsh

[suvbin]

@john_wang Am also facing the same  issue, could you please provide the script for the same . 

Thanks

[john_wang]

Hello @suvbin ,

Are you able to share the source and target DBs types? it's better with a table sample (obfuscating sensitive information as needed).

thanks,

John.