Manual export to Excel from Qlikview works but fails via NPrinting

[timsaddler]

hello all

I can export a chart to Excel from Qlikview but when i try to do this via NPrinting it fails.

The chart is large (19meg) and 212,000 rows.

The dev server i used to export manually has 32 gig ram

The NPrinting server has 16 gig ram

The error message for the failure isn't very helpful (see below)

Failed generation of report cde1fb0a-9705-42c5-9e6c-022b299029e0 for user 9d24661e-35a2-4295-9c90-f340c7b0f995. ERROR: NPrinting Engine failed to open document

Is there a way of "proving" that memory is an issue or is there something else i need to investigate?

Thanks in advance

[Ruggero_Piccoli]

Hi,

The Windows user who is running the Qlik NPrinting Engine service has not the rights to open the document.

Check that QlikView Desktop has a valid license activated for that user (Personal Edition is not supported). Left click on Qv.ex while keeping SHIFT+CTRL pressed and select Run as a different user. Insert the credentials of the Windows user who is running the Qlik NPrinting Engine and be sure the license is active. Leased licenses from a QlikView server are not supported.

Be also sure that the Windows user who is running the Qlik NPrinting Engine has access to the folder where the .qvw is if you are opening it as local or has the rights to open it from the QlikView Server.

Best Regards,

Ruggero

[timsaddler]

Hello @Ruggero_Piccoli 

Thanks for your prompt response.

I don't think it is an issue with permissions because :-

1. I have just run the report again with a "filter" to reduce the data to 5 meg and this worked as expected.

2. I have other reports generating daily from the same Qlikview App

Do you have any further suggestions?

many thanks

[Ruggero_Piccoli]

Hi,

In a xls workbook, the row limit is 65,536 and 256 columns. With xlsx workbooks (and xlsm), limits are 1,048,576 rows and and 16,384 columns. You are exporting more than 200.000 rows, so I suggest you to check that you selected the xlsx output format in the task.

If this also doesn't work, please open a support ticket with a link to this conversation. It is very important that in the ticket you will include all the information that let us to reproduce the issue. So, for example, the original .qvw also with scrambled data, the template, the versions of all softwares you are using, etc.

Best Regards,

Ruggero

[JonnyPoole]

Monitor task manager on the NP server as you perform the operation to see whether its a resource issue.

Another idea to 'parallelize' the extract : 

- establish a key for this QV table object where the key value changes for each record of the table 

- create 5-10 new table objects each with the key and just a subset of the columns

- in NPrinting designer, include fields from the multiple table objects side by side.  It will enable Nprinting to parallelize the data extraction in smaller chunks.

Try it with two objects first with a few columns each and then expand to more objects/columns.