4 Replies Latest reply: Apr 14, 2010 6:11 AM by Christian Schlettig RSS

    Export memory exceeded

    Christian Schlettig

      I use a qlikview macro to export data into a csv file from a simple table on a QV 9 SR3 7440.8 Server including QV Desktop.

      I have a row count from 50000 up to 1.4 mill. data rows.

       

      i use the .export function to write data to csv file.

      In many cases i got only 1 row with the text "Export memory exceeded" in it.

      My Actions were:

      I already set the export value in QV Developer to 2020.0 MB.

      The Memory Export and timeout Limits in QV Server were quadrupeld.

       

      i still get the Export memory exceeded message.

        • Export memory exceeded
          Oleg Troyansky

          I'm not sure how to solve he memory problem, but I wanted to mention that you can now store CSV files the same way you store QVD in your load script. If all you need is to export your raw data, this method would be much more robust and scalable than exporting from a table using a Macro.

            • Export memory exceeded
              Christian Schlettig

              Thanks, Oleg,

              but i try to export the results of bookmarks applied to the application. In fact i cycle through all bookmarks in the application and export the results to a csv to be processed further in another database.

               

              the history:

              Before exporting to csv i looped manually with vbscript macro through each row of the table and inserted via ODBC in the target database.

              That worked in most cases but it was too slow (took around 2 hours per 1 million rows) and sometimes the odbc connection hang and qv simple didn't write anything in the database anymore. I get the error message on the Console Desktop : "Execution of Macro failed"

               

              Would it be possible to extract data the other way round. From another application with qv-odbc driver to the qv application. Can i "apply Bookmarks" before i read from ODBC?

               

              Kind regards,

              Christian

               

                • Export memory exceeded
                  Oleg Troyansky

                  Christian,

                  I understand... If you have to cycle through bookmarks, then your options are limited...

                  - I'm not sure if you can apply bookmarks before you read from ODBC... I haven't used QV-ODBC in a while, but I remember that earlier versions of it were not too robust.

                  - If the number of your bookmarks is limited and known upfront, try using Actions instead of Macros (if you are on ver. 9.00)

                  - Try to step back and find an alternative solution all together...

                  As an idea to consider, here is what you can try:

                  Write a macro that does the following:

                  1. Cycles through bookmarks. For each bookmark:

                  2. Save the document with an alternative name

                  3. Apply bookmark and Reduce data to possible values

                  4. initiate partial reload

                  5. In your load script, add a section conditioned by function IsPartialReload(), that stores the necessary tables into CSV files

                  6. Discard the temporary document and go back to the original

                   

                  good luck!

                    • Export memory exceeded
                      Christian Schlettig

                      Hi Oleg,

                      thanks for your great input.

                      i solved the problem slighlty modified using the following steps:

                      1. exported to excel instead of csv in the Macro. (This has no Export Memory Problems)

                      this however gives the next task, since QlikView writes only 65536 rows on one sheet and so puts multiple sheets in the excel output.

                      2. Put the number of Rows in the Excel Filename like "Filename_1234.xls"

                      3. Open the excel file with another QlikView File via a looped load that runs through all sheets.

                      4. store the generated table into csv file

                      process the csv file further and store into DB.