5 Replies Latest reply on Aug 4, 2016 5:58 AM by khalid norat

    incomplete data extract

    Jonathan Marquez

      I have a small data set (11K+ rows).  However, in creating a data extract, tableau only gets 3,912 rows even if I didn't specify any filter.  Any ideas why this happens?

        • 1. Re: incomplete data extract
          Russell Christopher

          What is your data source?

           

          Do you see 11K rows in Tableau before you create the extract, or do you only see 3,912 rows when making a live connection, as well?

           

          My first guess is that you're using a text file that is somehow malformed...

          • 2. Re: incomplete data extract
            Jonathan Marquez

            Hi Russell,

             

            I'm connected to a Microsoft SQL Server installed in my desktop.

            I see the 11K rows before I create an extract.  I just lose the other rows in the process of creating the extract.

             

            Funny thing is, one of my colleague told me to re connect to my data.  But instead of using live connection and performing the extract, I will immediately create an extract for the data.  By following his suggestion, I was able to get all of my data in data extract.

             

            I'm still puzzled by my original problem and it still caused inefficiencies.  Do you know what possibly caused this?

            • 3. Re: incomplete data extract
              Russell Christopher

              I know you said you didn't specify a filter, but do you think one snuck in there somehow?

               

              Also, what version of Tableau Desktop and SQL Server are you on? Do you know what version of the SQL Server drivers you have running?

               

              I'm not in support at Tableau, but I'd be happy to give this a test on my side over the week-end if you can get me a backup of the database in question or a little bit more information (versions, driver info)

              • 4. Re: incomplete data extract
                Jilske Hupkes

                I seem to have the same problem.

                I have a Tableau report that I have used for the past 2 years, and every quarter I update the data in the SQL server database and then I update the Extract in Tableau.

                This has working fine until now, but this time the extra data in the database (via a custom query) is only partially added to the extract.

                 

                After some testing I found a workaround; I now add an extra row to the dataset (via a UNION ALL in the customer SQL query) and then I exclude this row from the dataset using the extract filter.

                When I follow this proces I do get all the records from the database in my extract.

                 

                Before this quarter my query resulted in max 5800 rows, and this quarter it results in 6475 rows (incluiding the extra 1 row).

                This is not the only dataset in my Tableau report, but the datasets are not mixed, except that they share some parameters.

                I use Tableau version 9.3. and SQL server 2012 on a remote server.

                 

                I have tested, and after again removing the UNION ALL and the filter my extract is still complete.

                Please be assured that I did update the extract in the proper way: Data > select the proper datasource > Extract > Refresh

                 

                My theorie is that updating the Extract doesn't always has an effect, but if you change the connection (other query and/or change an extract filter, going to live back to an extract) then the extract is refreshed

                • 5. Re: incomplete data extract
                  khalid norat

                  I came across the same issue - The work around I used and that worked for me was to place a datasource filter to bring back the top N number of rows.

                   

                  Use a number that is large enough for your data to never reach. I used 10 million where I only had 1 million records this is to avoid needing to change it as the data grows.

                   

                  Not the best of solutions but its quick and it works.

                   

                  Hope this helps for now and tableau can investigate why this is happening