Hi, actually we are encountering a strange error while exporting a report in CSV. As the data amount is big (round 170.000 records), we decided to virtualize the report (otherwise goes in OutOfMemory). In the file attached (report.csv) is enclosed a small subset of the entire recordset where is verified the error. The two lines who behave incorrectly are the following:1) 11;1;8;O;4;85416101;MOVIMENTAZIONE P.F. ;2702715;MAGA;MARIO;13/09/2010;L;8;0;19,02;13,53;22;;;;;;;;;1902;19/10/2010 17.232) 11;1;8;O;4;85416101;D;2702715;MAGA;MARIO;13/09/2010;L;8;0;19,02;13,53;22;;;;;;;;;1902;19/10/2010 17.23 Where the column “Descr. C.d.C” should have the entire value “MOVIMENTAZIONE P.F. D”, instead, the output file has the content of this record separated in two records. The first containing “MOVIMENTAZIONE P.F. ”, and the second “D”. We checked the data in the DB and the correct value is “MOVIMENTAZIONE P.F. D”, as is correctly exported in other previous rows of the dataset (see report.cvs).This “anomaly” is noticed first at row 47, then 9047, then 9304…with no periodicity.The record doesn’t contain special characters and isn’t larger than the rest of the records. A curiosity: Launching the report with smaller dataset (30.000 records) without the use of virtualization, the error isn’t present.So, I think is due to Virtualization.In this way, I used two different methods of virtualization:1) JRAbstractLRUVirtualizer vir = null;vir = new JRFileVirtualizer (50, sPath); if (vir != null){ parameters.put(JRParameter.REPORT_VIRTUALIZER, vir);} 2) JRAbstractLRUVirtualizer vir = null; JRSwapFile swapFile = new JRSwapFile(sPath, 4096 , 100);vir = new JRSwapFileVirtualizer(100, swapFile, true);if (vir != null){ parameters.put(JRParameter.REPORT_VIRTUALIZER, vir);} Any idea about what’s happening o where am I wrong? Could it be, because I’m missing any parameter property? (a part from the report virtualizer parameter)Any hint, would be helpful because it’s been 5 days that I’m working on it… Thanks in advance,