Hi,
I am creating a prototype with LC Output. The calling application will produce big XML file - Batch with more than1000 records.
Example XML data:
<?xml version="1.0" encoding="utf-8"?>
<BATCH>
<formDataRecords>
<LC_KOMA002>
<customerNameAddress>Jensens Biludlejning</customerNameAddress>
<customerNameAddress>Vestergade 21c</customerNameAddress>
<customerNameAddress>7100 Vejle</customerNameAddress>
</LC_KOMA002>
<LC_KOMA002>
<customerNameAddress>Pete Petersen</customerNameAddress>
<customerNameAddress>Vestergade 10</customerNameAddress>
<customerNameAddress>7100 Copenhagen</customerNameAddress>
</LC_KOMA002>
...
</formDataRecords>
...
In a setVariables step I extract the XML data to be merged with formTemplate. I extract all all nodes under <formDataRecords> in a new XML variable.
I use this xml as input data to a GeneratePDFOutput (LC Output). The GeneratePDFOutput is set up with 'multiple streams' and 'record level' 2 and 'Record name' = <LC_KOMA002>. I use the 'Output Location URI' to save the pdf with incremental filename xxx1,2,3,4.pdf
The workflow works fine with 500 records - here it produces 500 pdf files with 1 page.
Problem:
When running with 1000 records or more my process does not produce any pdf. The watchedfolder take my XML datafile. But after 3-4 min. it returns this error in the ERROR server log.
2010-08-12 11:52:00,484 WARN [com.arjuna.ats.arjuna.logging.arjLoggerI18N] [com.arjuna.ats.arjuna.coordinator.BasicAction_58] - Abort of action id a10ed08:ce4d:4c596be4:15f4fee invoked while multiple threads active within it.
It seems like GeneratePDFOutput run out of memory in some way when it handles more that 500 records.
I hope that you have ideas to configuration/settings that can be used on the Adobe LC server or CL Output so that it can handle large batch jobs??
I look forward to all you clever solutions - I really want to show the customer Adobe LC can do this one :)
/Thomas Groenbaek, Jyske Bank