Missing the FO Tree with FOP 0.20.1 and BEA Weblogic

[j-...@chronopost.fr]

Hi everyone!

I have a servlet which generates documents PDF from XML/XSL.,Elle goes well with
 FOP V0.20.1 under the application server BEA Weblogic V5.1 whose log is as
follows:

using SAX parser org.apache.xerces.parsers.SAXParser
building formatting object tree
setting up fonts
 [1] [2] [3] [4] [5] [6] [7] [8] [9>] [10>] [11>] [12>] [13>] [14>] [15>]
[16>>>>>] [17>>] [18>>>>>>] [19>] [20>] [21>] [22>] [23>] [24>] [25>]Parsing of
document complete, stopping renderer
Initial heap size: 6709Kb
Current heap size: 11703Kb
Total memory used: 4993Kb
  Memory use is indicative; no GC was performed
  These figures should not be used comparatively
Total time used: 15092ms
Pages rendererd: 25
Avg render time: 603ms/page
toPDF:  Durée(mm:ss,SSS) 00:58,244


Unfortunately, under the waiter of application BEA Weblogic V6.1, my servlet
creates a PDF empty whose size is 0 KB, which Acrobat Reader cannot it open
:"une error occurred at the time of the opening of this document.  Misreading of
 the file".
By looking at the log:

using SAX parser org.apache.xerces.parsers.SAXParser
toPDF:  Durée(mm:ss,SSS) 00:15,352 (filePDF.close();)


I have the impression that FO Tree  is not created. I do not understand at all.
  Why such suprise with the same/identical  servlet/XML/XSL-FO and FOP?

Do you have an idea ?
I thank you in advance.

______________________________________________
Jianxin KONG   ( j-kong@chronopost.fr )
CHRONOPOST INTERNATIONAL
Dir. systèmes d'information