IBR on separate server - Custom Conversions

Question

I have a custom conversion process that interupts the IBR process to do some work on a document before the rendering step. It works well while IBR is installed on the same server as the Content Server, but I am struggling to get it going when I move IBR onto a separate server.
When run on the same server, the "DocConverter.hda" contains the location of the original location of the document in the vault, but when on a separate server, everything points to a local folder. It appears that a pre-process copies in a copy of the document into a temp directory, and after the completion of the rendering, a post-process copies the pdf out to the weblayout folder. My process needs to know the location of the original document in the vault, but I cannot find any references to the pre/post processes in the documentation or even the logs.
Anyone know how this works?
I am running CS 7.5.1 and PDFConverter 7.6 on a Windows Server 2003 platform with a Sql Server 2005 back end.
Cheers,
Desmo

Answer 1

David is correct. Depending on exactly what you are changing (always should be in a component), you need to restart the Content Server first and then restart the refinery because of the way the conversion information gets passed to the refinery from the Content Server. If you are changing the top level file formats entry, you may need to check in a new version of the file vs just resubmitting the file to the refinery using the Repository Manger applet to make sure the right conversion gets called.
If you don't want to create this yourself, Fishbowl Solutions has a product called Native File Updater (part of the Document Automation Suite) that will likely work for you.
http://www.fishbowlsolutions.com/StellentSolutions/StellentComponents/fs_docauto_webcopy?WT.mc_id=fb_sig_to
Hopefully that helps,
Tim