Thank you very much for the details!
There are actually two issues here. The biggest problem is, that the PDF files does not have any text in them, they are only images scanned to a PDF document, so there is nothing to extract. Another issues is (it is related to the first one), because of the images the files are extremely large, so even if there was a text, it would be very hard to extract.
If the PDF files are converted to proper text documents with reduced sizes, then the PDF indexing will start working.