(I'm not even sure how it got in to this state...but anyway)
If the combined.pdf file used for the annotated PDF does not exist on the disk, but it does have a database record, then it appears that there is no checking of the existence of the content on disk.
The editpdf ajax.php file is attempting to get the number of pages in the combined.pdf, which makes a copy of the combined.pdf file.
It then loads the file copied file to count the number of pages.
At no point does it appear to check that the file can be or has been successfully copied.