It seems that this is NOT jD problem.
Filesonic probably has serious problem with their indexes.
I've got links that in jD:
a) in jD - Link Grabber says they are "All online"
b) jD can't download them because "Failed cloning connection"
The same links in browser:
Browser start to download them, but it downloads 0 bytes (so it is the reason for "failed cloning connection" because jD can't download 0 bytes in N chunks)
Their (Filesonic) linkchecker says that they are "Not Available" (deleted or removed)
So if they are unavailable why browser tries to download them and it is not redirected to "file not found" info.
So my conclusion is they have serious problems with their internal file structure - the links are treated differently depends on where you check them (browser, FS linkchecker, jD Link Grabber, jD downloader) - the answer should be always the same.
Example link:
- no matter if Premium or not Premium - browser tries to download it, and the result is NULL file.