Freaky occurrence involving a print file overridden to a PDF

Saw this at a customer site, and I'm not entirely sure what to make of it, or what to do about it.

We have an externally-described print file, which I'll call FOOBAR. And we have a CL program that does an OVRPRTF to create a PDF that I'll call /frobozz/pdfwork/FOOBAR12345.PDF (after having deleted any previous occurrence of that stream file). Then, it calls an RPG program (I'll call it FROBNITZ), that prints to that print file, generating the PDF.

Except that in a single freak occurrence, when the RPG program was called, it threw a CPF4208 message:

To module . . . . . . . . . : FROBNITZ
To procedure . . . . . . . : FROBNITZ
Statement . . . . . . . . . : 1000001
Message . . . . : File FOOBAR in FROBOZZ not opened.
Cause . . . . . : The open operation of file FOOBAR in library FROBOZZ
failed because of reason code 6: . . . 6 --
Specified stream file already exists. . . .
Technical description . . . . . . . . : errno = 3457.

and locked up the job.

But there isn't a "statement 1000001" to be found anywhere in the source member or compilation listing.

I will note that FOOBAR is *not* USROPN, and its F-spec declares it as an externally-described output file, of type PRINTER.

In the CL program, it's just
. . .
> DEL OBJLNK(&PDFPATH)
> MONMSG CPF0000
> OVRPRTF . . .
> MONMSG CPF0000
> CALL FROBNITZ . . .
. . .

Near as I can tell, two jobs were trying to generate the same PDF at the same time. Anybody have any insights?

--
JHHL