[cups] broken pipe but job successfull

[Helge Blischke]

> Am 04.09.2020 um 09:26 schrieb Lorenzo Milesi (maxxer) <lorenzo.milesi at gmail.com>:
> 
> I was wondering if the ErrorPolicy printer attribute could make any difference. I.e. setting it to retry-this-job could make it retry in case of errors?
> 
> I don't understand the difference between retry-job and retry-this-job. The latter explicitly states about this job, but the first one seems to imply it.
> 

No, it is definitely not a matter of the error policy attribute.

Digging one more into your error_log attachment, I realised that the pdftops filter 
used the (popper based) pdftops utility.

I tried to emulate your filter chain by calling the pdftopdf and pdftops filter separately 
and could not find any quirks in the resulting PostScript file that could have caused a failure.

You could try the command line option 
pdftops-renderer=xxxx
where xxxx is one of
gs				(Ghostscript)
pdftops			(popper’s pdftops utility, which is default for KONICA-MINOLTA printers)
pdftocairo		(poplar with cairo)
acroread			(probably no longer installed)
mupdf			(probably not installed)

Perhaps one of these settings might cause the print job to succeed.

Anyway, what the error_log tells me looks like a crash (and automatic restart) of the
printer’s PostScript interpreter.

For a final test, you could send me the PPD of the printer in question (off the list, as the list
deletes non-text attachments).

Helge