lost tray selection ability

Helge Blischke h.blischke at acm.org
Sat Jun 4 05:41:33 PDT 2011


Jeff wrote:

> Thanks Paul
> 
> I'll see if I can force the CUPS on the working box to upgrade without
> upgrading Ghostscript. That would give me CUPS 1.4.6 with GS 8.70.
> 
> Jeff
> 
> 
>> using Ghostscript 9.02 and 1.4.6 on RHEL 5.3, same results.  I didn't
>> bother breaking the laserjet III out, as i had a different laserjet
>> already set up.  same results, always pulled from default tray and not
>> the tray specified by the PPD.  are you able to get 8.70-1 GS and 1.4.6
>> on the same box to see if GS is the culprit?  in a pinch, you could write
>> a script filter probably to insert the tray commands in the job, but
>> that's really a hack way, somewhere along the line it does appear that
>> something broke.  I'll keep tinkering too and see if I can find the
>> smoking gun as well.
>>
>> > Added info:
>> >
>> > I just put CUPS 1.4.6 on a Fedora 14 box and tried the same setup and
>> > it's broken here too. No tray selection.
>> >
>> > Cups 1.4.6 on Fedora 14 using Ghostscript 8.71-16
>> >
>> > Jeff
>> >
>> >
>> >
>> > > Hello
>> > >
>> > > The working unit is using Ghostscript 8.70-1, it's running 1.3.11 on
>> > > a Fedora 10 box.
>> > >
>> > > The new server is using 8.71_6 running 1.4.6 on a freebsd 8.2 server.
>> > >
>> > > I will attempt to do the print to file thing next.
>> > >
>> > > Jeff
>> > >
>> > >
>> > >
>> > > > > > Paul wrote:
>> > > > > >
>> > > > > > > Well, this may take me a little longer to test...
>> > > > > > >
>> > > > > > > with 1.4.6 the mime.types and mime.convs appear to have
>> > > > > > > moved, but when I
>> > > > > > > found them, i don't see an issue.  whenever I lp any type of
>> > > > > > > file (text or
>> > > > > > > postscript) it says unsupported format.  looking through the
>> > > > > > > PPD and mime.convs file, I see a path to get there, but i
>> > > > > > > can't figure out why it
>> > > > > > > isn't getting there.  we're mostly a RAW shop but I'll keep
>> > > > > > > playing with it and let you know.
>> > > > > > >
>> > > > > >
>> > > > > > As the final content type requested by the filter specified in
>> > > > > > the printer's PPD (rastertohp), there must be a filter
>> > > > > > pstoraster (or gstoraster) installed. These filters are part of
>> > > > > > Ghostscript and get only installed if Ghostscript has been
>> > > > > > compiled with the cups device included. Missing these filters
>> > > > > > causes the unsupported format message. (I remember there is a
>> > > > > > RFE for cupsd to tell missing filters in cases like this, but
>> > > > > > this has been postponed to a future release).
>> > > > > >
>> > > > > > Helge
>> > > > > >
>> > > > >
>> > > > > At the risk of sounding stupid here (*NOTE* my expertise is more
>> > > > > around Printer communications / hardware as well as software /
>> > > > > script design, not compiling) I'm not finding any doc on how to
>> > > > > compile GhostScript with the cups device included.  I found the
>> > > > > chunk of code in the devs.mak file for CUPS, but not seeing how
>> > > > > to "turn it on"  can you point me in the right direction at all? 
>> > > > > I've googled, looked on cups.org, GhostScripts Site and came up
>> > > > > empty.
>> > > >
>> > > > Nevermind, default configure and make seem to do the trick and make
>> > > > the gstoraster.convs file for me as well
>> > >
>> >
>>

Looking into the sources of both Ghostscript's cups device (gs 8.64, 8.71, 
9.01) and CUPS' rastertohp (1.3.5, 1.4.5) filter gave no hints to the cause 
of th issue.
So I'd suggest to set up a "printer" using the file device and otherwise set 
up identical to the printer in question, switch the cups logging to debug, 
print a sample exhibiting the issue, and then post (an URL to) 
- the source file
- the PPD used
- the portion of the error_log containing the messages of the respective job
- the output file.

Helge





More information about the cups mailing list