MacOS print to CUPS Server Fails

[M Yudkowsky]

I'm attempting to print from a MacOS 10.4 laptop to a CUPS server running Debian (unstable). Although the MacOS laptop can print successfully to a slightly older installation of a CUPS server, the attempt to print to this new server (and new printer, a Samsung CLP-300) fails.

I've tried this with three different printer drivers for the CLP-300 (the Samsung unified drivers and two open source drivers) but all fail.

Here's the output from one of the more verbose printers. At the start, we get:

D [26/Mar/2008:18:58:13 -0500] [Job 72] Reading PostScript input ...
D [26/Mar/2008:18:58:13 -0500] [Job 72] --> This document is DSC-conforming!
D [26/Mar/2008:18:58:13 -0500] [Job 72] Job claims to be DSC-conforming, but "%%BeginProlog" was missing before first line with another "%%Begin..." comment (is this a TeX/LaT
eX/dvips-generated PostScript file?). Assuming start of "Prolog" here.
D [26/Mar/2008:18:58:13 -0500] [Job 72]
D [26/Mar/2008:18:58:13 -0500] [Job 72] -----------
D [26/Mar/2008:18:58:13 -0500] [Job 72] Found: %%BeginProlog
D [26/Mar/2008:18:58:13 -0500] [Job 72] Found: %%EndProlog
D [26/Mar/2008:18:58:13 -0500] [Job 72]
D [26/Mar/2008:18:58:13 -0500] [Job 72] -----------
D [26/Mar/2008:18:58:13 -0500] [Job 72] Found: %%BeginSetup
D [26/Mar/2008:18:58:13 -0500] [Job 72] Found: %%BeginFeature: *Quality normal


Later one we see:

D [26/Mar/2008:18:58:13 -0500] [Job 72] Starting renderer
D [26/Mar/2008:18:58:13 -0500] [Job 72] JCL: <job data>
D [26/Mar/2008:18:58:13 -0500] [Job 72]
D [26/Mar/2008:18:58:13 -0500] [Job 72] renderer PID kid4=29662
D [26/Mar/2008:18:58:13 -0500] [Job 72] renderer command: foo2qpdl-wrapper   -c -r600x600 -p0 -s1 -m0   -d1 -n1 -C10 -Gsamclp300-0.icm
D [26/Mar/2008:18:58:14 -0500] [Job 72] pcache: unable to open '~/.ghostscript/cache/gs_cache'
D [26/Mar/2008:18:58:14 -0500] [Job 72] pcache: unable to open '~/.ghostscript/cache/gs_cache'
D [26/Mar/2008:18:58:14 -0500] [Job 72] pcache: unable to open '~/.ghostscript/cache/gs_cache'
D [26/Mar/2008:18:58:14 -0500] [Job 72] pcache: unable to open '~/.ghostscript/cache/gs_cache'
D [26/Mar/2008:18:58:14 -0500] [Job 72] pcache: unable to open '~/.ghostscript/cache/gs_cache'
D [26/Mar/2008:18:58:14 -0500] [Job 72] pcache: unable to open '~/.ghostscript/cache/gs_cache'
D [26/Mar/2008:18:58:14 -0500] [Job 72] pcache: unable to open '~/.ghostscript/cache/gs_cache'
D [26/Mar/2008:18:58:14 -0500] [Job 72] pcache: unable to open '~/.ghostscript/cache/gs_cache'
D [26/Mar/2008:18:58:14 -0500] [Job 72] GPL Ghostscript 8.62: Unrecoverable error, exit code 1
D [26/Mar/2008:18:58:14 -0500] [Job 72] cat: write error: Broken pipe
D [26/Mar/2008:18:58:14 -0500] [Job 72] sed: couldn't write 65 items to stdout: Broken pipe
D [26/Mar/2008:18:58:14 -0500] [Job 72] tail process done writing data to STDOUT
D [26/Mar/2008:18:58:14 -0500] [Job 72] KID4 finished
D [26/Mar/2008:18:58:14 -0500] [Job 72] Read 194 bytes of print data...
D [26/Mar/2008:18:58:14 -0500] [Job 72] KID3 finished
D [26/Mar/2008:18:58:14 -0500] Discarding unused printer-state-changed event...
D [26/Mar/2008:18:58:14 -0500] [Job 72] Wrote 194 bytes of print data...
D [26/Mar/2008:18:58:14 -0500] Discarding unused printer-state-changed event...
D [26/Mar/2008:18:58:14 -0500] [Job 72] Wrote 1 pages...
D [26/Mar/2008:18:58:14 -0500] [Job 72] Found: %%PageTrailer
D [26/Mar/2008:18:58:14 -0500] [Job 72] --> Continue DSC parsing now.
D [26/Mar/2008:18:58:14 -0500] [Job 72]
D [26/Mar/2008:18:58:14 -0500] PID 29654 (/usr/lib/cups/filter/pstops) exited with no errors.

Eventually this exits from the driver with no reported errors and the file is reported as being printed -- that's a driver issue. However, the "GPL Ghostscript 8.62: Unrecoverable error, exit code 1" is a different problem.

Running the d-something file in /var/spool/cups through gs,

# cat /var/spool/cups/d00072-001 |  gs -dNOPAUSE -dBATCH -dSAFER -sOutputFile=- -sDEVICE=bitcmyk - > /dev/null

for example, eventually produces a segfault. (I tried this with a previous driver, not my current driver, since the current driver disposes of the spooled files.)

Any suggestions would be quite welcome.