[cups.bugs] [HIGH] STR #1424: Temporary file directory not set

Till Kamppeter till.kamppeter at gmx.net
Wed Feb 15 12:47:52 PST 2006


[STR New]

Now I am on rev 5108 and can finally print again, on a PostScript printer.
If I try on an HP PhotoSmart 2600 (HPLIP/HPIJS/GhostScript) I get no
printout, but GhostScript complaining in error_log (LogLevel debug):

E [15/Feb/2006:22:03:57 +0100] [Job 9] renderer command: gs -q -dBATCH
-dPARANOI
DSAFER -dQUIET -dNOPAUSE -sDEVICE=ijs -sIjsServer=hpijs
-sDeviceManufacturer="HE
WLETT-PACKARD" -sDeviceModel="deskjet 5600" -dDEVICEWIDTHPOINTS=595
-dDEVICEHEIG
HTPOINTS=842 -dDuplex=false -r300
-sIjsParams=Quality:Quality=0,Quality:ColorMod
e=2,Quality:MediaType=0,Quality:PenSet=2,PS:MediaPosition=7
-dIjsUseOutputFD -sO
utputFile=- -
E [15/Feb/2006:22:03:57 +0100] [Job 9] foomatic-gswrapper: gs '-dBATCH'
'-dPARAN
OIDSAFER' '-dQUIET' '-dNOPAUSE' '-sDEVICE=ijs' '-sIjsServer=hpijs'
'-sDeviceManu
facturer=HEWLETT-PACKARD' '-sDeviceModel=deskjet 5600'
'-dDEVICEWIDTHPOINTS=595'
 '-dDEVICEHEIGHTPOINTS=842' '-dDuplex=false' '-r300'
'-sIjsParams=Quality:Qualit
y=0,Quality:ColorMode=2,Quality:MediaType=0,Quality:PenSet=2,PS:MediaPosition=7'
 '-dIjsUseOutputFD' '-sOutputFile=/dev/fd/3' '/dev/fd/0' 3>&1 1>&2
E [15/Feb/2006:22:03:59 +0100] [Job 9] ESP Ghostscript 815.01: **** Could
not op
en temporary file /root/tmp/gs_5uOBAG
E [15/Feb/2006:22:03:59 +0100] [Job 9] ESP Ghostscript 815.01: Could not
open th
e scratch file /root/tmp/gs_5uOBAG.
E [15/Feb/2006:22:03:59 +0100] [Job 9] **** Unable to open the initial
device, q
uitting.
E [15/Feb/2006:22:03:59 +0100] [Job 9] renderer return value: 1

GhostScript being called by the foomatic-rip filter is running as the user
"lp" as every filter, but it tries to write into roots personal tmp
directory. This looks like that the environment variable for the temporary
directory is badly set.

I can remedy this problem by manually setting

TempDir /var/spool/cups/tmp

in /etc/cups/cupsd.conf

According to the documentation, this setting should be the default. It
seems also that it was the default up to last week's snapshot (rev 5083).

Link: http://www.cups.org/str.php?L1424
Version: 1.2-current





More information about the cups-devel mailing list