Hello, evince segfaults when trying to print PDF files on a network printer. It looks to be related to this bug report on Ubuntu: https://bugs.launchpad.net/ubuntu/+source/cups/+bug/660731 Printing a PDF file with a KDE application works fine. Printing (not a PDF file) with another gnome application works fine also. I did a few tracing with gdb (after having installed evince-dbg). Here it is below. Cheers Damien =============================== Starting program: /usr/bin/evince [Thread debugging using libthread_db enabled] [New Thread 0xb6219940 (LWP 10930)] [New Thread 0xb6056b70 (LWP 10933)] [New Thread 0xb529cb70 (LWP 10934)] [New Thread 0xb48ffb70 (LWP 10935)] [New Thread 0xb40feb70 (LWP 10936)] [Thread 0xb48ffb70 (LWP 10935) exited] [Thread 0xb40feb70 (LWP 10936) exited] [New Thread 0xb40feb70 (LWP 10937)] [Thread 0xb529cb70 (LWP 10934) exited] [New Thread 0xb529cb70 (LWP 10938)] [New Thread 0xb48ffb70 (LWP 10939)] [New Thread 0xb35e0b70 (LWP 10940)] [Thread 0xb40feb70 (LWP 10937) exited] [Thread 0xb35e0b70 (LWP 10940) exited] [Thread 0xb529cb70 (LWP 10938) exited] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0xb6219940 (LWP 10930)] 0xb2db5c4b in ppdFindChoice () from /usr/lib/libcups.so.2 (gdb) bt #0 0xb2db5c4b in ppdFindChoice () from /usr/lib/libcups.so.2 #1 0xb4a0d9c8 in ?? () from /usr/lib/gtk-2.0/2.10.0/printbackends/libprintbackend-cups.so #2 0x094a2b08 in ?? () #3 0x00000004 in ?? () #4 0xb2db94eb in ppdPageSize () from /usr/lib/libcups.so.2 #5 0xb71dae46 in gtk_printer_get_default_page_size () from /usr/lib/libgtk-x11-2.0.so.0 #6 0xb71e4f28 in ?? () from /usr/lib/libgtk-x11-2.0.so.0 #7 0x0942cc08 in ?? () #8 0xfffffffb in ?? () #9 0x00000001 in ?? () #10 0xbfa07358 in ?? () #11 0xffffffff in ?? () #12 0x00000011 in ?? () #13 0x00000010 in ?? () #14 0xb714310e in gtk_tree_view_get_type () from /usr/lib/libgtk-x11-2.0.so.0 #15 0xb6ca5c08 in g_cclosure_marshal_VOID__BOOLEAN () from /usr/lib/libgobject-2.0.so.0 #16 0xb6c9813a in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0 #17 0xb6cae61d in ?? () from /usr/lib/libgobject-2.0.so.0 #18 0x094c0e18 in ?? () #19 0x00000000 in ?? () (gdb) info threads 8 Thread 0xb48ffb70 (LWP 10939) 0xb77cb424 in __kernel_vsyscall () 2 Thread 0xb6056b70 (LWP 10933) 0xb77cb424 in __kernel_vsyscall () * 1 Thread 0xb6219940 (LWP 10930) 0xb2db5c4b in ppdFindChoice () from /usr/lib/libcups.so.2
Hello, I have upgraded my cups packages to testing, it seems that the problem is fixed. If you have interested I have kept a core file for the problem I have reported yesterday. Currently (with issue fixed) the list of cups packages on my computer is: $ dpkg -l '*cups*' | grep ^i ii cups 1.4.4-7 Common UNIX Printing System(tm) - server ii cups-client 1.4.4-7 Common UNIX Printing System(tm) - client pro ii cups-common 1.4.4-7 Common UNIX Printing System(tm) - common fil ii cups-driver-gutenprint 5.2.6-1 printer drivers for CUPS ii cups-ppdc 1.4.4-7 Common UNIX Printing System(tm) - PPD manipu ii cupsddk 1.4.4-7 Common UNIX Printing System (transitional pa ii ghostscript-cups 8.71~dfsg2-6 The GPL Ghostscript PostScript/PDF interpret ii libcups2 1.4.4-7 Common UNIX Printing System(tm) - Core libra ii libcups2-dev 1.4.4-7 Common UNIX Printing System(tm) - Developmen ii libcupscgi1 1.4.4-7 Common UNIX Printing System(tm) - CGI librar ii libcupsdriver1 1.4.4-7 Common UNIX Printing System(tm) - Driver lib ii libcupsimage2 1.4.4-7 Common UNIX Printing System(tm) - Raster ima ii libcupsmime1 1.4.4-7 Common UNIX Printing System(tm) - MIME libra ii libcupsppdc1 1.4.4-7 Common UNIX Printing System(tm) - PPD manipu ii libcupsys2 1.3.8-1+lenny8 Common UNIX Printing System (transitional pa ii libgnomecups1.0-1 0.2.3-3 GNOME library for CUPS interaction ii python-cups 1.9.31-1 Python bindings for CUPS ii python-cupsutils 1.0.0-4lenny1 Python utility modules around the CUPS print Cheers, Damien
severity 606685 important thanks Le vendredi 10 décembre 2010 à 23:25 +0100, Damien Couroussé a écrit : Thanks for your report. Are you sure only cups and its libraries were upgraded when you attempted to install cups from squeeze? The other possible culprit would be GTK+, which has seen a number of network-printing-related bug fixes in the meantime. Cheers,
Hi, Yes, while debugging a bit this problem, I have seen that the segfault in libcups comes from a call in libgtk. However, only cups and the associated libs were upgraded for my concern: I asked for an upgrade of libcups2 from lenny to squeeze (I am mostly based on lenny but use a few packages from squeeze and/or lenny-backports). Here is the log from /var/log/aptitude: ################################################## Will install 11 packages, and remove 14 packages. 91.1MB of disk space will be freed =============================================================================== [REMOVE, NOT USED] foomatic-db-gutenprint [REMOVE, NOT USED] foomatic-db-hpijs [REMOVE, NOT USED] ijsgutenprint [REMOVE, NOT USED] libgutenprint2 [REMOVE, NOT USED] libijs-0.35 [REMOVE, NOT USED] openprinting-ppds [REMOVE, NOT USED] python-foomatic [REMOVE, NOT USED] python-ipy [HOLD, DEPENDENCIES] dpkg-dev [HOLD, DEPENDENCIES] gcc-4.4-base [HOLD, DEPENDENCIES] libicu44 [HOLD, DEPENDENCIES] libmagick++10 [HOLD, DEPENDENCIES] libmagick10 [HOLD, DEPENDENCIES] libpurple0 [HOLD, DEPENDENCIES] libx11-6 [HOLD, DEPENDENCIES] libx11-dev [HOLD, DEPENDENCIES] libxml2-utils [HOLD, DEPENDENCIES] linux-image-2.6.26-2-686 [HOLD, DEPENDENCIES] linux-libc-dev [HOLD, DEPENDENCIES] linux-patch-debian-2.6.26 [HOLD, DEPENDENCIES] linux-source-2.6.26 [HOLD, DEPENDENCIES] linux-source-2.6.32 [HOLD, DEPENDENCIES] linux-support-2.6.26-2 [HOLD, DEPENDENCIES] mysql-common [HOLD, DEPENDENCIES] openssl [HOLD, DEPENDENCIES] pidgin-data [HOLD, DEPENDENCIES] python-libxml2 [HOLD, DEPENDENCIES] xserver-xorg-core [INSTALL, DEPENDENCIES] libgssapi-krb5-2 [INSTALL, DEPENDENCIES] libk5crypto3 [INSTALL, DEPENDENCIES] libkrb5-3 [INSTALL, DEPENDENCIES] libkrb5support0 [REMOVE, DEPENDENCIES] cups [REMOVE, DEPENDENCIES] cups-driver-gutenprint [REMOVE, DEPENDENCIES] hal-cups-utils [REMOVE, DEPENDENCIES] hplip-cups [REMOVE, DEPENDENCIES] printconf [HOLD] base-files [HOLD] dpkg [HOLD] exim4 [HOLD] exim4-base [HOLD] exim4-config [HOLD] exim4-daemon-light [HOLD] imagemagick [HOLD] libssl0.9.8 [HOLD] libstdc++6 [HOLD] libxml2 [HOLD] linux-tree-2.6.26 [HOLD] pidgin [REMOVE] libcupsys2 [UPGRADE] libcups2 1.3.8-1+lenny8 -> 1.4.5-1 [UPGRADE] libgcrypt11 1.4.1-1 -> 1.4.5-2 [UPGRADE] libgnutls26 2.4.2-6+lenny2 -> 2.8.6-1 [UPGRADE] libgpg-error0 1.4-2 -> 1.6-1 [UPGRADE] libkrb53 1.6.dfsg.4~beta1-5lenny4 -> 1.8.3+dfsg-3 [UPGRADE] libtasn1-3 1.4-1 -> 2.7-1 [UPGRADE] python-cups 1.9.31-1 -> 1.9.48-1 =============================================================================== Log complete. Aptitude 0.4.11.11: log report Fri, Dec 10 2010 21:11:15 +0100 IMPORTANT: this log only lists intended actions; actions which fail due to dpkg problems may not be completed. ################################################## I do use the following libs in GTK: $ dpkg -l '*libgtk*' | grep ^i ii libgtk-vnc-1.0-0 0.3.8-2 A VNC viewer widget for GTK+ (runtime libraries) ii libgtk2-perl 1:1.190-1 Perl interface to the 2.x series of the Gimp Too ii libgtk2.0-0 2.18.6-1~bpo50+1 The GTK+ graphical user interface library ii libgtk2.0-0-dbg 2.18.6-1~bpo50+1 The GTK+ libraries and debugging symbols ii libgtk2.0-bin 2.12.12-1~lenny2 The programs for the GTK+ graphical user interfa ii libgtk2.0-common 2.12.12-1~lenny2 Common files for the GTK+ graphical user interfa ii libgtkhtml2-0 2.11.1-2 HTML rendering/editing library - runtime files ii libgtkhtml3.14-19 3.18.3-1 HTML rendering/editing library - runtime files ii libgtkmm-2.4-1c2a 1:2.12.7-1 C++ wrappers for GTK+ 2.4 (shared libraries) ii libgtksourceview-common 1.8.5-1 common files for the GTK+ syntax highlighting wi ii libgtksourceview1.0-0 1.8.5-1 shared libraries for the GTK+ syntax highlightin ii libgtksourceview2.0-0 2.2.2-1 shared libraries for the GTK+ syntax highlightin ii libgtksourceview2.0-common 2.2.2-1 common files for the GTK+ syntax highlighting wi ii libgtkspell0 2.0.13-1+b1 a spell-checking addon for GTK's TextView widget Hope it helps, Cheers, Damien
reassign 606685 gtk+2.0 thanks Le lundi 13 décembre 2010 à 18:41 +0100, Damien Couroussé a écrit : Indeed, thanks for checking. HAH ! So this is the culprit. I knew this was suspicious, since PDF printing works perfectly on my lenny systems. My guess goes for an insufficient build-dependency on libcups2-dev in GTK+, leading the backport to not actually work with the cups version in lenny.