Both openoffice-1.1.3 and openoffice-bin-1.1.3 return the following when I type in 'ooffice' in a terminal window: $ ooffice running openoffice.org setup... /usr/bin/ooffice: line 178: 29675 Segmentation fault /opt/OpenOffice.org/program/setup ${UPDATEFLAG} -R:/etc/openoffice/autoresponse-1.1.3.conf 1>&/dev/null setup failed.. abort openoffice-1.1.2 and openoffice-bin-1.1.2 both run on my system without error. I have also removed /etc/openoffice, ~/.openoffice and ~/.sversionrc before re-emerging openoffice-bin-1.1.3. I still get the above error. I will attach an "emerge info" output and a strace of my launch of ooffice. This system has had 'emerge -u --deep world' run on it about weekly. The only ACCEPT_KEYWORDS="~x86" packages are mozilla-firefox and mozilla-launcher. The kernel: 2.4.27-rr03 is my third compile of the vanilla-sources-2.4.27 package. If I need to provide any more information, please ask. Reproducible: Always Steps to Reproduce: 1.ooffice 2. 3. Actual Results: running openoffice.org setup... /usr/bin/ooffice: line 178: 29675 Segmentation fault /opt/OpenOffice.org/program/setup ${UPDATEFLAG} -R:/etc/openoffice/autoresponse-1.1.3.conf 1>&/dev/null setup failed.. abort Expected Results: Opened the Open Office main window.
Created attachment 42186 [details] Output of "emerge info"
Created attachment 42187 [details] strace output (last 100 lines of about 300)
Any progress on this? Ive just submitted 68631 because OO.org 1.1.2 has just been removed! See also 65406.
What's the "-pipe -W -Wall" part in your CFLAGS for? Could you try to re
Sorry forget the last comment, this was posted by accident...
Same problem on my installation. Running oosetup as root works fine. Perhaps this is a rights problem? Regards, -Martin.
Yes.. running OO.o 1.1.3 works as root. This was not the case with 1.1.2.
Sorry, as in.. it works as root with 1.1.3, root and users in 1.1.2.
copying the /root/.openoffice dir into the user's home and creating an appropriate .sversionrc seems to work :)
I attempted to run ooffice (v. 1.1.3) as root. I still receive the same 'Segmentation Fault'
Anyone caring to fix this thing?
http://forums.gentoo.org/viewtopic.php?t=224306
Same problem. I think this is severity "critical" because we can't use OpenOffice at all. # qpkg -I -v openoffice app-office/openoffice-bin-1.1.3 * # emerge --info Portage 2.0.51-r2 (gcc34-x86-2004.2, gcc-3.4.2, glibc-2.3.4.20041021-r0, 2.6.9-ck2 i686) ================================================================= System uname: 2.6.9-ck2 i686 AMD Athlon(tm) XP 1800+ Gentoo Base System version 1.6.4 ccache version 2.3 [enabled] Autoconf: sys-devel/autoconf-2.59-r5 Automake: sys-devel/automake-1.8.5-r1 Binutils: sys-devel/binutils-2.15.92.0.2-r1 Headers: sys-kernel/linux26-headers-2.6.8.1-r1 Libtools: sys-devel/libtool-1.5.2-r5 ACCEPT_KEYWORDS="x86 ~x86" AUTOCLEAN="yes" CFLAGS="-march=athlon-xp -O2 -pipe" CHOST="i686-pc-linux-gnu" COMPILER="" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3/share/config /usr/lib/mozilla/defaults/pref /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-march=athlon-xp -O2 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs candy ccache distlocks fixpackages sandbox userpriv usersandbox" GENTOO_MIRRORS="ftp://ftp.tu-clausthal.de/pub/linux/gentoo ftp://sunsite.informatik.rwth-aachen.de/pub/Linux/gentoo/" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/opt/portage /opt/portage-kde" SYNC="rsync://rsync.de.gentoo.org/gentoo-portage" USE="3dnow X aalib acl acpi alsa apm arts artswrappersuid avi berkdb bitmap-fonts cdr crypt cups dga divx4linux dvb dvd encodef77 faad flac foomaticdb gdbm gif gnome gphoto2 gpm gstreamer gtk gtk2 hal imap imlib ipv6 java jpeg kde lcms libg++ libwww mad maildir mikmod mmx mng mono motif mozilla mozsvg mpeg mule ncurses nls nptl offensive oggvorbis opengl oss pam pcre pdflib perl pic png python qt quicktime readline sasl scanner sdl slang spell sse ssl svg tcltk tcpd tetex theora threads tiff truetype unicode usb x86 xml2 xmms xv xvid zlib" $ gdb /opt/OpenOffice.org/program/setup.bin GNU gdb 6.2.1 Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i686-pc-linux-gnu"...(no debugging symbols found)...Using host libthread_db library "/lib/tls/libthread_db.so.1". (gdb) run Starting program: /opt/OpenOffice.org/program/setup.bin (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...[Thread debugging using libthread_db enabled] [New Thread -1241569600 (LWP 27045)] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1241569600 (LWP 27045)] 0x00000000 in ?? () (gdb) bt #0 0x00000000 in ?? () #1 0xb6f551b3 in _STL::basic_string<char, _STL::char_traits<char>, _STL::allocator<char> >::reserve () from /opt/OpenOffice.org/program/libstlport_gcc.so #2 0xb6f92643 in _STL::_Init_timeinfo () from /opt/OpenOffice.org/program/libstlport_gcc.so #3 0xb6f9982f in _STL::_Locale_impl::make_classic_locale () from /opt/OpenOffice.org/program/libstlport_gcc.so #4 0xb6f9a39e in _STL::locale::_S_initialize () from /opt/OpenOffice.org/program/libstlport_gcc.so #5 0xb6f9a2f5 in _STL::ios_base::_Loc_init::_Loc_init () from /opt/OpenOffice.org/program/libstlport_gcc.so #6 0xb6f9b627 in _STL::locale::locale () from /opt/OpenOffice.org/program/libstlport_gcc.so #7 0xb6f9b64a in _STL::locale::locale () from /opt/OpenOffice.org/program/libstlport_gcc.so #8 0xb6fad255 in _STL::basic_string<wchar_t, _STL::char_traits<wchar_t>, _STL::allocator<wchar_t> >::replace<wchar_t const*> () from /opt/OpenOffice.org/program/libstlport_gcc.so #9 0xb6f4cc0a in _init () from /opt/OpenOffice.org/program/libstlport_gcc.so #10 0xb7ff6a54 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2 #11 0xb7ff6b0e in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2 #12 0xb7feb865 in ?? () from /lib/ld-linux.so.2
changing the permissions to /dev/dri/card0 to rw for everyone worked for me
I have no /dev/dri at all. My X-server is configured to use the nv driver and I have full rw-permissions to /dev/nvidia0 and /dev/nvidiactl.
Reinstalling openoffice-bin fixed the issue here. I can't explain it, because the binary setup.bin which segfaulted should be the same as before.
usually if you use strace to start the program, you can see what it tries to do right before it crashes. If it is trying to open some sort of device you can make sure the permissions are correct or see if the device is even available. For example /dev/dri/card0 is created with the agpgart driver. Nvidia has their own devices I believe, but can use the agpgart driver if it is loaded first.
rt_sigprocmask(SIG_BLOCK, [CHLD], [RTMIN], 8) = 0 rt_sigaction(SIGINT, {0x8077213, [], 0}, {SIG_DFL}, 8) = 0 rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0 rt_sigaction(SIGINT, {SIG_DFL}, {0x8077213, [], 0}, 8) = 0 rt_sigprocmask(SIG_BLOCK, NULL, [RTMIN], 8) = 0 stat64("/home/craig/.openoffice/1.1.3", 0xbfffebd0) = -1 ENOENT (No such file or directory) rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0 fstat64(1, {st_mode=S_IFCHR|0600, st_rdev=makedev(136, 2), ...}) = 0 old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40183000 write(1, "running openoffice.org setup...\n", 32running openoffice.org setup... ) = 32 rt_sigprocmask(SIG_BLOCK, [INT CHLD], [RTMIN], 8) = 0 fork() = 29697 rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0 rt_sigprocmask(SIG_BLOCK, [CHLD], [RTMIN], 8) = 0 rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0 rt_sigprocmask(SIG_BLOCK, [CHLD], [RTMIN], 8) = 0 rt_sigaction(SIGINT, {0x8077213, [], 0}, {SIG_DFL}, 8) = 0 waitpid(-1, [WIFSIGNALED(s) && WTERMSIG(s) == SIGSEGV], 0) = 29697 fstat64(2, {st_mode=S_IFCHR|0600, st_rdev=makedev(136, 2), ...}) = 0 old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40184000 uname({sys="Linux", node="amilo", ...}) = 0 write(2, "/usr/bin/oosetup: line 178: 2969"..., 161/usr/bin/oosetup: line 178: 29697 Segmentation fault /opt/OpenOffice.org/program/setup ${UPDATEFLAG} -R:/etc/openoffice/autoresponse-1.1.3.conf >&/dev/null ) = 161 rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0 --- SIGCHLD (Child exited) @ 0 (0) --- waitpid(-1, 0xbfffea5c, WNOHANG) = -1 ECHILD (No child processes) sigreturn() = ? (mask now [RTMIN]) rt_sigaction(SIGINT, {SIG_DFL}, {0x8077213, [], 0}, 8) = 0 write(1, "setup failed.. abort\n", 21setup failed.. abort ) = 21 rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0 munmap(0x40184000, 4096) = 0 munmap(0x40183000, 4096) = 0 exit_group(1) = ? Any ideas? I've also set rw on /dev/dri/card0 and /dev/radeon
xorg 6.8.0-r1 didnt help either...
err.. r2
Alternatively, if you have Goudy Old Style font installed (mine is from a good quality source mind you, Corel CD), remove it and the crashes might stop. I had to installed oo.org 1.1.1 then hack (S&R) the setup checker script of 1.1.3 once reinstalled to make it look for my 1.1.1 setup dir as 1.1.2 is no longer installed on gentoo. Once I ran it, I was then getting the oo.org report crash back to oo.org dialog . From there I could get a good strace to find the font (which is good and works in 1.1.1 and 1.1.2)
I just successfully: ACCEPT_KEYWORDS="~x86" emerge openoffice and now have openoffice-1.1.4 on my system. Unfortunatly I STILL get the following when I try to run openoffice. (I did remove the ~/.openoffice from openoffice-1.1.1 first) ooffice running openoffice.org setup... /usr/bin/ooffice: line 178: 32461 Segmentation fault /opt/OpenOffice.org/program/setup ${UPDATEFLAG} -R:/etc/openoffice/autoresponse-1.1.4.conf 1>&/dev/null setup failed.. abort Should I open a new bug for openoffice-1.1.4? I ran "emerge -u --deep world" before the above list emerge. openoffice-1.1.1 (or openoffice-bin-1.1.1) is still the latest openoffice that will work on my system.
Could you please emerge sync and try again, I've modified the wrapper-script in regards to fonts, so this might fix it. To all the others, commenting about the /dev/dri/card0 issue. This is another issue as Richard seems to have, please comment here instead: http://bugs.gentoo.org/show_bug.cgi?id=70397
@Richard: Could it be that you don't have a mounted /proc filesystem? open("/proc/sys/kernel/ngroups_max", O_RDONLY) = -1 ENOENT (No such file or directory) open("/proc/sys/kernel/ngroups_max", O_RDONLY) = -1 ENOENT (No such file or directory) open("/proc/sys/kernel/ngroups_max", O_RDONLY) = -1 ENOENT (No such file or directory)
I have /proc/sys/kernel, but no ngroups_max.
Solved: It was a font issue, but not with any fonts I emerged via Portage. The problem was some old Adobe fonts that I have in /usr/local/lib/fonts. Althouth I successfully ran mkfontscale mkfontdir -e /usr/share/fonts/encodings -e /usr/share/fonts/encodings/large in /usr/local/lib/fonts, OpenOffice crashed on startup if these fonts were present. (OpenOffice 1.1.1 did not have this problem.) These old Adobe fonts came on a 3.5" floppy I bought about 10 years ago and seem to work with other apps. The fonts include such standards as Helvetica, Times and Courier and the files look like: hv______.afm hv______.pfb hvb_____.afm hvb_____.pfb hvbo____.afm hvbo____.pfb hvo_____.afm hvo_____.pfb I wonder why those old Adobe fonts have issues? I don't know if this bug is 'resolved'. Should the checking look for non portage installed fonts?
I had similarly named hv* fonts at one point for testing a Scribus/EPS issue, and also found them to be suspect and nuked them.
@Richard: The problem is, that the setup script can't know where you put all your shiny fonts... Just put them in the defaul location /usr/share/fonts and they all should be found by OOo (I've modifiede the startup script to do this a few days ago). About the segfault: This is definately an upstream problem, if you want it solved, please comment here: http://www.openoffice.org/issues/show_bug.cgi?id=37703 Looks like the same bug to me at least. Not a lot we can do, also there is a workaround so closing. Thanks for tracking that one down!