Openbox woes (SOLVED)

[manicman01]

I'm starting to think I have bad luck

Openbox happens to be my favourite window manager. I've managed to compile the latest version from source several times with other distros, but Gentoo is another story; neither the tarball nor the older version in the portage tree want to work for me. (That being the case, I'm still trying to get 3.4.5 -- compiled manually -- up and running). What's puzzling is, I don't get any errors until after "make install" is completed.

This time around, I get

[didumos]

Well, it looks like openbox is trying to allocate almost 2GB of memory for some reason

Firstly, openbox-3.4.5 is in portage but it is marked as ~arch (unstable). So, add "x11-wm/openbox" to /etc/portage/package.keywords to install it through portage, if you wish

The only thing I can think to suggest to fix your actual problem would be to try openbox with a clean config and see if it still happens - i.e. move ~/.config/openbox/ somewhere else for the moment so that openbox opens with default settings.

[manicman01]

Good idea, except I haven't created any config files yet, so the settings should already be the defaults.

Emerging 3.4.5 results in an "all your base are belong to us" signal-11 error message, so I don't know what's up.

Bump?

Harry

EDIT: oops, I take it back. When I emerge 3.4.5, it's "Failed to open the display from the DISPLAY environment variable."

[didumos]

Hmm, can you post the output of emerge --info please.

[manicman01]

Portage 2.1.3.19 (default-linux/x86/2007.0, gcc-4.1.2, glibc-2.6.1-r0, 2.6.23-gentoo-r6 i686)
=================================================================
System uname: 2.6.23-gentoo-r6 i686 Intel(R) Core(TM)2 CPU 6400 @ 2.13GHz
Timestamp of tree: Sun, 27 Jan 2008 19:00:01 +0000
app-shells/bash: 3.2_p17-r1
dev-lang/python: 2.4.4-r6
dev-python/pycrypto: 2.0.1-r6
sys-apps/baselayout: 1.12.10-r5
sys-apps/sandbox: 1.2.18.1-r2
sys-devel/autoconf: 2.13, 2.61-r1
sys-devel/automake: 1.5, 1.7.9-r1, 1.8.5-r3, 1.10
sys-devel/binutils: 2.18-r1
sys-devel/gcc-config: 1.4.0-r4
sys-devel/libtool: 1.5.24
virtual/os-headers: 2.6.23-r3
ACCEPT_KEYWORDS="x86"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-Os -march=i686 -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc"
CONFIG_PROTECT_MASK="/etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/revdep-rebuild /etc/terminfo /etc/udev/rules.d"
CXXFLAGS="-Os -march=i686 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="distlocks metadata-transfer sandbox sfperms strict unmerge-orphans userfetch"
GENTOO_MIRRORS="ftp://mirrors.tera-byte.com/pub/gentoo http://gentoo.mirrors.tera-byte.com/ "
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages --filter=H_**/files/digest-*"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="acl berkdb bitmap-fonts cli cracklib crypt cups dri fortran gdbm gpm iconv ipv6 isdnlog midi mudflap ncurses nls nptl nptlonly openmp pam pcre perl pppd python readline reflection session spl ssl tcpd truetype-fonts type1-fonts unicode x86 xorg zlib" ALSA_CARDS="ali5451 als4000 atiixp atiixp-modem bt87x ca0106 cmipci emu10k1 emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel intel8x0 intel8x0m maestro3 trident usb-audio via82xx via82xx-modem ymfpci" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mulaw multi null plug rate route share shm softvol" APACHE2_MODULES="actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias" ELIBC="glibc" INPUT_DEVICES="keyboard mouse evdev" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" USERLAND="GNU" VIDEO_CARDS="apm ark chips cirrus cyrix dummy fbdev glint i128 i740 i810 imstt mach64 mga neomagic nsc nv r128 radeon rendition s3 s3virge savage siliconmotion sis sisusb tdfx tga trident tseng v4l vesa vga via vmware voodoo"
Unset: CPPFLAGS, CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, LDFLAGS, LINGUAS, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS

[didumos]

How are you actually trying to start openbox?

I start it using the startx command from the console with the following in ~/.xinitrc

[manicman01]

I've tried starting openbox-session in my ~/.xinitrc, just openbox by itself from a terminal window... both have the same result.

Also, for some unexplained reason, I'm now getting the Signal 11 error again.

Ah, the joys of Linux

It's a good thing I'm not easily discouraged.

[didumos]

Does openbox --debug tell us anything useful?

Another cause of it segfaulting could be that something it depends on has changed, so does revdep-rebuild -p (as root) pick up anything?

[manicman01]

SESSION: Connecting to SM with id: (null)
SESSION: Connected to SM with id: (null)
DEBUG: Failed to connect to session manager: SESSION_MANAGER environment variable is not defined
How are you gentlemen? All your base are belong to us. (Openbox received signal 11)
Aborted

(That last bit really isn't necessary, I just think it's funny )

Oops, I missed the last line of your post. I'll give it a try.

EDIT: revdep-rebuild -p turns up nothing, I'm afraid... everything is consistent.

[manicman01]

Google gives me the impression that said environment variable isn't vital, either, so I'm stumped, for the moment.

[didumos]

Yes, the session manager isn't vital at all and shouldn't segfault openbox.

I take it that other window managers (e.g. twm) work and this isn't a problem with X?

If not, the next thing to try would be to get a backtrace using gdb:

1) emerge gdb (if it isn't installed already)

2) Open a shell/terminal and run ulimit -c unlimited. This (hopefully) allows openbox to do a core dump when it crashes.

3) Run openbox from the same shell. When it crashes it should produce the core dump in the directory you run it from.

4) Run gdb /path/to/openbox /path/to/core/file

5) Type bt full and press return to (hopefully) get a backtrace that you can paste here.

(I hope this works but you might have to re-compile openbox with ./configure --enable-debug to get something useful, we'll see )

Edit: I just tried it myself and you will need to re-compile openbox with debugging enabled.

[manicman01]

I was able to install pekwm, so I think this is a problem with just Openbox. Xorg runs fine; I have "exec urxvt" in my .xinitrc currently, so that it actually starts up.

[manicman01]

This is a bit off-topic, but I haven't seen anything recent on it. Do you know of a command that would allow me to unmerge a package along with any unused dependencies? It's the only thing *.deb-based distributions have going for them, IMHO. (I'm thinking of "aptitude".)

Harry

[didumos]

Yeah, I don't think portage can do that on a per-package basis, but what you can do after you unmerge a package is run emerge --depclean -p. This will show the unused packages on your system. If you want to go ahead and remove them then remove the '-p' flag.

[manicman01]

Just recompiled. Should I now take another core dump?

[manicman01]

The really odd thing is that when I run those commands now, I don't end up with a file ... bah humbug

I should add that Openbox says "core dumped," but there isn't one this time. openbox --debug gives the same old message.

[didumos]

Hmm, very weird. Just before you run openbox make sure the output of ulimit -c says unlimited. That's all you should have to do.

[manicman01]

Yep. ls -a shows diddly, I'm afraid.

I solved the problem (partly, anyway). When I run Openbox as a normal user, the core file shows up

[didumos]

Eek, I should have said that it should be run as a normal user, oops. Ok so can you run gdb over that file then and get a backtrace.

[manicman01]

As far as I can tell, the only interesting line is this:

#0 0xffffe402 in __kernel_vsyscall ()

the rest goes

No symbol table info available.
#1 0xb7a64101 in ?? ()
No symbol table info available.
#2 0xb7b66ff4 in ?? ()
No symbol table info available.
#3 0xb7a396c0 in ?? ()
No symbol table info available.
#4 0xbf8cb2f8 in ?? ()
No symbol table info available.
#5 0xb7a658e8 in ?? ()
No symbol table info available.
#6 0x00000006 in ?? ()
No symbol table info available.
#7 0xbf8cb26c in ?? ()
No symbol table info available.
#8 0x00000000 in ?? ()
No symbol table info available.

[didumos]

Are you sure you are running it with the openbox binary that you compiled with --enable-debug and not the one in /usr/bin/openbox?

e.g: something like this, with the openbox-3.4.5 package extracted to ~/openbox-3.4.5/

> cd ~/openbox-3.4.5/
> ./configure --enable-debug
> make
> cd openbox/.libs/
> ulimit -c unlimited
> ./openbox
> gdb ./openbox core

Edit: had the wrong path above for the binary, fixed now

[manicman01]

I wouldn't be surprised if I made a silly mistake like that. I'll try again.

[manicman01]

All goes well until the last command; again, it doesn't want to give me a core file. Says it's there, but it isn't -- this time as a normal user. Pardon my French, but WTF?

[didumos]

Ok, I'm running out of ideas now I don't know why openbox wouldn't produce a core dump under those conditions.

Let's take a look at which package versions you have installed, can you run emerge -epvt openbox > emerge.out and upload the file emerge.out to somewhere like http://omploader.org. Here is mine for comparison: http://omploader.org/vYzJ0

[manicman01]

Here goes:
http://omploader.org/vYzJ5/emerge.out