Back
[00:32:35] <linuxcnc-build> build #1314 of deb-lucid-rt-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-lucid-rt-binary-i386/builds/1314 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[14:50:57] <KGB-linuxcnc> 03Sebastian Kuzminsky 05seb/ubc3-deb ac88141 06linuxcnc 10(12 files) deb: new debian/configure and debian/control* * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=ac88141
[14:50:57] <KGB-linuxcnc> 03Sebastian Kuzminsky 05seb/ubc3-deb e9e0714 06linuxcnc 10debian/configure deb/conf: accept -a to mean -r * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=e9e0714
[14:50:57] <KGB-linuxcnc> 03Sebastian Kuzminsky 05seb/ubc3-deb 242ff7e 06linuxcnc 10debian/control.in deb/control: get rid of ancient Provides in control * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=242ff7e
[14:50:59] <KGB-linuxcnc> 03Sebastian Kuzminsky 05seb/ubc3-deb 84fcbca 06linuxcnc 10debian/control.posix.in 10debian/control.rtai.in 10debian/control.rtpreempt.in deb/control: relax the Conflicts * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=84fcbca
[14:51:03] <KGB-linuxcnc> 03Sebastian Kuzminsky 05seb/ubc3-deb 96f1723 06linuxcnc 10debian/rules.in deb/rules: dont use cpio when cp will do * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=96f1723
[14:51:07] <KGB-linuxcnc> 03Sebastian Kuzminsky 05seb/ubc3-deb 02ef845 06linuxcnc 10debian/linuxcnc.files.in 10src/Makefile install rsyslogd.conf file * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=02ef845
[14:51:11] <KGB-linuxcnc> 03Sebastian Kuzminsky 05seb/ubc3-deb 2dd468e 06linuxcnc 10src/configure.in 10src/rtapi/ulapi_autoload.c give dlopen the path to the ulapi library * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=2dd468e
[14:51:15] <KGB-linuxcnc> 03Sebastian Kuzminsky 05seb/ubc3-deb 07e9d60 06linuxcnc 10src/Makefile 10src/rtapi/Submakefile put ulapi-$flavor.so straight into rtlib/, not lib, so the dlopen can look in EMC2_RTLIB_DIR * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=07e9d60
[15:37:04] <cradek> cmorley: I also get 3.8.0 as glade-gtk2. this comes from precise/universe and I have no other alternatives listed
[15:58:57] <linuxcnc-build> build #25 of deb-precise-xenomai-binary-x86 is complete: Failure [4failed apt-get-update shell_1] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-xenomai-binary-x86/builds/25 blamelist: dummy, Sebastian Kuzminsky <seb@highlab.com>
[15:59:01] <linuxcnc-build> build #25 of deb-precise-xenomai-binary-amd64 is complete: Failure [4failed apt-get-update shell_1] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-xenomai-binary-amd64/builds/25 blamelist: dummy, Sebastian Kuzminsky <seb@highlab.com>
[16:01:01] <linuxcnc-build> build #25 of deb-precise-rtpreempt-binary-x86 is complete: Failure [4failed apt-get-update shell_1] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-rtpreempt-binary-x86/builds/25 blamelist: dummy, Sebastian Kuzminsky <seb@highlab.com>
[16:01:18] <linuxcnc-build> build #25 of deb-precise-rtpreempt-binary-amd64 is complete: Failure [4failed apt-get-update shell_1] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-rtpreempt-binary-amd64/builds/25 blamelist: dummy, Sebastian Kuzminsky <seb@highlab.com>
[16:49:44] <linuxcnc-build> build #1315 of deb-lucid-rt-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-lucid-rt-binary-i386/builds/1315 blamelist: dummy, Sebastian Kuzminsky <seb@highlab.com>
[18:09:32] <seb_kuzminsky> he was promoted
[19:29:35] <cmorley> cradek: Looking in the 'software manager' GUI there is glade and glade-gtk2 available.
[19:29:53] <cmorley> oh wait damn I'm using mint - sorry I forgot.
[19:30:57] <cmorley> surely mint and ubuntu wouldn't change the runtime name of the a common package...
[19:32:11] <cmorley> either way having 'glade-3' as the launch name seems weird
[20:06:22] <cmorley> maybe i should quit when behind - it seems 12.04 didn't include the option to have glade 3.10+ and 3.8 in parallel.
[20:08:23] <cmorley> so I guess 'glade-3' does make sense. arg Mint 13 (based on 12.04) does allow them both but then that means a launch name change.
[20:08:52] <cmorley> everyone should move to Mint anyways :)
[20:16:18] <cradek> oh 3.8 is the one that works! I misremembered what you said. So yes I confirm that you can get 3.8.
[20:16:30] <cradek> the install I tested this on is xubuntu 12.04
[20:17:51] <cradek> http://pastebin.com/raw.php?i=dLsUa8wd
[20:18:19] <cmorley> yes I see in ubuntu 12.10 they allowed both versions of glade
[20:18:29] <cmorley> yes 3.8 is good!
[20:18:44] <cradek> then I still don't understand the problem owhite has, do you?
[20:18:58] <cmorley> no I don't
[20:19:36] <cmorley> we can get him to look in a couple places to see whats up if he comes back
[20:19:44] <owhite> i be here
[20:20:00] <cradek> oh hey
[20:20:03] <owhite> what would you like me to test.
[20:20:09] <cradek> hey buddy what's your problem
[20:20:23] <owhite> Better grab a pen and paper.
[20:21:53] <cmorley> look in /usr/share/glade3/catalogs for hal_python.xml
[20:22:12] <owhite> checking...
[20:23:12] <owhite> I have: gtksourceview-glade.xml gtk+.xml hal_python.xml /usr/share/glade3/catalogs
[20:23:38] <owhite> I have: gtksourceview-glade.xml gtk+.xml hal_python.xml IN /usr/share/glade3/catalogs, that is.
[20:23:49] <cmorley> if you launch the glade editor do you see the HAL widgets?
[20:23:56] <owhite> no.
[20:24:36] <owhite> should I try setting some sort of environmental variable?
[20:24:41] <cmorley> open a terminal, launch glade and see if you can catch any info about it.
[20:24:48] <cmorley> no
[20:24:53] <owhite> coming up...
[20:25:24] <cmorley> there will be lots of debug warnings
[20:25:40] <cmorley> probably
[20:25:50] <owhite> not too much:
http://pastebin.ca/2571043
[20:26:38] <owhite> maybe I'll get more if I try to open one of the .ui files....hang on.
[20:26:42] <cmorley> k we may need to turn on debugging
[20:26:49] <cmorley> no not yet
[20:26:58] <owhite> ok.
[20:27:33] <owhite> when I run the editor, should an instance of linuxcnc be running?
[20:28:07] <cmorley> not necessary
[20:28:11] <owhite> ok
[20:28:41] <cmorley> try again in terminal with -v added
[20:28:52] <owhite> aye cap'n
[20:29:50] <owhite> wait - did you want the version?
[20:30:15] <cmorley> no verbose debugging
[20:30:37] <owhite> okay -v didnt create any additional output.
[20:30:50] <cmorley> ok to bad
[20:31:30] <cmorley> did you export an environment variable before?
[20:31:43] <cmorley> I seem to recall you did
[20:32:23] <owhite> yes. But hang on.
[20:32:29] <cmorley> ok
[20:32:39] <cmorley> hanging on :)
[20:33:29] <owhite> when I try to open a glade file in configs/sim.gmoccapy I am getting this: "But this version of Glade is for GTK+ 3 only. Make sure you can run this project with Glade 3.8 with no deprecated widgets first. Specially because there are 4 objects that can not be build with types HAL_Meter and HAL_LED"
[20:34:18] <owhite> and, it's throwing this: vcp_box.glade targets Gtk+ 2.16
[20:34:58] <cmorley> That warning is miss-leading. it assumes GTK3 but it's cause glade is not picking up the hal_python_xml file
[20:35:10] <owhite> ok.
[20:35:31] <cmorley> hal_python.xml defines gladevcp widgets.
[20:35:37] <owhite> so what should I try next? I could dig up the environmental variable I tried before.
[20:35:40] <cmorley> so that glade knows about them
[20:35:47] <owhite> ...or I could slap the screen upside the head.
[20:35:56] <cmorley> both...:)
[20:36:32] <cmorley> try exporting the catalog path to the actual catalog path...
[20:37:00] <owhite> yeah so I used this before :: export GLADE_CATALOG_PATH=/usr/lib/pymodules/python2.7/gladevcp
[20:37:03] <cmorley> I have never done that so outside my known territory
[20:37:08] <owhite> which is probably wrong, correct?
[20:37:13] <cmorley> ya thats not right.
[20:37:30] <owhite> okay, well hang on....
[20:40:49] <owhite> so you figure I should export GLADE_CATALOG_PATH=/usr/share/glade3/catalogs ??
[20:40:54] <owhite> 'snot working.
[20:41:55] <cmorley> I figured, yes
[20:42:33] <owhite> whoa whoa whoa.
[20:42:39] <cmorley> You actually shouldn't have to do that at all, as that is the standard place it looks
[20:43:20] <cmorley> whoa?
[20:43:32] <owhite> so I do not claim to understand any of the version stuff. But I also installed glade-3 at some point.
[20:43:40] <owhite> And it looks like that's getting the hal widgets.
[20:44:03] <cmorley> oh wow. you are using 12.04 yes?
[20:44:41] <owhite> and, it only works if I use that export. the version is...
[20:45:21] <owhite> glade3 3.8.0
[20:45:24] <owhite> go figure.
[20:46:35] <cmorley> it's a pain in the ass - the naming. glade used to be called glade3
[20:47:00] <owhite> seems pretty ridiculous.
[20:47:16] <cmorley> if you look in /usr/bin/
[20:47:36] <cmorley> see if there are more then one file name glade or glade something
[20:48:01] <owhite> glade glade-gtk2 glade-previewer gladevcp :-)
[20:48:34] <owhite> and glade-3 is in /usr/LOCAL/bin
[20:48:43] <owhite> (emphasis mine)
[20:49:39] <cmorley> that indicates (to me) you have two versions of glade, and as far as I can tell, 12.04 doesn't handle that.
[20:50:04] <cmorley> can you launch the 'glade' file and check it version in the about box
[20:50:18] <owhite> for which one? The one that works?
[20:50:52] <cmorley> there is usr/bin/glade and usr/bin/glade-gtk2
[20:51:00] <owhite> my glade-gtk2 works too.
[20:51:12] <cmorley> but are they both the same version
[20:51:42] <cmorley> glade-gtk2 will be glade 3.8
[20:51:54] <cmorley> not sure about plain glade
[20:52:51] <owhite> my glade-3 is also 3.8.0
[20:53:27] <cmorley> glade-3 is in usr/local/bin ? yes?
[20:53:51] <owhite> yes.
[20:54:06] <memleak> hello everyone
[20:54:08] <cmorley> what about usr/bin/glade ? 3.8?
[20:54:18] <memleak> $ which glade
[20:55:00] <cmorley> memleak: please be more specific
[20:55:14] <memleak> ^bash command
[20:55:42] <owhite> so /usr/bin/glade is... 3.12.1
[20:55:51] <cmorley> ahh ok there we go
[20:56:17] <cmorley> 3.12 is for gtk3 and doesn't work with linuxcnc.
[20:56:30] <owhite> ok, got it.
[20:56:53] <memleak> rm -rf /usr/bin/glade && ln -sfv /usr/local/bin/glade /usr/bin/glade ?
[20:57:02] <owhite> so it seems like we've solved the problem for my situation. Should I 'deactivate' the bug ticket?
[20:57:06] <cmorley> my very quick look on the internet says that 12.04 wasn't set up for having both versions.
[20:57:44] <cmorley> I will close the bug sure.
[20:58:15] <cmorley> 12.10 can have both versions and should not be a problem.
[20:58:20] <owhite> ok - if you can take care of it that works for me.
[20:58:35] <cmorley> I tried mint 13 with both versions and it worked.
[20:58:54] <cmorley> how will you fix your glade problem?
[20:58:56] <owhite> what I know about the difference between /usr/bin and /usr/local/bin and glade version names could fit in a bottlecap.
[20:59:29] <cmorley> lol i don't know much more - believe it or not
[20:59:41] <owhite> how will I ? Probably by just by deleting /usr/bin/glade
[21:00:39] <owhite> ...but I do appreciate all the time of yours this took.
[21:00:58] <cmorley> I was curious - glad I could help a bit anyways
[21:01:30] <owhite> you sure did - by the way have you used the editor? It doesnt look like that much fun.
[21:02:00] <cmorley> I use it lots. yes it takes some time to learn the little stuff.
[21:02:07] <cradek> what the heck installed a glade in /usr/local?
[21:02:25] <cradek> that's insanity
[21:02:40] <owhite> I just really like the look of the gmoccapy interface - and I think it would really useful to use it with a touch screen.
[21:02:44] <cmorley> what is usr/local actually for Chris?
[21:03:13] <cradek> it depends on the distribution, but in debian-based distros, packages aren't supposed to install there
[21:03:13] <cmorley> yes that was what it was made for mostly I think
[21:04:28] <cmorley> owhite: you don't need the glade editor to run gmoccapy - only to modify it
[21:04:46] <owhite> oh yeah I got that - I definitely would like to change it.
[21:04:54] <cmorley> oh ok I see
[21:05:18] <owhite> they're set up for machines unlike mine - the closest is the one for plasma cutters.
[21:05:22] <owhite> this is for a laser.
[21:06:12] <owhite> in the past I always relied on pyvcp - to make sliders for the power setting and monitoring data coming from the laser.
[21:12:56] <cmorley> I found glade much easier for anything beyond trivial panels (vrs pyvcp)
[21:13:51] <cmorley> it can get a little painful on really complicated panels.
[21:14:48] <owhite> I'll give it a shot.
[21:15:57] <cmorley> ya i'd suggest modifying the small sample panels, then build you own then try Gmoccapy.
[21:16:07] <cmorley> good luck!
[21:19:15] <cmorley> memleak: thanks for that - I am not versed so much on a lot of command line stuff - that's handy
[21:19:41] <memleak> i was hoping that'd help. i love the command line
[21:20:39] <mozmck> memleak: do you know anything about using 4 monitors on a recent Xorg with Nvidia drivers?
[21:21:15] <cradek> pretty sure the nvidia configurator will handle that fine
[21:21:21] <memleak> what he said^
[21:21:24] <cmorley> I use the command line everyday. i just never venture far from what i always use.
[21:21:40] <memleak> if not, xrandr
[21:21:52] <memleak> xrandr can get very lengthy sometimes
[21:22:03] <cradek> mozmck: do you have a more specific question?
[21:22:11] <mozmck> I just did that for my boss and looked like the only way to do it was set it up as two X screens with twinview
[21:22:32] <mozmck> I wanted one large screen ideally.
[21:22:51] <cradek> did you try turning on xinerama?
[21:22:52] <mozmck> Cinnamon also did not like the 2 X screens and crashes :)
[21:23:00] <mozmck> no, I didn't know what it was.
[21:23:33] <cradek> twinview is nvidia's own idea; xinerama is the usual and non-nvidia way of making one screen span devices
[21:23:38] <memleak> xrandr --output DVI0 --auto --output DVI1 --auto --right-of HDMI1 etc etc
[21:23:53] <cradek> memleak: sure but something persistent would be nice
[21:24:05] <mozmck> I see. does xrandr work with xinerama?
[21:24:07] <memleak> you can add it to your xorg start up script
[21:24:22] <mozmck> does xorg even have startup scripts anymore?
[21:24:22] <cradek> mozmck: yes I think so
[21:24:40] <cradek> yes you can still coerce it into running something in your home directory
[21:24:46] <memleak> yup^
[21:24:53] <cradek> it's not usual to do that anymore
[21:25:11] <memleak> or you can also add it to your DMEstartup script and have KDE or GNOME do it
[21:25:19] <memleak> *DM startup script
[21:25:23] <mozmck> I haven't seen an xorg.conf for a while.
[21:26:05] <cradek> the nvidia configurator writes a normalish xorg.conf
[21:26:07] <mozmck> ok, I'll do some more reading and playing. stuff I found after a little searching just left me confused :)
[21:26:13] <memleak> As of August 2013, Xinerama is broken with proprietary NVIDIA driver from 319 on.
[21:26:21] <cradek> yeah at least try xinerama
[21:26:21] <mozmck> bummer.
[21:26:23] <cradek> oh
[21:26:33] * cradek shrugs
[21:26:41] <memleak> works only with Linux kernels earlier than 3.10
[21:26:43] <cradek> try poking all the settings :-/
[21:28:02] <memleak> this might be of a little help:
https://devtalk.nvidia.com/default/topic/612517/?comment=3980519
[21:30:19] <memleak> i normally do all my multi monitor adjustments using the GUI based screen resolution tools but if that doesnt work i resort to xrandr
[21:30:49] <memleak> if that doesn't work for some reason then i start moving my monitors around to how linux wants them arranged (swapping DVI cables etc)
[21:31:42] <cradek> definitely hook them all up with digital (not vga)
[21:32:12] <memleak> heh i ran a dual monitor VGA set-up once, took me several hours
[21:32:25] <memleak> my xorg.conf file was nearly 2MB
[21:32:52] <memleak> i've read tom clancy novels that were shorter.
[21:54:25] <mozmck> thanks for the pointers.
[22:02:46] <memleak> hey mozmck just note i'm more of a radeon guy so im really not the one to ask about specific nouveau issues but anything X related in general and not a specific GPU problem i should be able to help out with. 2=< monitors are pretty much set up the same regardless of the graphics card though