IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 27 March 2008 (all times are UTC)
| 00:28 | rjune__ has joined #ltsp | |
| 00:29 | rjune_ has quit IRC | |
| 00:29 | rjune has quit IRC | |
| 00:29 | rjune_ has joined #ltsp | |
| 00:38 | chupa has joined #ltsp | |
| 00:40 | chup has quit IRC | |
| 01:27 | chup has joined #ltsp | |
| 01:27 | chupa has quit IRC | |
| 02:10 | chup has quit IRC | |
| 02:10 | chup has joined #ltsp | |
| 02:11 | deavid has joined #ltsp | |
| 02:20 | plamengr has joined #ltsp | |
| 02:31 | monteslu has quit IRC | |
| 02:40 | Pascal_1 has joined #ltsp | |
| 02:45 | <Pascal_1> salut
| |
| 02:57 | chupa has joined #ltsp | |
| 02:57 | chup has quit IRC | |
| 02:58 | plamengr has quit IRC | |
| 03:08 | <gvy> halo
| |
| 03:09 | monteslu has joined #ltsp | |
| 03:45 | Q-FUNK has joined #ltsp | |
| 03:46 | chup has joined #ltsp | |
| 03:46 | chupa has quit IRC | |
| 04:00 | bobby_C has joined #ltsp | |
| 04:10 | chup has quit IRC | |
| 04:10 | chup has joined #ltsp | |
| 04:18 | Pascal_1 has quit IRC | |
| 04:19 | Pascal_1 has joined #ltsp | |
| 04:36 | chup has quit IRC | |
| 04:36 | chup has joined #ltsp | |
| 04:38 | psycodad_ has quit IRC | |
| 05:10 | chup has quit IRC | |
| 05:40 | sep has quit IRC | |
| 06:04 | sep has joined #ltsp | |
| 06:22 | Q-FUNK has quit IRC | |
| 06:23 | Q-FUNK has joined #ltsp | |
| 06:24 | spectra has joined #ltsp | |
| 06:34 | subir has quit IRC | |
| 06:35 | ogra_cmpc_ has joined #ltsp | |
| 06:37 | ogra_cmpc has quit IRC | |
| 06:42 | vagrantc has joined #ltsp | |
| 06:55 | Guaraldo has joined #ltsp | |
| 06:58 | cliebow has joined #ltsp | |
| 06:59 | <daduke> vagrantc: howdy! I'm playing with debian live now...
| |
| 07:00 | <vagrantc> daduke: drop in irc.debian.org #debian-live too
| |
| 07:04 | jammcq has quit IRC | |
| 07:09 | <daduke> vagrantc: thanks, I'm in.
| |
| 07:16 | vagrantc: in my regular ltsp chroot, I have ltsp-client, ltsp-client-core, ltspfsd and ldm as ltsp-specific packages. Would you think that ldm is sufficient for LTSP operation provided that booting is performed by other means than NFS?
| |
| 07:19 | <ogra_cmpc_> daduke, as long as it knows the server ip it should connect to it wont have a prob
| |
| 07:19 | <vagrantc> daduke: i would actually install ltsp-client (which will pull in ltsp-client-core, ltspfsd and ldm)
| |
| 07:20 | daduke: but what ogra_cmpc_ said is true ... it should basically work.
| |
| 07:20 | <daduke> vagrantc, ogra_cmpc_ : I'm just worried that the NFS booting voodoo will mess up my CD boot..
| |
| 07:20 | <vagrantc> daduke: but if you want local devices and sound support and such, the easiest would be to install ltsp-client
| |
| 07:21 | <tarzeau> HELP!
| |
| 07:21 | <vagrantc> daduke: the only thing that you'll have to do is ensure that /etc/ltsp_chroot gets created before ltsp-client-core is installed.
| |
| 07:21 | tarzeau: see:
| |
| 07:21 | !question
| |
| 07:21 | <ltspbot> vagrantc: "question" is if you have a question about ltsp, please go ahead and ask it, and people will respond if they can. please also mention the linux distro and release you're using. :)
| |
| 07:21 | <tarzeau> vagrantc: i just said hello
| |
| 07:22 | * vagrantc compares "HELP!" to "hello" | |
| 07:22 | <daduke> vagrantc: I already argued with tarzeau....
| |
| 07:22 | * vagrantc doesn't notice a match | |
| 07:23 | <tarzeau> use -ignore-case and you find the first letter is equal
| |
| 07:23 | <daduke> vagrantc: your ltsp_chroot comment is something I will have to think about... not sure what you're trying to say yet.
| |
| 07:24 | <vagrantc> daduke: well, if the file /etc/ltsp_chroot doesn't exist, ltsp-client-core will fail to install. it's a check to make sure that ltsp-client-core doesn't get installed on a "regular" system, as it could do harmful things.
| |
| 07:24 | daduke: normally, ltsp-build-client handles that.
| |
| 07:25 | daduke: you'll have to figure out how to do that from debian-live. it's got a million hooks, so surely there's a way.
| |
| 07:26 | daduke: probably best to ask about that in #debian-live
| |
| 07:28 | <Pascal_1> klausade ??
| |
| 07:30 | K_O-Gnom has joined #ltsp | |
| 07:31 | otavio has joined #ltsp | |
| 07:32 | <tarzeau> vagrantc: in ldm how is the background color/picture defined?
| |
| 07:33 | vagrantc: it seems it takes bg.png for a tenths of a second, then makes it all white
| |
| 07:33 | i tried playing with greeter-gtkrc but no luck
| |
| 07:33 | <vagrantc> tarzeau: you're using the backports?
| |
| 07:33 | <tarzeau> we got a logo.png which is our own, a tux with a transparent background
| |
| 07:33 | ldm 0.1~bzr20080308-1~40.etch.0
| |
| 07:34 | <vagrantc> tarzeau: see http://wiki.debian.org/LTSP/Howto, look in the known issues section.
| |
| 07:34 | <ogra_cmpc_> tarzeau, there is a bg.png as well
| |
| 07:34 | avlis has joined #ltsp | |
| 07:34 | <tarzeau> i noticed the bg.png
| |
| 07:34 | does it matter if it's a file in place or a symlink?
| |
| 07:34 | <vagrantc> tarzeau: bg.png is broken for the etch backports. no idea how to fix it.
| |
| 07:34 | <tarzeau> we've got a symlinked bg.png it seems be loaded then overdrawn by solid white
| |
| 07:34 | i see
| |
| 07:34 | <ogra_cmpc_> it shouldnt (i never tested that though)
| |
| 07:35 | <tarzeau> i don't care about bg.png, but can i change the solid color somehow?
| |
| 07:35 | <ogra_cmpc_> not without patching the greeter
| |
| 07:35 | <tarzeau> i see
| |
| 07:35 | <vagrantc> and the interesting thing was, at one point i installed the exact same binary package on lenny and the bg.png worked fine.
| |
| 07:35 | <tarzeau> i heard a new version is coming soon (fixing the x forwarding stuff)
| |
| 07:35 | funny
| |
| 07:35 | <vagrantc> so it's a bug in one of the libraries on etch
| |
| 07:36 | <tarzeau> can you figure which?
| |
| 07:36 | the thing is, when i start ldm, i can see bg.png be drawn
| |
| 07:36 | but only for a little part of a second
| |
| 07:36 | <vagrantc> tarzeau: if i knew which, do you think i'd tell you? :P
| |
| 07:36 | <tarzeau> certainly not
| |
| 07:36 | <vagrantc> tarzeau: yeah, known issue.
| |
| 07:37 | <tarzeau> well ldd /usr/sbin/ldm doesn't list THAT many
| |
| 07:37 | <daduke> vagrantc: thanks, now I understand. the interesting thing is that ltsp-client-core installs in live just fine... I'll have to check that
| |
| 07:38 | <vagrantc> daduke: then ltsp-client-core is broken...
| |
| 07:40 | tarzeau: regarding a new version, i just uploaded a new version to sid yesterday. it fixes a big security bug when using LDM_DIRECTX (which was fixed in the version you're using when not using LDM_DIRECTX)
| |
| 07:40 | tarzeau: i'll see about a backport, but i'm going to be traveling a bit this next week.
| |
| 07:41 | although i'll actually be specifically working on ltsp for part of that time- not sure if i'll have a chance to test the backports.
| |
| 07:43 | <tarzeau> vagrantc: ah cool, i'll make my own backport:
| |
| 07:43 | <vagrantc> tarzeau: there's some patches specific to the backport you may need to apply and update.
| |
| 07:43 | <tarzeau> which?
| |
| 07:43 | it's building fine
| |
| 07:45 | http://krum.ethz.ch/backports/ldm/
| |
| 07:45 | <daduke> tarzeau: do we really want to do this ourselves? We can wait some days no prob if vagrantc is working on it anyway...
| |
| 07:45 | <vagrantc> tarzeau: have you installed it?
| |
| 07:45 | <tarzeau> daduke: debian makes no backport, you never have a warranty that there will be a backport
| |
| 07:45 | vagrantc: i need to build it for i386 first, then test it
| |
| 07:45 | <ogra_cmpc_> tarzeau, daduke ... in any case send us your patches :)
| |
| 07:45 | <daduke> tarzeau: vagrantc's my warranty.
| |
| 07:46 | <vagrantc> i pretty much always make backports if there's a reason to. i intend to make this backport.
| |
| 07:46 | slidesinger has joined #ltsp | |
| 07:46 | <daduke> tarzeau: my point is: testing == production in our current setup.
| |
| 07:47 | <vagrantc> tarzeau: the debian dir is in a bzr branch: http://llama.freegeek.org/~vagrant/bzr/ltsp/ltsp-etch-backports-packaging
| |
| 07:47 | <tarzeau> vagrantc: that url is empty?
| |
| 07:47 | <ogra_cmpc_> thats a bzr branch
| |
| 07:48 | use it with bzr :)
| |
| 07:48 | <vagrantc> tarzeau: and you'll need to merge patches from: http://bzr.debian.org/bzr/pkg-ltsp/ldm/ldm-debian-packaging
| |
| 07:48 | and ogra explained the rudiments.. :)
| |
| 07:49 | tarzeau: or if you build it without the patches and it works, so much the better :)
| |
| 07:50 | tarzeau: for some reason neither RyanRyan52 or i could figure out, we had to move some code from a function into the main() in order to get it to work.
| |
| 07:51 | <tarzeau> vagrantc: the backport just worked
| |
| 07:51 | <vagrantc> tarzeau: ah, maybe it just broke autologin and/or guest login, then.
| |
| 07:51 | <tarzeau> yes maybe, i haven't tested autologin/guest logins
| |
| 07:51 | <vagrantc> ah, my changelog entry was: * apply patch from Ryan Niebur to fix autologin and guest login on etch
| |
| 07:52 | if you don't use them, then you're probably fine.
| |
| 07:54 | <tarzeau> i think we use the autologin for some kiosks
| |
| 07:55 | <daduke> tarzeau: indeed we do
| |
| 07:55 | <vagrantc> i'll see if i can't get a backport built and tested today.
| |
| 07:55 | <ogra_cmpc_> autologin is slightly crak still anyway ... until we add a proper delay
| |
| 07:55 | <vagrantc> ogra_cmpc_: if you need it, you need it.
| |
| 07:56 | <ogra_cmpc_> you cant cleanly log out/shut down with it
| |
| 07:56 | vagrantc, indeed, just ointing out the drawbacks
| |
| 07:56 | *pointing
| |
| 07:56 | <vagrantc> ogra_cmpc_: although the same patch is needed to get guest login to work, which doesn't have that problem.
| |
| 07:57 | hmmm...
| |
| 07:57 | <ogra_cmpc_> it likely suffices to have a wait loop in there
| |
| 07:57 | <vagrantc> if we basically had a timeout for the guest login, it would obsolete autologin
| |
| 07:57 | <ogra_cmpc_> probably with a configurable value
| |
| 07:57 | <vagrantc> well, the autologin code doesn't even display the greeter
| |
| 07:58 | <ogra_cmpc_> for timeouts use timeoutd :)
| |
| 07:58 | <vagrantc> so we'd have to muck around with that a bit.
| |
| 07:58 | mhterres has joined #ltsp | |
| 07:59 | <vagrantc> oops. i've never published my ldm-etch-backports branch ...
| |
| 07:59 | <ogra_cmpc_> i have a littel few line app from linus that logs you out after a configurable time ... he sent it to me for edubuntu but didnt bother to give it any license
| |
| 07:59 | <vagrantc> tarzeau: gave you the wrong URL ...
| |
| 08:03 | mhterres has quit IRC | |
| 08:03 | mhterres has joined #ltsp | |
| 08:04 | <vagrantc> tarzeau: the correct URL is: http://llama.freegeek.org/~vagrant/bzr/ltsp/ldm-etch-backports-packaging
| |
| 08:05 | Faithful has joined #ltsp | |
| 08:06 | tux_440volt has joined #ltsp | |
| 08:13 | mccann has joined #ltsp | |
| 08:13 | jammcq has joined #ltsp | |
| 08:13 | * tarzeau looks at the ldm/src now | |
| 08:13 | <jammcq> g'morning friends
| |
| 08:15 | <tarzeau> i still don't get who/where ldm is launched ?
| |
| 08:16 | and what's most annoying is that a wrong password doesn't allow you another try
| |
| 08:16 | and that ldm isn't restarted when killed after such a bad entry
| |
| 08:16 | <ogra_cmpc_> tarzeau, its started by the ltsp-client-core initscript
| |
| 08:17 | patches for the password stuff would be really appreciated
| |
| 08:18 | it did allow to re-enter the pw three times but the implementation ws a bit weird so its disabled by a one line patch until we fixed it properly
| |
| 08:20 | tarzeau, bzr diff -r807..808 on the ldm source has the change
| |
| 08:20 | <tarzeau> the ltsp-client-core starts screen_session (looking at that now)
| |
| 08:20 | ogra_cmpc_: thanks but i don't use bzr
| |
| 08:20 | <ogra_cmpc_> err
| |
| 08:20 | how do you get the source code than from vagrants branches ??
| |
| 08:20 | *then
| |
| 08:21 | <tarzeau> i don't
| |
| 08:21 | Q-FUNK has quit IRC | |
| 08:21 | Q-FUNK has joined #ltsp | |
| 08:21 | <tarzeau> i only got the source debian package from sid
| |
| 08:21 | <ogra_cmpc_> ah well
| |
| 08:22 | in any case the change is in revision 808 in case you want to revert it ... its searchable in the branches web interface on launchpad in the upstream branch
| |
| 08:24 | Gadi has joined #ltsp | |
| 08:27 | ogra_cmpc_ is now known as ogra_cmpc | |
| 08:34 | <tarzeau> seems the background stuff is gtk related
| |
| 08:39 | elisboa has joined #ltsp | |
| 08:40 | <ogra_cmpc> tarzeau, which gtk version does your chroot have ?
| |
| 08:40 | <vagrantc> tarzeau: if you come up with a patch that either works in both newer and older versions of gtk (and isn't too ugly), i can hopefully include it upstream, or a patch that just works on etch, i can make a backport with it.
| |
| 08:42 | <ogra_cmpc> gdk_pixbuf_new_from_file_at_scale is only supported from gtk 2.6 on
| |
| 08:44 | <tarzeau> ogra_cmpc: i've got gtk 2.8.20(-7)
| |
| 08:44 | ogra_cmpc: it's supported, but for some reason fails, checking the gnome docs...
| |
| 08:44 | <ogra_cmpc> that should suffice
| |
| 08:44 | <tarzeau> gtk is not my favourite graphics toolkit
| |
| 08:44 | the code is ugly
| |
| 08:49 | <ogra_cmpc> thanks
| |
| 08:49 | whats ugly about it (so i can prevent uglyness in teh future)
| |
| 08:50 | <tarzeau> ogra_cmpc: it's just a paradigm "do the gui stuff graphically", if you know cocoa...
| |
| 08:50 | nothing you can do something about
| |
| 08:50 | <vagrantc> you know, some code is just born ugly.
| |
| 08:50 | <ogra_cmpc> oh, you mean gtk in general
| |
| 08:50 | <tarzeau> yes
| |
| 08:51 | <ogra_cmpc> i thoguht you mean the greeter specifically :)
| |
| 08:51 | <tarzeau> i can't edit the gui elements in a sane way
| |
| 08:51 | <ogra_cmpc> depends what you are used to
| |
| 08:51 | i like programming gtk
| |
| 08:51 | we used to use glade
| |
| 08:52 | <vagrantc> ogra_cmpc: was glade a big part of why it was *so slow* ?
| |
| 08:52 | <ogra_cmpc> but that proved to be to heavyweight, the code tries to use the bare minimum of gtk
| |
| 08:52 | <tarzeau> well the background picture problem is somewhere in gtkgreet/greeter.c
| |
| 08:53 | <ogra_cmpc> vagrantc, part of it ... glade itself was more about ram usage though
| |
| 08:53 | <vagrantc> ogra_cmpc: ah.
| |
| 08:53 | tarzeau: thanks for digging into this issue :)
| |
| 08:54 | <tarzeau> i want to see if i can figure where the authentication happens
| |
| 08:54 | <ltsppbot> "ogra" pasted "your prob is somewhere here" (9 lines) at http://pastebot.ltsp.org/486
| |
| 08:54 | <ogra_cmpc> thats the bit responsible for the bg.png
| |
| 08:55 | unless ldm_theme_file is doing something wrong with assembling the right path
| |
| 08:55 | <tarzeau> ogra_cmpc: that was where i kept looking at
| |
| 08:55 | i didn't dig into ldm_theme_file since it works for logo.png
| |
| 08:55 | <vagrantc> since it actually displays, but then gets overwritten by the white, and the same binary worked on etch and lenny, i was wondering if it was a change in the order that gtk does things
| |
| 08:55 | <ogra_cmpc> it does for all other files
| |
| 08:56 | <vagrantc> ogra_cmpc: it actually displays the bg.png, it just gets overwritten by something else
| |
| 08:56 | <ogra_cmpc> gdk_pixbuf_unref(rawpic);
| |
| 08:56 | try commenting that line
| |
| 08:56 | <tarzeau> the gnome docs said something about deprecated for new code on the gdk_pixbuf_unref
| |
| 08:57 | <ogra_cmpc> well, it works fine in hardy
| |
| 08:57 | which should have the very latest gtk
| |
| 08:59 | <tarzeau> ogra_cmpc: commenting it out had no effect
| |
| 08:59 | <ogra_cmpc> right
| |
| 09:00 | sonjag has joined #ltsp | |
| 09:00 | <tarzeau> it probably just frees up some memory?
| |
| 09:01 | tux_440volt has quit IRC | |
| 09:02 | <ogra_cmpc> hmm
| |
| 09:02 | i dont have the theme patches in ubuntu i just noticed
| |
| 09:02 | so its likely introduced by the patch in revision 806
| |
| 09:03 | <tarzeau> so you can fix it easily?
| |
| 09:03 | <vagrantc> ogra_cmpc: it's been an issue long before that, though.
| |
| 09:03 | <ogra_cmpc> no idea its no code i use in ubuntu
| |
| 09:03 | <vagrantc> ogra_cmpc: it's been an issue since the first backport of C ldm.
| |
| 09:03 | <ogra_cmpc> vagrantc, well, i dont see it on any ubuntu release
| |
| 09:04 | <vagrantc> ogra_cmpc: yes, it's only present on etch.
| |
| 09:04 | ogra_cmpc: same code on etch works fine on lenny.
| |
| 09:04 | <ogra_cmpc> so its likely a version mistmatch or a missing dep
| |
| 09:04 | <vagrantc> ogra_cmpc: or sid.
| |
| 09:04 | <ogra_cmpc> hmm
| |
| 09:04 | weird
| |
| 09:05 | <vagrantc> ogra_cmpc: and at one point, the same exact *binary* worked on lenny but not etch.
| |
| 09:06 | so it's not necessarily a problem in the code... but how the code interacts with different versions of depending libraries, or maybe a missing dependency ...
| |
| 09:07 | but i would think a missing dependency would cause it to fail entirely...
| |
| 09:08 | <ogra_cmpc> it wouldnt display the pic first place
| |
| 09:09 | <vagrantc> but it actually displays the picture for a split second... something else ends up writing over it ...
| |
| 09:09 | <ogra_cmpc> right
| |
| 09:10 | comment everything after gdk_window_set_back_pixmap((GdkWindow *) window->window, pic, 0);
| |
| 09:10 | until right before gtk_widget_show_all(window);
| |
| 09:10 | and then move the top comment sign down UI element by UI element
| |
| 09:11 | <vagrantc> right ...
| |
| 09:11 | <ogra_cmpc> that should show you what overwrites it
| |
| 09:15 | * gvy is back (gone 04:00:23) | |
| 09:15 | <gvy> ogra_cmpc, vagrantc, ! :)
| |
| 09:15 | <ogra_cmpc> hey
| |
| 09:22 | gonzaloaf has quit IRC | |
| 09:25 | mccann has quit IRC | |
| 09:52 | Q-FUNK has quit IRC | |
| 09:52 | Q-FUNK has joined #ltsp | |
| 10:00 | Skarmeth has joined #ltsp | |
| 10:16 | steph_ has joined #ltsp | |
| 10:17 | <cyberorg> hi guys http://en.opensuse.org/Easy-LTSP#Enhancement_requests fill it up :)
| |
| 10:19 | <gvy> hi cyberorg
| |
| 10:19 | <cyberorg> hiya gvy :)
| |
| 10:30 | staffencasa has joined #ltsp | |
| 10:31 | steph_ has quit IRC | |
| 10:32 | prpplague has joined #ltsp | |
| 10:32 | <prpplague> jammcq: greetings
| |
| 10:32 | jammcq: got a couple questions for you when you a minute
| |
| 10:33 | mccann has joined #ltsp | |
| 10:37 | K_O-Gnom has quit IRC | |
| 10:41 | gentgeen__ has joined #ltsp | |
| 10:42 | <jammcq> prpplague: hey
| |
| 10:42 | <prpplague> jammcq: hey, i'm tinkering with one of our small board designs
| |
| 10:43 | jammcq: http://www.elinux.org/Nail_Board
| |
| 10:43 | <jammcq> cool
| |
| 10:43 | <prpplague> jammcq: if you had something with similiar size and shape, but two usb hosts on one end, and vga out on the other, would that be viable for a small thin client in your mind?
| |
| 10:44 | <jammcq> well, if it were in a small box that could be stuck to the back of a monitor, that'd be pretty cool
| |
| 10:44 | what about ethernet?
| |
| 10:45 | <prpplague> jammcq: we have two options, we could add the ethernet internally, or do something similiar to the AMD PIC device and use a usb->ethernet adapter
| |
| 10:45 | <jammcq> well, if you can avoid a bunch of adapters, that'd be best
| |
| 10:46 | <prpplague> jammcq: ahh
| |
| 10:46 | jammcq: we'll we can add 10baseT for really cheap, 100baseT adds another few dollars
| |
| 10:46 | <jammcq> but you could do mouse and keyboard on usb
| |
| 10:46 | ethernet port
| |
| 10:46 | how much ram ?
| |
| 10:47 | <prpplague> jammcq: the base model we are testing with is very low resource, 32mb ram and 16mb flash, but we'd have a larger model for production with 128mb ram and 128mb flash
| |
| 10:47 | <jammcq> and if yer gonna do an ethernet port, it should have PXE or Etherboot built-in
| |
| 10:47 | flash wouldn't be necessary for LTSP
| |
| 10:48 | <prpplague> jammcq: well PXE and etherboot are part of the code for the device, not really in the ethernet(atleast in embedded systems)
| |
| 10:48 | <jammcq> what about power supply?
| |
| 10:48 | where's the code? in flash?
| |
| 10:48 | <prpplague> jammcq: yea in the flash
| |
| 10:48 | <jammcq> i've seen PXE in the bios
| |
| 10:48 | what's the CPU ?
| |
| 10:48 | <prpplague> jammcq: equiv to bios on x86
| |
| 10:48 | jammcq: s3c2410 arm920t
| |
| 10:49 | <jammcq> you love those strongarms, eh?
| |
| 10:49 | <prpplague> jammcq: power supply is just a standard +5v wall wort
| |
| 10:49 | jammcq: hehe, actually this arm9 not strongarm
| |
| 10:49 | jammcq: but yea i love arm, it pays the bills
| |
| 10:49 | * jammcq doesn't know the difference | |
| 10:49 | <ogra_cmpc> sad ... none of the ltsp5 distros has arm support
| |
| 10:50 | <vagrantc> not entirely true
| |
| 10:50 | <jammcq> hmm
| |
| 10:50 | <prpplague> jammcq: its like comparing AMD products to Intel
| |
| 10:50 | <vagrantc> i just added patches for arm on debian
| |
| 10:50 | but i don't have any hardware to test
| |
| 10:50 | <ogra_cmpc> vagrantc, your ltsp5 packages run on arm ?
| |
| 10:50 | <vagrantc> ogra_cmpc: in theory
| |
| 10:50 | <ogra_cmpc> right
| |
| 10:50 | <prpplague> jammcq: i.e. difference between saying x86 and pentium
| |
| 10:50 | <jammcq> vagrantc: hmmm, maybe prpplague can get you some arm hardware for testing
| |
| 10:50 | * ogra_cmpc has two old ipaqs lying around here ... | |
| 10:51 | <vagrantc> ogra_cmpc: i wrote some network boot code based on the code in debian-installer
| |
| 10:51 | <ogra_cmpc> but no netboot opportunoty for them
| |
| 10:51 | * gvy . o O ( r701 ) | |
| 10:51 | <prpplague> ogra_cmpc: yea it works on arm, i did several arm installations with ltsp and ltsp/hybrid stuff, but thats been years ago
| |
| 10:52 | <ogra_cmpc> prpplague, ltsp5 uses the host systems kernel ... ubuntu doesnt support arm at all, debian has arm support but nobody tested ltsp ever on arm there
| |
| 10:52 | not sure about fedora which is supposed to have an ltsp5 implementation with next months release
| |
| 10:53 | but i dont think warren took other areches than i386 and amd64 into account
| |
| 10:53 | <prpplague> ogra_cmpc: yea, you'd need to generate a seperate kernel and rootfs specific for the arm devices
| |
| 10:53 | bobby_C has quit IRC | |
| 10:53 | <ogra_cmpc> which is fine and exactly what we do anyway in ltsp
| |
| 10:53 | <prpplague> ogra_cmpc: i never used ltsp's system for creating the rootfs anyway
| |
| 10:54 | <warren> I added code so other arches have a chance of working on fedora
| |
| 10:54 | <ogra_cmpc> but you need something you can modify ... ubuntu doesnt have it so its a no go for me
| |
| 10:54 | (we might go for arm in teh next release)
| |
| 10:54 | not sure though
| |
| 10:54 | <prpplague> so the base ltsp would be problematic for standized usage
| |
| 10:54 | hmm
| |
| 10:55 | <vagrantc> prpplague: why would it be problematic?
| |
| 10:56 | arm is supported on debian. therefore, arm is (theoretically) supported in ltsp5 in debian.
| |
| 10:56 | doing cross-architecture install is the tricky part, i guess.
| |
| 10:56 | but i've had decent luck with qemu-system-arm
| |
| 10:59 | <prpplague> vagrantc: just stating that since ogra_cmpc seems to think ubuntu doesn't have arm support, that not everyone would be able to use the standard ltsp
| |
| 11:00 | vagrantc: not saying that the support to couldn't be tweaked and fix
| |
| 11:00 | <vagrantc> prpplague: on ubuntu ...
| |
| 11:00 | <prpplague> vagrantc: correct
| |
| 11:00 | <ogra_cmpc> ltsp5 is deitro specific
| |
| 11:00 | *distro
| |
| 11:01 | <vagrantc> well, i'd be curious about testing it for debian.
| |
| 11:01 | <ogra_cmpc> if you find distros using ltsp5 and having arm support, all is fine
| |
| 11:01 | <vagrantc> and if debian gets it working, then if ubuntu ever adds support for arm, it'll basically work for ubuntu without many, if any, changes.
| |
| 11:01 | <prpplague> yea, we have some prototype designs we were looking at, trying to decide if it would be a practical device to put into production
| |
| 11:02 | Pascal_1 has quit IRC | |
| 11:02 | <warren> arm based thin clients?
| |
| 11:02 | <prpplague> warren: yea its a small arm based device, not nessarily a thin client
| |
| 11:03 | <vagrantc> i thought otavio had an arm thin client at some point
| |
| 11:03 | <prpplague> its 5" x 1"
| |
| 11:03 | basically an oversized thumb drive
| |
| 11:03 | <mnemoc> did Ubuntu get sample fulong-2F boxes?
| |
| 11:04 | <prpplague> jammcq: think there would be any interest in using it as a thin client?
| |
| 11:05 | <vagrantc> the keychain thin-client...
| |
| 11:07 | <ogra_cmpc> teah, carry a usb disk and a hacked LCD photo frame with you ... can replace your laptop
| |
| 11:07 | <laga> yeah, also make sure to bring your server ;)
| |
| 11:07 | <ogra_cmpc> pocket laptop kit ;)
| |
| 11:09 | * ogra_cmpc is waiting for a USB key sized mini projector that works at daylight and produces a 14" screen since years | |
| 11:09 | <ogra_cmpc> but these slackers simply dont invent it ... damned
| |
| 11:11 | <warren> ogra_cmpc, its in the movie Paycheck
| |
| 11:11 | <ogra_cmpc> heh
| |
| 11:11 | <jammcq> prpplague: well, for it to work as an LTSP workstation, it needs to be supported by the major distros. that's probably an issue, at least for fedora and ubuntu. prolly suse too
| |
| 11:11 | debian would probably work, and dunno about gentoo and the others
| |
| 11:12 | <ogra_cmpc> gentoo would ... if their ltsp was done
| |
| 11:12 | there is movement but its still early
| |
| 11:13 | <prpplague> jammcq: fedora has arm support with no problems, i'd have to look into ubuntu
| |
| 11:13 | jammcq: yea debian has arm support as well
| |
| 11:13 | <jammcq> don't need to look much further than ogra_cmpc for questions about Ubuntu support
| |
| 11:13 | <ogra_cmpc> prpplague, there is nothing to look into on ubuntu ... the arch simply isnt available
| |
| 11:13 | <prpplague> jammcq: we are talking something like $189 for the board
| |
| 11:14 | <jammcq> hmm, seems expensive
| |
| 11:14 | <ogra_cmpc> yeah
| |
| 11:14 | <prpplague> jammcq: we did the original design for a company that wanted to replace a bunch of old wyse 50 devices
| |
| 11:14 | <jammcq> there's $100 terminals out there
| |
| 11:14 | what's the performance like on that arm chip?
| |
| 11:15 | <prpplague> jammcq: around 100 bogomips, no where near what you are probably used to on a pc
| |
| 11:15 | <ogra_cmpc> that wont be fun on ltsp5
| |
| 11:15 | <jammcq> and what video chip would you use?
| |
| 11:15 | <prpplague> jammcq: which is kind of why i was wondering about the practicality of using it with ltsp
| |
| 11:15 | <jammcq> it prolly would have worked ok on ltsp-4.2, but that is pretty dead
| |
| 11:15 | <prpplague> jammcq: the s3c2410 has it's own video controller, as do most arm soc's
| |
| 11:16 | <jammcq> is there Xorg support for that chip?
| |
| 11:16 | <prpplague> jammcq: yea
| |
| 11:16 | jammcq: i already run x on it
| |
| 11:17 | jammcq: are the $100 terminals actually available these days?
| |
| 11:17 | <gvy> prpplague, ex-amd pic?
| |
| 11:17 | <vagrantc> ogra_cmpc: well ... i made it as far as UserPrompt = gtk_label_new("");
| |
| 11:17 | <prpplague> jammcq: i've seen alot of stuff on linuxdevices about them, but i didn't think they were available
| |
| 11:17 | <vagrantc> ogra_cmpc: still nothing overwriting bg.png
| |
| 11:17 | <jammcq> prpplague: DisklessWorkstations has sold a bunch of them
| |
| 11:17 | and has them in stock afaik
| |
| 11:18 | <vagrantc> ogra_cmpc: don't know what in that jumble would be meaningful ...
| |
| 11:18 | <prpplague> gvy: i didn't think that the company that purchase the pic design from amd was still selling them
| |
| 11:18 | <ogra_cmpc> vagrantc, so you have it properly sitting in the background ?
| |
| 11:18 | <prpplague> jammcq: ahh i'll have to have a look
| |
| 11:18 | <vagrantc> ogra_cmpc: yeah... displaying nothing but the background ...
| |
| 11:18 | ogra_cmpc: i don't really understand gtk well enough to guess which lines are meaningful ...
| |
| 11:18 | ogra_cmpc: testing it one line at a time is a little tedious. :)
| |
| 11:19 | <prpplague> jammcq: this design had some special considerations, which is why they didn't use an off the shelf x86 terminal
| |
| 11:19 | <gvy> prpplague, they seemed to do a few months ago
| |
| 11:19 | * prpplague wasn't too impressed with the PIC design | |
| 11:20 | <ogra_cmpc> vagrantc, yeah, the intresting pieces are the gtk_box_pack_start lines which actually add the elements
| |
| 11:20 | <gvy> prpplague, https://store.dataevolution.com/ProductDetails.asp?ProductCode=DT-7001
| |
| 11:20 | <vagrantc> ogra_cmpc: ah, cool.
| |
| 11:20 | <prpplague> jammcq: big difference between this design and the x86 stuff is the power consumption, which was a priority for the customer
| |
| 11:21 | alekibango has quit IRC | |
| 11:22 | <prpplague> gvy: ahh interesting
| |
| 11:22 | gvy: i wonder if they are just selling off stock or are they still in actual production
| |
| 11:23 | <ogra_cmpc> thats pretty old ... likely stock only
| |
| 11:23 | i suspect
| |
| 11:23 | <prpplague> gvy: as cable&wireless had originally planned to order atleast 50k units, and ended up only purchasing around 2k units
| |
| 11:24 | <ogra_cmpc> and they are something like 3 years old already i think
| |
| 11:24 | <prpplague> ogra_cmpc: actually closer to 4
| |
| 11:25 | <ogra_cmpc> yup
| |
| 11:27 | <gvy> prpplague, guess the initial stock would make no sense a year later
| |
| 11:27 | don't know
| |
| 11:27 | (readin up) ah, even so?
| |
| 11:27 | funny...
| |
| 11:28 | <prpplague> well, we already got paid for the design, i was just curious about using it with ltsp and such so not that big of a deal
| |
| 11:28 | <ogra_cmpc> if you ever have lpia devices i'll be your man with ubuntu :)
| |
| 11:28 | indradg has joined #ltsp | |
| 11:28 | <prpplague> lpia?
| |
| 11:28 | <ogra_cmpc> menlo
| |
| 11:29 | intels new mobile CPU
| |
| 11:29 | *menlow
| |
| 11:29 | <prpplague> ahh
| |
| 11:29 | <lns> is that the new geode competitor?
| |
| 11:29 | <prpplague> not likely
| |
| 11:29 | <ogra_cmpc> thats somnething ubuntu supports :)
| |
| 11:29 | <lns> that all-in-one integrated chipset
| |
| 11:30 | <prpplague> ogra_cmpc: we generally do only arm, mips, or ppc, i can't even remember the last time we did an x86 design
| |
| 11:30 | <ogra_cmpc> no idea i havent had one in myu hands yet (i have seen a fridge sized test setup though ... not very "mobile" :) )
| |
| 11:30 | <r3zon8> i used the debian tarball for ltsp5 to install on RHEL5, and ive got everything working fine now..its just that the login screen shows Debian logos/etc...and i rather it show RH logos..any ideas?
| |
| 11:30 | <lns> ha
| |
| 11:30 | mikkel has joined #ltsp | |
| 11:31 | <ogra_cmpc> prpplague, sad, none of the arches ubuntu supports
| |
| 11:31 | ppc got dropped two releases ago ... this round we got rid of sparc
| |
| 11:31 | <laga> i thought sparc packages are still available and built, but not officially supported anymore?
| |
| 11:32 | <ogra_cmpc> right
| |
| 11:32 | like ppc
| |
| 11:32 | <laga> ah
| |
| 11:32 | <ogra_cmpc> but sionce thats community maintained you cant predict if i9t will b e like thatr next release ... or if anyone will fix your bugs
| |
| 11:32 | (ppc community is big, but sparc is rather small)
| |
| 11:34 | tux_440volt has joined #ltsp | |
| 11:45 | * mnemoc has huge hopes on lemote mips boxes | |
| 11:53 | alekibango has joined #ltsp | |
| 11:57 | makghosh has joined #ltsp | |
| 11:59 | mccann_ has joined #ltsp | |
| 12:08 | indradg has quit IRC | |
| 12:09 | indradg has joined #ltsp | |
| 12:10 | DonSilver has joined #ltsp | |
| 12:10 | mccann has quit IRC | |
| 12:27 | TelnetManta has joined #ltsp | |
| 12:30 | elisboa has quit IRC | |
| 12:32 | elisboa has joined #ltsp | |
| 12:35 | gonzaloaf_test has joined #ltsp | |
| 12:55 | MacIver has quit IRC | |
| 13:03 | DonSilver has quit IRC | |
| 13:03 | DonSilver has joined #ltsp | |
| 13:06 | mccann_ has quit IRC | |
| 13:10 | mccann has joined #ltsp | |
| 13:16 | deavid has quit IRC | |
| 13:24 | avlis has quit IRC | |
| 13:31 | <vagrantc> well, tested the new ldm backport
| |
| 13:32 | prpplague has left #ltsp | |
| 13:33 | <cyberorg> hi vagrantc is suse plugin pushed?
| |
| 13:34 | <vagrantc> i haven't pushed anything ...
| |
| 13:34 | last i recall, we had discussed all sorts of things, the exact details i don't remember ...
| |
| 13:35 | i've gotta run at the moment ...
| |
| 13:35 | <cyberorg> vagrantc, you wanted bzr tree with individual commits :)
| |
| 13:35 | you got that
| |
| 13:35 | <vagrantc> cyberorg: yes.
| |
| 13:36 | <cyberorg> http://cyberorg.kicks-ass.org/ltsp/ltsp-trunk/ that was it
| |
| 13:36 | <warren> Wow, pack-0.92 is way faster
| |
| 13:36 | <vagrantc> warren: yeah, seems to be.
| |
| 13:38 | cyberorg: looks like it's been merged.
| |
| 13:38 | cyberorg: er ...
| |
| 13:39 | <cyberorg> vagrantc, cool, i'll pull and add more stuff
| |
| 13:39 | <vagrantc> cyberorg: actually ... bzr: ERROR: Connection error: curl connection error (Couldn't resolve host 'cyberorg.kicks-ass.org')
| |
| 13:39 | cyberorg: i just have revision 677 ...
| |
| 13:39 | cyberorg: which just looks like a mirror of ltsp-trunk
| |
| 13:40 | cyberorg: sorry, got to run now ...
| |
| 13:40 | vagrantc has quit IRC | |
| 13:40 | <cyberorg> ok, ciao
| |
| 13:44 | deavid has joined #ltsp | |
| 13:47 | <cyberorg> i didn't know bzr needed push, thought commit should take care of it :(
| |
| 13:49 | <gvy> cyberorg, warren, again just in case: r702 was pushed to http://ftp.linux.kiev.ua/pub/Linux/ALT/people/led/bzr/ltsp/ltsp-altlinux/
| |
| 13:49 | <warren> gvy, I am not looking at it for at least 2 weeks
| |
| 13:49 | <gvy> that seems to be all of the main part, although there are some more cleanups in the pipe
| |
| 13:49 | warren, i remember :)
| |
| 13:50 | warren, btw who merges into ltsp-trunk?
| |
| 13:50 | <cyberorg> gvy, we need to ask for commit access :)
| |
| 13:50 | <gvy> cyberorg, well i'd better have some more training time (led probably too) :)
| |
| 13:50 | botching up a shared repo is not the best thing to do
| |
| 13:51 | <cyberorg> gvy, don't worry, nothing can go wrong that cannot be reverted with source control
| |
| 13:51 | <gvy> warren, does vagrantc merge into trunk?
| |
| 13:51 | cyberorg, well that can be messy too
| |
| 13:51 | <cyberorg> gvy, there are projects with 100s of devs pushing all the time
| |
| 13:52 | * gvy pushes all the time, *but still* | |
| 14:19 | Q-FUNK has quit IRC | |
| 14:22 | sonjag has quit IRC | |
| 14:23 | * gvy is away: ~ | |
| 14:23 | <gvy> bb folks
| |
| 14:35 | Skarmeth has quit IRC | |
| 14:41 | Skarmeth has joined #ltsp | |
| 14:52 | praveer_cool has quit IRC | |
| 15:01 | Gadi has left #ltsp | |
| 15:02 | bobby_C has joined #ltsp | |
| 15:09 | Sockra_Tease has joined #ltsp | |
| 15:13 | <cliebow> kinda quiet at the moment..lot of the Europeans are peob resting
| |
| 15:15 | <Sockra_Tease> if I compile a binary for slackware 12 for LTSP would you guys want to use it/store it?
| |
| 15:15 | Pascal_1 has joined #ltsp | |
| 15:16 | <Sockra_Tease> i know most projects dont like to hold onto bin packages for slackware due to the nature of the distro, so offer stands
| |
| 15:16 | <cliebow> there have been a couple guys working on ltsp5 in slack..jammcq would know..you need to read up on ltsp..
| |
| 15:16 | <Sockra_Tease> doing so atm
| |
| 15:16 | <laga> yes, read up on "project muekow".. binary packages probably won't do much good
| |
| 15:18 | <Sockra_Tease> muekow like moocow? #moocow?
| |
| 15:18 | <cliebow> Sockra_Tease, quite a few distros are working together to create a common codebase for this thin client solution..
| |
| 15:19 | tux_440volt has quit IRC | |
| 15:19 | <Sockra_Tease> why would the distros be responsible for that?
| |
| 15:19 | rather the maintainers
| |
| 15:19 | <laga> Sockra_Tease: www.ltsp.org and read up on muekow
| |
| 15:19 | <Sockra_Tease> ok ok
| |
| 15:21 | <cliebow> Sockra_Tease, once upon a time we created a chroot from redhat7.3 binaries..it was crazy trying to support it..so jammcq went to creating an lbe--linuxbuild environment so all glibcs and stuff matched..ltsp5 is step forward in that each distro provides its own binaries and ltsp just provides the "glue"
| |
| 15:22 | <laga> LBE was nice.. but working with a "real" distro is a lot nicer for local apps
| |
| 15:22 | <cliebow> yes for sure
| |
| 15:22 | <laga> i still remember when i tried to add Qt and MythTV to the LBE
| |
| 15:22 | * laga goes catatonic | |
| 15:23 | <cliebow> hahah!!
| |
| 15:23 | ihad my share adding x11vnc
| |
| 15:23 | <laga> heh
| |
| 15:23 | LBE wasn't too bad, but i just wasn't up to the task back then
| |
| 15:24 | <cliebow> nor i..and now totally lost
| |
| 15:24 | <jammcq> how do you think I feel? I wrote LBE. Now, all I can do is watch as these guys work their magic
| |
| 15:25 | <Sockra_Tease> cliebow: its crazy trying to support anything redhat lol they dont even support their OWN stuff lol....good luck finding help with their RHGB
| |
| 15:25 | i tried to port it to slackware a couple years ago
| |
| 15:25 | it was a big mess of spaghetti code
| |
| 15:25 | <laga> jammcq: heh. didnt mean to thrash your work :)
| |
| 15:25 | <warren> From a realistic standpoint, major distros couldn't ship LBE-based LTSP to customers and have it be supportable for 7+ years.
| |
| 15:26 | LTSP5 makes it possible to support LTSP
| |
| 15:26 | <laga> what linux distros have support for 7+ years?
| |
| 15:26 | <warren> laga, RHEL
| |
| 15:26 | <laga> oh, nice
| |
| 15:27 | <jammcq> laga: hmm, didn't see any trashing
| |
| 15:27 | talntidwrk has joined #ltsp | |
| 15:27 | <laga> jammcq: good :)
| |
| 15:27 | <talntidwrk> hey, for the thin client manager.. to view the screens...
| |
| 15:27 | what is recommended?
| |
| 15:28 | <cliebow> Sockra_Tease, meet warren from fedora 8~)
| |
| 15:29 | laterrr..
| |
| 15:29 | cliebow has quit IRC | |
| 15:29 | * warren ? | |
| 15:32 | <dberkholz> jammcq: do you feel like you get to sit back and sip a mai tai? =)
| |
| 15:32 | <Sockra_Tease> lol hy warren :)
| |
| 15:32 | <warren> dberkholz, this is where we pay him daily royalty payments
| |
| 15:33 | <jammcq> dberkholz: I wish. got enough stuff going on here to keep 10 of me busy
| |
| 15:35 | <dberkholz> jammcq: i know the feeling.
| |
| 15:44 | TelnetManta has quit IRC | |
| 15:53 | vagrantc has joined #ltsp | |
| 16:04 | Guaraldo has left #ltsp | |
| 16:05 | Skarmeth has quit IRC | |
| 16:12 | RyanRyan52 has quit IRC | |
| 16:22 | topslakr is now known as Topslakr | |
| 16:22 | soneyka has quit IRC | |
| 16:23 | Pascal_1 has quit IRC | |
| 16:32 | slidesinger has quit IRC | |
| 16:42 | mccann has quit IRC | |
| 16:48 | abadger1999 has quit IRC | |
| 16:49 | abadger1999 has joined #ltsp | |
| 16:49 | abadger1999 has joined #ltsp | |
| 17:11 | <vagrantc> updated the ldm backport for etch: http://wiki.debian.org/LTSP/Howto (see the section mentioning backports)
| |
| 17:13 | there's one issue to get LDM_DIRECTX working which i'll document in the known issues section
| |
| 17:22 | <Ryan52> vagrantc: Did you find/fix the bug that was making it act wierd?
| |
| 17:23 | <vagrantc> Ryan52: act weird?
| |
| 17:23 | Ryan52: still had to apply that one patch with the autologin/guestlogin function ...
| |
| 17:23 | <Ryan52> oh, ya...that was what I meant.
| |
| 17:23 | <vagrantc> not the "it's just totally broken" issue you were having at freegeek ?
| |
| 17:24 | you were getting gtk errors and such?
| |
| 17:25 | Ryan52: i didn't actually test if it worked without the *login patch since it applied cleanly... but it would be good to minimize the diff between sid and the backport
| |
| 17:25 | <Ryan52> yes...I wonder if any of the changes will make it start to work at GF...
| |
| 17:25 | err, FG
| |
| 17:27 | <vagrantc> Ryan52: you had those issues with an upgraded environment, or a clean install?
| |
| 17:27 | one of these days i'm going to have to work on the etch -> lenny upgrade path.
| |
| 17:28 | i'm pretty sure there's some bugs at the moment.
| |
| 17:28 | Ryan52: oh! it might have been an issue with the default theme being set to something that didn't exist
| |
| 17:35 | Sockra_Tease is now known as cpunches | |
| 17:57 | deavid has quit IRC | |
| 17:59 | jammcq has quit IRC | |
| 17:59 | <Ryan52> ya...I never got to looking into it...
| |
| 18:00 | it was upgraded and clean
| |
| 18:00 | both on glacier though
| |
| 18:00 | maybe using a new server might have worked better
| |
| 18:01 | <vagrantc> Ryan52: version of ltsp-server on glacier was current?
| |
| 18:02 | slipttees has joined #ltsp | |
| 18:03 | <Ryan52> yes, it was the latest from backports
| |
| 18:04 | <vagrantc> Ryan52: the theme is bad
| |
| 18:06 | Ryan52: well, in /opt/ltsp/i386.bad, the theme points to a file that doesn't exist
| |
| 18:06 | <Ryan52> oh...that might have something to do with it...
| |
| 18:07 | * Ryan52 is too busy to care today anyway...but he will look into it next week or something. | |
| 18:07 | <vagrantc> Ryan52: it's an issue i've been meaning to fix, actually ... when you upgrade ldm, it doesn't contain all the same themes
| |
| 18:08 | and so if the old theme is bad, it gets borked...
| |
| 18:08 | s,bad,missing,
| |
| 18:09 | mikkel has quit IRC | |
| 18:30 | slipttees has quit IRC | |
| 18:42 | DonSilver has quit IRC | |
| 18:42 | cpunches has quit IRC | |
| 18:42 | talntidwrk has quit IRC | |
| 18:46 | talntidwrk has joined #ltsp | |
| 18:48 | bobby_C has quit IRC | |
| 18:49 | johnny has joined #ltsp | |
| 18:50 | Egyptian[Home1 has quit IRC | |
| 18:51 | Egyptian[Home] has joined #ltsp | |
| 18:52 | staffencasa has quit IRC | |
| 19:18 | <warren> vagrantc, hmm, I can use most of update-kernels if we make the part that edits the yaboot, pxelinux.cfg and other boot loaders optional
| |
| 19:19 | <vagrantc> warren: sure
| |
| 19:19 | warren: could even modularize or plugin-ify it.
| |
| 19:19 | <warren> OTOH I'm considering doing it entirely from outside the chroot instead
| |
| 19:19 | * vagrantc fought long and hard for doing it inside the chroot | |
| 19:19 | pilipo has joined #ltsp | |
| 19:19 | <warren> what benefits do you have inside?
| |
| 19:20 | there's otherwise no reason to install mkelfimage within the chroot
| |
| 19:20 | <johnny> i'm back..
| |
| 19:20 | <vagrantc> you're not depending on the server to have the right software to prepare your thin client for network booting
| |
| 19:20 | <warren> ok
| |
| 19:20 | <vagrantc> i like keeping everything relevent to the chroot within the chroot ...
| |
| 19:21 | <warren> btw
| |
| 19:21 | do you have anything to clean up old kernels from your tftpdir later?
| |
| 19:21 | <vagrantc> i.e. (at least on debian) all your boot related files are in /boot ... ltsp shouldn't be any different.
| |
| 19:21 | warren: nope. :)
| |
| 19:21 | <warren> I have 30 something kernel versions in my tftpdir =_
| |
| 19:21 | =)
| |
| 19:21 | <vagrantc> it needs a lot of improvements, but it's the kind of thing i'd like to work on.
| |
| 19:22 | <warren> ltsp-update-kernels only copies whatever is in the client /boot then
| |
| 19:22 | <vagrantc> that was the idea. keeping it really simple.
| |
| 19:23 | could move aside the old dir and just copy everything over fresh.
| |
| 19:23 | otavio has quit IRC | |
| 19:24 | <warren> that doesn't work fo rme
| |
| 19:24 | I use pxelinux.cfg entirely controlled by the sysadmin
| |
| 19:24 | never do my scripts touch it
| |
| 19:25 | <vagrantc> oh well.
| |
| 19:25 | <warren> #!/bin/sh -e
| |
| 19:25 | what does -e do?
| |
| 19:25 | <vagrantc> stop execution on error
| |
| 19:28 | <warren> why is that desired here?
| |
| 19:28 | jammcq has joined #ltsp | |
| 19:28 | <vagrantc> why is it undesireable?
| |
| 19:28 | <jammcq> good evening gents
| |
| 19:29 | <pilipo> hi all
| |
| 19:29 | <warren> well, you could just implement proper error checking and error messages
| |
| 19:29 | and control failure paths
| |
| 19:29 | <vagrantc> warren: patches accepted :)
| |
| 19:30 | <warren> vagrantc, does debian have some way to automatically run this whenever a kernel is updated in the chroot?
| |
| 19:30 | vagrantc, or it has to be run from something external to the chroot?
| |
| 19:33 | <vagrantc> warren: admin has to run the server-side copying.
| |
| 19:33 | warren: it's possible to set up your configuration file and tftp server so you don't have to update
| |
| 19:34 | i worked on that a while ... but it needs some more care.
| |
| 19:34 | warren: the nbi.img generation and such happens on the fly
| |
| 19:37 | i.e. when installing a new kernel, or removing a kernel ..
| |
| 19:52 | yay. got my hands on a network bootable powerpc :)
| |
| 19:53 | soon, it will be running ltsp.
| |
| 19:53 | of course, it has to install it's own root filesystem ...
| |
| 19:58 | <warren> jammcq, do you sell anything with linuxbios?
| |
| 20:01 | http://bazaar.launchpad.net/~ltsp-upstream/ltsp/ltsp-trunk/files
| |
| 20:01 | hmm, not sure what broke this.
| |
| 20:01 | <jammcq> nope
| |
| 20:01 | actually, I don't sell anything. I'm not part of DisklessWorkstations anymore
| |
| 20:02 | but disklessworkstations doesn't sell anything with a linuxbios
| |
| 20:02 | <warren> jammcq, really?
| |
| 20:02 | <jammcq> yep
| |
| 20:03 | <warren> according to our kernel hacker, linuxbios is perfectly capable of booting a kernel and initrd image, the manufacturers who loaded linuxbios must have done something stupid.
| |
| 20:04 | <jammcq> I know some of the linuxbios guys, and I fully believe that it should work just fine
| |
| 20:04 | <warren> then why are we required to use mkelfimage to wrap it?
| |
| 20:05 | mhterres has quit IRC | |
| 20:06 | <jammcq> umm, dunno why
| |
| 20:06 | TelnetManta has joined #ltsp | |
| 20:08 | MacIver has joined #ltsp | |
| 20:08 | <vagrantc> so we just hand it a single image?
| |
| 20:09 | probably shouldn't require that if there's another way ...
| |
| 20:43 | petre has joined #ltsp | |
| 20:43 | <petre> evening all
| |
| 21:18 | rjune__ has quit IRC | |
| 21:18 | pilipo has quit IRC | |
| 21:18 | spectra has quit IRC | |
| 21:49 | petre has quit IRC | |
| 21:58 | makghosh has quit IRC | |
| 22:03 | makghosh has joined #ltsp | |
| 22:08 | RyanRyan62 has joined #ltsp | |
| 22:13 | Ryan52 has quit IRC | |
| 22:16 | Ryan52 has joined #ltsp | |
| 22:17 | RyanRyan62 has quit IRC | |
| 22:17 | RyanRyan52 has joined #ltsp | |
| 22:21 | praveer_cool has joined #ltsp | |
| 22:57 | gonzaloaf has joined #ltsp | |
| 23:05 | <vagrantc> wahoo.
| |
| 23:05 | debian sid powerpc thin-client ...
| |
| 23:06 | it's like it's its' own grampa.
| |
| 23:15 | makghosh has quit IRC | |
| 23:16 | gonzaloaf has quit IRC | |
| 23:18 | gonzaloaf has joined #ltsp | |
| 23:21 | captain_magnus has quit IRC | |
| 23:55 | <johnny> vagrantc, word
| |
| 23:59 | <dberkholz> johnny: ah, there you are
| |
| 23:59 | johnny: got an ltsp overlay set up
| |
| 23:59 | <johnny> sweet
| |
| 23:59 | i was on vacation
| |
| 23:59 | but my computer went off at some point
| |
| 23:59 | so, what kinda overlay is it?
| |