IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 8 April 2008 (all times are UTC)
| 00:00 | <AzMoo> sutula: I'm trying to understand how this all works. My X is failing to start on the client so I ctrl+alt+f1 to a terminal and try to log in. It fails login, I'm assuming because the user wasn't created inside of the chroot, but everybody keeps telling me that it should just be using the users on the server.
| |
| 00:02 | <sutula> AzMoo: Normally, nobody would be logging in on the client, so all the above is true.
| |
| 00:03 | AzMoo: But, as you point out, you need to in order to debug the X failure, so adding a user to the chroot (or just using "passwd root" there) will allow you to log into the client
| |
| 00:03 | chupa has joined #ltsp | |
| 00:04 | <lns> (AzMoo, sutula, of course after running ltsp-update-image (if applicable))
| |
| 00:04 | <sutula> Also, (I think) ssh daemon is not normally installed in the chroot (client's env) so if you want to ssh in, you'd need to install it there
| |
| 00:04 | <AzMoo> lns: Absolutely, I've got that bit down.
| |
| 00:05 | sutula: I thought that was how the X session was started? ssh -C server etc. ?
| |
| 00:05 | * sutula nods to lns ...sutula has an older etch-based installation so still forgets the new stuff | |
| 00:05 | chup has quit IRC | |
| 00:05 | <sutula> AzMoo: Ah...that could also be a change from what I have
| |
| 00:06 | AzMoo: In any case, so you can ssh to the client, right? What else do you need?
| |
| 00:06 | <lns> so AzMoo, you're having a problem logging in on the console (ctl-alt-f1) with root?
| |
| 00:06 | <AzMoo> OK, so I've got the user bit down now. How about logging? One of my clients is panicking when it boots the image. Is there anywhere that's logged?
| |
| 00:06 | lns: No, I've fixed that now and understand what I was doing wrong.
| |
| 00:06 | <lns> oh
| |
| 00:06 | <sutula> dmsg on the client?
| |
| 00:06 | <lns> AzMoo, the best log is what it says when it panics (I would think)
| |
| 00:06 | <AzMoo> sutula: Yeah, I can't though because it's panicking before I get a chance to see anything.
| |
| 00:07 | Ubuntu's got its stupid fbsplash.
| |
| 00:07 | <johnny> you can disable that
| |
| 00:08 | in /var/lib/tftpboot/ltsp/i386/pxelinux.cfg/default
| |
| 00:08 | remove quiet and splash
| |
| 00:08 | <lns> johnny, is that in etch too?
| |
| 00:08 | <johnny> uhmm.. no idea
| |
| 00:08 | <AzMoo> johnny: Awesome, thanks. I was looking for it in /boot/
| |
| 00:09 | <lns> ah nm, that was sutula that said he used etch =p
| |
| 00:09 | * lns is still learning the current debian distro names | |
| 00:09 | <johnny> if you look in the pxelinux.cfg/default file.. you can find out
| |
| 00:09 | <AzMoo> etch for stable, lenny for testing, sid for development
| |
| 00:10 | <johnny> we need to revamp the ltsp wiki :(
| |
| 00:10 | <sutula> lns: Don't get me wrong...at this point in Debian's lifecycle, I wish I was on something newer. But I don't have the time to keep debugging stuff when the users are counting on it working :)
| |
| 00:10 | <AzMoo> Yeah, totally. There's no docs for this and it's really frustrating.
| |
| 00:10 | <lns> sutula, i agree with you
| |
| 00:10 | i love debian on the server
| |
| 00:11 | just never tried ltsp on it (always used ubuntu)
| |
| 00:11 | <AzMoo> On an upside, I'm learning an incredible amount about linux :D
| |
| 00:11 | <johnny> i run all my personal servers on gentoo
| |
| 00:11 | <sutula> lns: At least vagrantc has done etch-backports
| |
| 00:11 | <lns> I just installed etch yesterday on a replacement bind/apache2/ssh/ftp/etc server of mine
| |
| 00:11 | <johnny> but i'd prolly use debian elsewhere
| |
| 00:11 | <|Paradox|> LTSP could soon become even more desirable a solution for running terminals across the net : http://www.dailytech.com/CERN+Develops+Possible+Internet+Replacement+Unfathomably+Faster/article11394.htm
| |
| 00:11 | * sutula likes ubuntu on stand-alone laptops | |
| 00:12 | <lns> that link crashed my firefox :( haha...of course what doesn't these days
| |
| 00:14 | <AzMoo> God, CERN are so awesome.
| |
| 00:15 | |Paradox|: The reason I'm trying to understand all this is to get LTSP working with FreeNX, which judging by their demos, will already allow LTSP over the net.
| |
| 00:15 | <lns> AzMoo, it's been stated by some of the ltsp devs that freenx isn't really a viable solution to couple with ltsp
| |
| 00:15 | without heavy hacking anyway
| |
| 00:15 | * AzMoo is up for heavy hacking. | |
| 00:16 | <AzMoo> And there's been lots of instances of people around the net doing it, just nobody's put any directions on how in.
| |
| 00:17 | <lns> right
| |
| 00:17 | i've used it on a testbed so i'm not sure how well it scales
| |
| 00:17 | i heard it is kind of resource intensive
| |
| 00:17 | at least compared to straight-up ltsp
| |
| 00:17 | <AzMoo> That's ok, I have some great hardware here.
| |
| 00:18 | And we're only looking at about 20 users at the most, it's just we need it going across a WAN.
| |
| 00:19 | <lns> ah that's not bad
| |
| 00:19 | <AzMoo> I did a test with standard LTSP the other day and we were hitting about 45 minutes to boot and get ssh up and going.
| |
| 00:19 | er, not ssh, the entire X session.
| |
| 00:19 | Which is obviously not acceptable.
| |
| 00:20 | tux_440volt has quit IRC | |
| 00:21 | <lns> what speed link are you using?
| |
| 00:21 | and are you actually pxe booting over a wan? or just contacting an x server?
| |
| 00:23 | <AzMoo> pxe booting over the WAN. The initial image was OK, that took about 10 minutes, but after that it was opening the applications and some basic clicking around. It was a 512Kbps SDSL connection which was being maxed out the entire time. That can be upgraded though, just not too far.
| |
| 00:26 | <lns> AzMoo, that's understandable for all that stuff
| |
| 00:26 | i mean the image isn't that big
| |
| 00:26 | well 45min...jeez even for 512Kbps is really odd
| |
| 00:27 | <AzMoo> Yeah, it took forever. Scrolling around was almost impossible.
| |
| 00:28 | Loading pages in web apps was fine, but as soon as there was any kind of scrolling or window movement it all went to hell.
| |
| 00:28 | So, I figure I need more compression.
| |
| 00:29 | <lns> AzMoo, you're not using ldm_directx in lts.conf are you?
| |
| 00:29 | ssh really does compress it a lot..still might be real choppy over 512 but still
| |
| 00:29 | <AzMoo> lns: Nope. My lts.conf is practically empty. It had a user and a password and that's about it.
| |
| 00:30 | Even the window manager was minimal. No toolbars, no menus, just firefox and a couple of other little app windows.
| |
| 00:31 | <lns> well not that you can compare, but i use ssh tunneling to clients of mine and run VNC sessions through it
| |
| 00:31 | and it's very smooth over 1.5Mbps
| |
| 00:31 | now of course that's vnc, not x11 protocol
| |
| 00:32 | but x11 is still very lightweight afaik
| |
| 00:32 | and compressed with ssh is even smaller
| |
| 00:32 | indradg has quit IRC | |
| 00:32 | <lns> i remember doing straight X11 session over SSH, over a 56k modem connection
| |
| 00:33 | running KDE
| |
| 00:33 | it was hooooorible...but i mean, it didn't take no 45 minutes
| |
| 00:33 | maybe 10 minutes at the longest to display the background
| |
| 00:33 | lol
| |
| 00:33 | man that was horrible
| |
| 00:35 | <AzMoo> lns: I've been doing a lot of reading and apparently it's not the X11 protocol that's the problem, but the way apps are ignoring the network functionality of it nowdays./
| |
| 00:37 | <lns> AzMoo, unfortunately, yes...many apps are not aware of their own roots in x11. :( it sucks when you think that x11 was designed for network use, and now it's hard to find apps that don't crash under a network
| |
| 00:44 | <johnny> firefox3 is supposed to be better right?
| |
| 00:45 | <AzMoo> johnny: It's still in beta, but yeah, that's the plan.
| |
| 00:58 | <johnny> here we go..
| |
| 00:58 | see if i can go from boot to login screen ..
| |
| 00:58 | yay1
| |
| 01:00 | * AzMoo applauds! | |
| 01:00 | <johnny> i'm learning the internals of ltsp deeply now ..
| |
| 01:00 | reimplementing it is interesting.
| |
| 01:01 | chupa has quit IRC | |
| 01:01 | chupa has joined #ltsp | |
| 01:02 | <johnny> dberkholz, my newest chroot can go from boot to ldm now
| |
| 01:03 | still need to integrate my newest modifications tho
| |
| 01:04 | now if only some other ldm expert was here now .. :)
| |
| 01:06 | <dberkholz> johnny: rockin
| |
| 01:06 | johnny: are your changes uploaded?
| |
| 01:07 | <johnny> no.. i have to integrate them still
| |
| 01:07 | need to test it from a from scratch chroot now
| |
| 01:07 | <dberkholz> johnny: not sure what you mean by integrate...
| |
| 01:07 | <johnny> they are local modifications
| |
| 01:08 | ldm requires ubuntulooks in ldm-trunk
| |
| 01:09 | so, i emerged gtk-engines
| |
| 01:09 | and then symlinked clearlooks overtop of the greeter-gtkrc
| |
| 01:10 | that's probably not required
| |
| 01:12 | gtk_rc_add_default_file(ldm_theme_file("/greeter-gtkrc")); seems to be unconditionally done
| |
| 01:13 | deavid has joined #ltsp | |
| 01:13 | <johnny> also LDM_THEME wasn't defined, i'm still investigating why
| |
| 01:13 | i had to add it to my lts.conf to get it to work
| |
| 01:16 | ldm doesn't check to make sure there is valid theme before trying to include the images files
| |
| 01:23 | plamengr has joined #ltsp | |
| 01:45 | <AzMoo> lns: It seems you're right. FreeNX is going to be too hard. I'm going to have to rebuild half of the distro to do it.
| |
| 01:48 | But now it's a challenge. It's just that I'm going to have to do it at home.
| |
| 01:48 | ;)
| |
| 02:16 | <johnny> exit
| |
| 02:17 | oops
| |
| 02:21 | deavi1 has joined #ltsp | |
| 02:21 | deavid has quit IRC | |
| 02:27 | ogra_cmpc has joined #ltsp | |
| 02:30 | plamengr has quit IRC | |
| 02:34 | AzMoo has quit IRC | |
| 02:39 | jamesjames has joined #ltsp | |
| 02:40 | jamesjames has quit IRC | |
| 02:47 | mikkel has joined #ltsp | |
| 02:49 | exodos has joined #ltsp | |
| 02:49 | Q-FUNK has joined #ltsp | |
| 02:50 | exodos has quit IRC | |
| 02:51 | exodos has joined #ltsp | |
| 02:51 | bobby_C has joined #ltsp | |
| 02:52 | Q-FUNK has quit IRC | |
| 03:08 | plamengr has joined #ltsp | |
| 03:15 | jamesjames has joined #ltsp | |
| 03:33 | Q-FUNK has joined #ltsp | |
| 03:33 | viking-ice has quit IRC | |
| 03:52 | jamesjames has quit IRC | |
| 04:11 | viking-ice has joined #ltsp | |
| 04:17 | topslakr_ has quit IRC | |
| 04:18 | topslakr_ has joined #ltsp | |
| 04:46 | cbrmonger has joined #ltsp | |
| 04:57 | cbrmonger has left #ltsp | |
| 05:16 | viking-ice_ has joined #ltsp | |
| 05:21 | <Q-FUNK> ogra_cmpc: seems that geode 2.8.0-4 is ready to sync. :)
| |
| 05:29 | <ogra_cmpc> yay
| |
| 05:29 | i et bryce is subscribed to the bug ?
| |
| 05:29 | *bet
| |
| 05:30 | <Q-FUNK> he is
| |
| 05:30 | <ogra_cmpc> great :)
| |
| 05:30 | <Q-FUNK> but I just noticed an oversight in the package description. just uploaded -5 to mentors, now
| |
| 05:30 | <ogra_cmpc> he already asked if i had objections last week ...
| |
| 05:30 | <Q-FUNK> still, it's only a cosmetic change in the package description. -4 works just as well
| |
| 05:31 | ah ok
| |
| 05:31 | did you get a chance to test it with your sample?
| |
| 05:36 | here, it works perfectly fine, except if usplash is used. usplash seems to do something nasty to the frame buffer that temporarily distorts the image one initially gets when launching X
| |
| 05:37 | if you switch vcons and back, though, it disappears
| |
| 05:37 | chupa has quit IRC | |
| 05:37 | <ogra_cmpc> ouch
| |
| 05:37 | chupa has joined #ltsp | |
| 05:37 | <ogra_cmpc> thats pretty bad
| |
| 05:37 | there wont be ubuntu systems withjout usplash
| |
| 05:38 | unless yu use server, but then you wont care for X
| |
| 05:39 | <Q-FUNK> true
| |
| 05:39 | gotta admit that usplash keeps on getting worse with age, though
| |
| 05:39 | <ogra_cmpc> iirc there is a chvt call in the usplash initscript
| |
| 05:39 | it didnt change since three releases
| |
| 05:39 | the framebuffer handling in the kernel did though
| |
| 05:40 | <Q-FUNK> well, everything starting with 0.4 has exposed various bugs on varioius hardware, for me
| |
| 05:40 | ever since they started using a config file to define the expected rez, usplash has been a real PITA on most hardware I have
| |
| 05:44 | daduke: tarzeau said that you have some GX/LX ahrdware to try the new driver with?
| |
| 05:44 | <daduke> Q-FUNK: indeed we do
| |
| 05:45 | <Q-FUNK> on Etch, was it?
| |
| 05:45 | <daduke> Q-FUNK: exactly
| |
| 05:45 | <Q-FUNK> which hardware, again?
| |
| 05:45 | <daduke> Q-FUNK: it's this one: http://pcengines.ch/alix1c.htm
| |
| 05:45 | <Q-FUNK> ah yes
| |
| 05:46 | <daduke> Q-FUNK: have it on my desk right here, what do you want me to do?
| |
| 05:46 | <Q-FUNK> we make this one here: http://www.thincan.com
| |
| 05:46 | daduke: we're looking for people to test this on as wide a range of hardware as possible
| |
| 05:47 | <daduke> Q-FUNK: errr isn't the hw given?
| |
| 05:47 | Q-FUNK: or is there some fancy software involved?
| |
| 05:47 | <Q-FUNK> testing -geode on the alix is welcome, since it's a fairly wide-spread board
| |
| 05:48 | with our board and the OLPC, it's also the only Geode hardware on the market that can boot off coreboot
| |
| 05:49 | <daduke> Q-FUNK: well I'm always interested in revolutions ;)
| |
| 05:51 | Q-FUNK: but again: what shall I do?
| |
| 05:52 | <Q-FUNK> as mentioned above, there's some odd bugs which surface on an LX with usplash.
| |
| 05:52 | I'm wondering if you also get these on your hardware.
| |
| 05:52 | <daduke> Q-FUNK: tell me how and I'm gonna try
| |
| 05:54 | <Q-FUNK> install usplash with some usplash theme. reconfigure the linux image. install the geode driver. reboot
| |
| 05:54 | do you get bugs in the image when you reboot back into X?
| |
| 05:55 | if you do, try purging usplash, then rebooting. do you still get bugs in the image?
| |
| 05:55 | if you don't, this might indicate that this is a product-specific bug
| |
| 05:56 | <daduke> Q-FUNK: and who's usplash?
| |
| 05:57 | <Q-FUNK> whose usplahs theme? any
| |
| 05:57 | <daduke> Q-FUNK: no. I have no idea what usplash is.
| |
| 05:57 | <Q-FUNK> I don't think that this is specific to any particular theme
| |
| 05:57 | <ogra_cmpc> is it even packaged in debian ?
| |
| 05:57 | <Q-FUNK> ah. usplash is a package that displays a boot splash during bootup
| |
| 05:57 | yes
| |
| 05:58 | <ogra_cmpc> i doubt it wrks with the debian kernel though
| |
| 05:58 | <Q-FUNK> there are just fewer themes in debian
| |
| 05:58 | <daduke> Q-FUNK: all right, now we're getting somewhere ;)
| |
| 05:58 | <Q-FUNK> it does
| |
| 05:58 | <ogra_cmpc> it requires qa certain frambuffer setup i doubt debian has
| |
| 05:58 | so testing it there is moot
| |
| 05:59 | <daduke> Q-FUNK, ogra_cmpc: you guys sort it out and I'll come back when you know, ok?
| |
| 05:59 | <Q-FUNK> it works on debian too. my development environment is on debian.
| |
| 05:59 | <ogra_cmpc> daduke, if Q-FUNK says it works i have to belive him, i dont have debian anywhere around
| |
| 05:59 | <Q-FUNK> the new default debian usplash theme looks quite nice on it, too
| |
| 06:00 | hm.. firefox froze. brbr
| |
| 06:00 | Q-FUNK has quit IRC | |
| 06:00 | <ogra_cmpc> but i know that debian allows all sorts of framebiuffers we dont in the ubuntu kernel, so the result is very likely not accurate to what you would get with the ubuntu kernel
| |
| 06:00 | (but its surely worth a try)
| |
| 06:01 | Q-FUNK has joined #ltsp | |
| 06:03 | <daduke> Q-FUNK: ok, I guess we just try.
| |
| 06:03 | <Q-FUNK> IIRC in Etch, the only usplash theme available was for debian-edu
| |
| 06:03 | <ogra_cmpc> just an apt-get install usplash in the chroot and ltsp-update-kernels should suffice
| |
| 06:04 | (assuming the deps in etch are right)
| |
| 06:04 | <daduke> Q-FUNK, ogra_cmpc: I'm not gonna do it on ltsp production, but I have an etch CF here
| |
| 06:04 | <Q-FUNK> the image too, to get the progression during boot, no?
| |
| 06:05 | <ogra_cmpc> Q-FUNK, usplash should provide a default theme (or depend on one)
| |
| 06:09 | <Q-FUNK> IIRC that was an issue in Etch
| |
| 06:09 | TelnetManta has quit IRC | |
| 06:09 | <Q-FUNK> themes came during the freeze
| |
| 06:10 | <ogra_cmpc> ah
| |
| 06:15 | <Q-FUNK> IIRC only the one for skolelinux/debian-edu was ready on time
| |
| 06:26 | <daduke> Q-FUNK: it's installing and initramfs'ing now
| |
| 06:28 | vagrantc has joined #ltsp | |
| 06:29 | <daduke> Q-FUNK: I rebooted and everything looks like before.
| |
| 06:29 | <Q-FUNK> no boot splash?
| |
| 06:29 | <daduke> Q-FUNK: nope.
| |
| 06:29 | <Q-FUNK> hm
| |
| 06:29 | vagrantc has joined #ltsp | |
| 06:29 | <Q-FUNK> ah, wait. etch. I'm on Lenny at home.
| |
| 06:29 | <daduke> Q-FUNK: initrd is fresh tho
| |
| 06:29 | <Q-FUNK> that might explain it
| |
| 06:30 | * daduke waves to vagrantc | |
| 06:30 | <Q-FUNK> yes, initramfs is rebuilt when you install or remove usplash or a uspplash theme
| |
| 06:30 | * vagrantc waves | |
| 06:31 | plamengr has quit IRC | |
| 06:31 | <daduke> Q-FUNK: so? what more do I need to do?
| |
| 06:31 | <Q-FUNK> hmm
| |
| 06:31 | I haven't touched Etch in ages
| |
| 06:32 | is vesafb loaded?
| |
| 06:32 | or some other frame buffer?
| |
| 06:33 | <daduke> Q-FUNK: no, I'm gonna try to boot with vga=791 now
| |
| 06:35 | Q-FUNK: I'm in fb mode now, but still no module loaded and no change from usplash
| |
| 06:35 | jammcq has quit IRC | |
| 06:36 | <daduke> Q-FUNK: vesafb seems to be in the kernel. fb0: VESA VGA frame buffer device
| |
| 06:38 | vagrantc, ogra_cmpc: with the ltsp live cd up and running, I'm ready for the next task: dpms screen power safe in ldm. DPMS in xorg.conf is not enough on our hw (as is the case for a lot of other systems too: http://www.shallowsky.com/linux/x-screen-blanking.html). I now plan to integrate vbetool to shut down the screen. any thoughts on this?
| |
| 06:39 | jonkke has joined #ltsp | |
| 06:39 | <vagrantc> daduke: i have no idea how to do it, but good luck :)
| |
| 06:40 | <jonkke> Hi
| |
| 06:40 | <muh2000> swesome kicking asses, my geode works :) x is running, music is playing :D
| |
| 06:40 | <vagrantc> daduke: it shouldn't be much different from a "normal" machine.
| |
| 06:41 | <muh2000> all i hjave to do is wait a few hours till x comes up ^^
| |
| 06:41 | hmmbut damn sound is only mono asnd left channel
| |
| 06:42 | <daduke> vagrantc: thing is, with the 'old' ldm (before the last update), I could xhost +localhost and start xscreensaver which allowed me to use some hooks to control vbetool. however, the latest ldm seems to prevent this... X connection refused.
| |
| 06:42 | muh2000: congrats!
| |
| 06:43 | <vagrantc> daduke: that is because the old LDM allowed anybody and everybody to connect to it :)
| |
| 06:43 | <jonkke> is there any way of displaying thin client name (ws001)? I found this: https://lists.ubuntu.com/archives/edubuntu-users/2007-January/000601.html , but when i type "echo $DISPLAY" it displays client ip address.
| |
| 06:43 | <vagrantc> daduke: probably didn't even need xhost
| |
| 06:44 | <daduke> vagrantc: not that old. the one in between.
| |
| 06:44 | <muh2000> ah right chanel works, plug wasnt plugged right.
| |
| 06:44 | <vagrantc> daduke: using LDM_DIRECTX ?
| |
| 06:44 | <muh2000> daduke: thnx :) ( all i did was letting the system load over night :D )
| |
| 06:44 | <daduke> vagrantc: yep
| |
| 06:45 | <vagrantc> daduke: well, without knowing which versions you're talking about, the "one in between" still disabled all security measures for LDM_DIRECTX.
| |
| 06:46 | <daduke> jonkke: echo $HOSTNAME
| |
| 06:46 | <vagrantc> daduke: can you log in at all?
| |
| 06:47 | <jonkke> daduke: it prints servers hostname
| |
| 06:48 | <daduke> vagrantc: everything is working fine, I just cannot get xscreensaver (running as root, since it's ldm and nobodys has logged in) to connect to :7.
| |
| 06:48 | johnny: not if you do it locally on the thin client
| |
| 06:49 | <vagrantc> daduke: you might be able to use the cookie generation code to generate an xauth cookie for localhost ...
| |
| 06:49 | <jonkke> daduke: I have screen03 as shell and there "echo $HOSTNAME" prints client hostname, but if i'm on Gnome, it displays server hostname. i have ubuntu and ltsp4.2
| |
| 06:49 | <vagrantc> daduke: look in the ldm screen.d script, and the ldm rc.d script
| |
| 06:50 | <jonkke> daduke: i'm trying to create something to make one printer A default to one classroom and printer B default to another
| |
| 06:51 | <muh2000> is this: "VOLUME = 84" still a valid value for lts.conf ?
| |
| 06:51 | (ltsp5)
| |
| 06:51 | <daduke> vagrantc: I think that would be something like http://www.jwz.org/xscreensaver/faq.html#root-solution, right? I tried, but I still get connection refused.
| |
| 06:52 | <jonkke> daduke: if you have some better ways to do that than script in /etc/profile, i would be happy to hear. and sorry for my bad english :)
| |
| 06:52 | <daduke> jonkke: sorry, I have no idea about 4.2
| |
| 06:53 | <jonkke> daduke: ok. i have this system daily in use, so i have no time to upgrade to ltsp5. i will do that when kids go to summer vacation. and maybe i leave this problem wait for ltsp5 too.
| |
| 06:54 | <vagrantc> daduke: sorry, i'm still in transit and need to look at other things at the moment...
| |
| 06:54 | <daduke> jonkke: can't you just set each user's $PRINTER to whatever you need?
| |
| 06:55 | vagrantc: no prob. I was just wondering if nobody has ever thought about turning of screens in ldm - it saves so much power
| |
| 06:56 | cliebow has joined #ltsp | |
| 06:56 | <daduke> Q-FUNK: I haven't forgotten about you, but what can I do?
| |
| 06:56 | <muh2000> hmmm geode does not seem to be able to playback simple avi files :/ but at least music is working :)
| |
| 06:56 | <vagrantc> daduke: yeah, i'd like to hear about your sucesses with it, and we can hopefully make it an lts.conf option.
| |
| 06:57 | <daduke> vagrantc: fair enough ;)
| |
| 06:57 | mhterres has joined #ltsp | |
| 06:57 | <jonkke> daduke: it must be per ws, because users can login from anywhere at school.
| |
| 06:57 | <daduke> vagrantc: OTOH, it might be easiest to just integrate it into ldm.
| |
| 06:58 | <vagrantc> daduke: once you know how to do it, we'll figure out how best to integrate it :)
| |
| 06:59 | <daduke> vagrantc: well then system('vbetool dpms off') in ldm.c
| |
| 06:59 | <vagrantc> daduke: why not from the screen script?
| |
| 06:59 | oh, right ... ldm doesn't start the X server
| |
| 06:59 | er, ldm starts the X server
| |
| 07:01 | <daduke> vagrantc: we need to have something that monitors kb and mouse. if we do it outside of ldm, we need xscreensaver. we could create a screensaver program that executes vbetool, but then we need some 'screensaver exit' hook to turn on the screen again. I think I found that already. but still, the permission problem.
| |
| 07:01 | vagrantc: it might just be easier to look for kb and mouse events in ldm directly
| |
| 07:08 | <Q-FUNK> daduke: I'm wondering. Etch is so far back in mind that I can barely remember what was it that
| |
| 07:09 | <daduke> Q-FUNK: come on, it's still what they call stable ;)
| |
| 07:10 | jammcq has joined #ltsp | |
| 07:10 | <jammcq> g'morning folks
| |
| 07:13 | * vagrantc notes that some people are unstable | |
| 07:15 | <Q-FUNK> daduke: it's what i call "so obsolete that william shatner might have particiipated in the release" :)
| |
| 07:16 | <laga> star trek TOS is not obsolete
| |
| 07:16 | <tarzeau> FLASH GORDON > star trek
| |
| 07:16 | deavi1 has quit IRC | |
| 07:16 | <laga> tarzeau: yeah, but only because queen made the sound track for the latest movie
| |
| 07:18 | <Q-FUNK> flash! flaaaaaaaaaaaaaaaaaaaaaaaaaaash!
| |
| 07:18 | <daduke> it's interesting how many ppl get active as soon as a flamewar starts...
| |
| 07:18 | plamengr has joined #ltsp | |
| 07:19 | <Q-FUNK> and wasn't it Toto?
| |
| 07:19 | <laga> Q-FUNK: queen.
| |
| 07:19 | seriously!
| |
| 07:20 | * Q-FUNK is going all radio gaga again | |
| 07:20 | <laga> and while we're at it, flash just killed my browser.
| |
| 07:20 | Q-FUNK: the original flash gordon series from the 1930ies are nice, too
| |
| 07:21 | <Q-FUNK> I have a feeling that we're not in kansas anymore
| |
| 07:21 | Guaraldo has joined #ltsp | |
| 07:21 | * cliebow cliebow notes that jammcq is Not unstable 8~) | |
| 07:22 | <daduke> Q-FUNK: got it. I installed debian-edu-artwork-usplash and now I get gfx at least when powering down.
| |
| 07:23 | <Q-FUNK> was it dario that had a blog called "debian gnu/human unstable" ?
| |
| 07:25 | <tarzeau> daduke: http://wiki.debian.org/DebianLive/Howto/Custom_Artwork says: "Usplash is a pain."
| |
| 07:25 | da steht noch: LH_BOOTAPPEND_LIVE="quiet splash vga=792"
| |
| 07:26 | <Q-FUNK> evet
| |
| 07:26 | <tarzeau> isn't splashy better than usplash?
| |
| 07:26 | <vagrantc> and has had RC bugs for ages
| |
| 07:26 | <tarzeau> splashes are why i quit windows
| |
| 07:26 | i wanted to see what happens
| |
| 07:27 | splashy as well as usplash have 6 important bugs in debian bts (sid)
| |
| 07:28 | usplash has a bigger version number than splashy
| |
| 07:28 | <daduke> Q-FUNK: all right, tarzeau's hints did it. now it's working. what do you want to have tested now?
| |
| 07:28 | <laga> tarzeau: you can easily disable usplash on ubuntu
| |
| 07:28 | <tarzeau> laga: i know, i prefer debian
| |
| 07:29 | <Q-FUNK> daduke, whether you get odd distortion in the image, when X initially launches
| |
| 07:29 | <daduke> Q-FUNK: please stand by
| |
| 07:34 | elisboa has quit IRC | |
| 07:34 | elisboa has joined #ltsp | |
| 07:37 | <Q-FUNK> here, when I boot and let GDM start, X initially has some trashy pixels in the image.
| |
| 07:37 | howeve,r if I boot without usplash, e.g. by selecting the maintenanc eimage in grub, i don't get those trashy pixels
| |
| 07:38 | <vagrantc> would anyone want to kill me if i renamed /lib/udev/add_fstab_entry and remove_fstab_entry to something more like *_ltspfs_entry ?
| |
| 07:39 | we cna include symlinks for backwards compatibility
| |
| 07:41 | <daduke> Q-FUNK: does this answer you questions? https://www.phys.ethz.ch/~daduke/usplash.avi
| |
| 07:41 | slidesinger has joined #ltsp | |
| 07:42 | <Q-FUNK> ssl certificate problem trying to download this
| |
| 07:43 | and error 403 if I bypass the ssl certificate warning
| |
| 07:44 | <daduke> Q-FUNK: sorry, perm probs. retry now
| |
| 07:48 | jonkke has quit IRC | |
| 07:48 | <gvy> hi folks
| |
| 07:54 | Q-FUNK has quit IRC | |
| 07:55 | rjune_ has quit IRC | |
| 07:57 | Q-FUNK has joined #ltsp | |
| 08:01 | <daduke> Q-FUNK: silence == satisfaction ?
| |
| 08:04 | chup has joined #ltsp | |
| 08:05 | chupa has quit IRC | |
| 08:06 | <Q-FUNK> that avi killed totem and froze the desktop to death too
| |
| 08:07 | pretty scary stuff
| |
| 08:07 | I could move my mouse, but the keyboard became unresponsive
| |
| 08:07 | I could not even change to a vcons
| |
| 08:08 | mikkel has quit IRC | |
| 08:14 | <daduke> Q-FUNK: I'm sorry. it's whatever my pentax produces. mplayer plays it fine. don't sue me now.
| |
| 08:16 | <Q-FUNK> :)
| |
| 08:16 | anyhow. what was I supposed to see? :)
| |
| 08:23 | <daduke> Q-FUNK: the system booting, progress bar and all, and X coming up. I don't see an obvious problem
| |
| 08:23 | <muh2000> lol i am waiting now 1:20 hours/minuts for x to come up :D
| |
| 08:23 | sep has quit IRC | |
| 08:24 | <Q-FUNK> daduke: ok, good to know
| |
| 08:24 | <daduke> muh2000: on an alix? it's more like 40 sec here
| |
| 08:25 | <muh2000> daduke: yes :) alix3c3. i dont know why it takes that long. i found out when i started the box yesterday and switched the monitor on again 10 hours later - there was the ltsp login smiling at me :D
| |
| 08:26 | <daduke> muh2000: and what does it do in the meantime?
| |
| 08:26 | <Q-FUNK> muh2000: which OS was it again?
| |
| 08:26 | <muh2000> daduke: showing a black screen :)
| |
| 08:26 | Q-FUNK: server: debian lenny; chroot/ltsp: debian sid
| |
| 08:26 | kernel 2.6.22.something
| |
| 08:26 | * vagrantc is so happy to see a recent ldm hit lenny | |
| 08:27 | <vagrantc> ogra_cmpc: you saw the ldm upload to experimental?
| |
| 08:28 | <ogra_cmpc> i saw there was one, yes
| |
| 08:28 | sorry, extremly busy today ...
| |
| 08:28 | <vagrantc> ogra_cmpc: no problem.
| |
| 08:31 | sep has joined #ltsp | |
| 08:34 | Guaraldo has left #ltsp | |
| 08:37 | muh2000 has quit IRC | |
| 08:40 | muh2000 has joined #ltsp | |
| 08:42 | otavio has quit IRC | |
| 09:03 | TelnetManta has joined #ltsp | |
| 09:06 | cyberorg has quit IRC | |
| 09:09 | mnemoc has quit IRC | |
| 09:09 | mnemoc has joined #ltsp | |
| 09:22 | Q-FUNK has quit IRC | |
| 09:33 | <warren> ogra_cmpc: how does sshd do consolekit on ubuntu?
| |
| 09:33 | vagrantc: is sshd doing CK on debian now?
| |
| 09:34 | ogra_cmpc: how do I install the absolute latest ubuntu so I can take a look for myself?
| |
| 09:35 | <cliebow> warren:ogra is tied up today..
| |
| 09:35 | <warren> does anyone else know how to install the latest ubuntu?
| |
| 09:35 | (the unreleased one)
| |
| 09:37 | <ogra_cmpc> warren, with the patch to sshd i showed you ages ago ...
| |
| 09:37 | <warren> oh shit
| |
| 09:37 | ogra_cmpc: yeah, we aren't allowed to use that unfortunately.
| |
| 09:37 | <ogra_cmpc> yep, i know
| |
| 09:38 | but the ssh maintainer in ubuntu and debian is the same ... i'm pretty sure he merged the patch to debian as well, to save doubled work
| |
| 09:39 | Gadi has joined #ltsp | |
| 09:39 | <ogra_cmpc> but i'd appreciate a fallback functionallity i.e. if i connect to non debian/ubuntu servers
| |
| 09:40 | but the solution shouold be done client side ... the rc.d scripts can easily operate through the ssh socket to run anything you want or get any info you want from the server
| |
| 09:45 | mccann has joined #ltsp | |
| 09:47 | <gvy> vagrantc, warren, q: is it really neccessary to do #!/bin/sh -e in update-kernels? (simple command bails out the whole script with -e)
| |
| 09:47 | and that doesn't help with that many tests
| |
| 09:47 | <vagrantc> gvy: if you handle errors properly, no.
| |
| 09:47 | <gvy> the result might be e.g. staying with no initrd
| |
| 09:47 | vagrantc, *if* :)
| |
| 09:48 | <vagrantc> gvy: but currently, i think it is better with -e
| |
| 09:48 | <warren> gvy: yeah, I don't like it either.
| |
| 09:48 | <gvy> okay, thanks
| |
| 09:48 | <warren> gvy: I haven't touched update-kernels yet, I intend on soon
| |
| 09:48 | <gvy> warren, tnx :)
| |
| 09:49 | <vagrantc> i've been meaning to re-write *update-kernels for over a year now, but it's worked "good enough" for me that i've focused on other things.
| |
| 09:52 | chup has quit IRC | |
| 09:52 | chup has joined #ltsp | |
| 09:56 | deavid has joined #ltsp | |
| 10:07 | bobby_C has quit IRC | |
| 10:14 | elisboa has quit IRC | |
| 10:14 | otavio has joined #ltsp | |
| 10:17 | otavio has quit IRC | |
| 10:18 | otavio has joined #ltsp | |
| 10:35 | staffencasa has joined #ltsp | |
| 10:42 | otavio has quit IRC | |
| 10:44 | chupa has joined #ltsp | |
| 10:44 | otavio has joined #ltsp | |
| 10:45 | chup has quit IRC | |
| 10:48 | * vagrantc heads for the train... | |
| 10:48 | vagrantc has quit IRC | |
| 10:52 | praveer_cool has joined #ltsp | |
| 11:00 | <ogra_cmpc> gah, to late for hardy .... https://lists.ubuntu.com/archives/edubuntu-users/2008-April/003873.html
| |
| 11:00 | praveer_cool has quit IRC | |
| 11:08 | exodos has quit IRC | |
| 11:11 | <Gadi> ogra_cmpc: you deeal with livecds a lot - do you know a bootparam to add a delay before mounting the rootfs (to let the drivers settle)?
| |
| 11:12 | * Gadi is not sure rootdelay is implemented in casper | |
| 11:12 | <ogra_cmpc> there is a timeout value
| |
| 11:12 | might be rootdelay, yes
| |
| 11:13 | <Gadi> ${LIVEMEDIA_TIMEOUT}
| |
| 11:13 | maybe thats it
| |
| 11:14 | <ogra_cmpc> it compares with some $timeout value and then delays ...
| |
| 11:15 | ah, $timeout is the sequence number from mountroot
| |
| 11:15 | <Gadi> right
| |
| 11:15 | I see that
| |
| 11:16 | I guess LIVEMEDIA_TIMEOUT has to be set in casper.conf
| |
| 11:16 | <ogra_cmpc> ${LIVEMEDIA_TIMEOUT} seems to be right then
| |
| 11:16 | <Gadi> wish it could be a kernel arg
| |
| 11:16 | <ogra_cmpc> you can just use it as kernel bootoption
| |
| 11:16 | <Gadi> ah, really?
| |
| 11:16 | <ogra_cmpc> the kernel usually passes env values through
| |
| 11:16 | <Gadi> ah, okey dokey
| |
| 11:16 | will give it a whirl
| |
| 11:16 | thx
| |
| 11:16 | <ogra_cmpc> try it, not sure it still does
| |
| 11:18 | viking-ice_ has quit IRC | |
| 11:18 | viking-ice has quit IRC | |
| 11:18 | <ogra_cmpc> https://bugs.launchpad.net/ubuntu/+source/ltsp/+bug/212802
| |
| 11:18 | mumble mumble ...
| |
| 11:19 | K_O-Gnom has joined #ltsp | |
| 11:19 | <ogra_cmpc> why oh why is g-p-m so silly ...
| |
| 11:20 | <muh2000> the mouse thing?
| |
| 11:20 | ahhh after 4 hours and 20 minutes: there is a login screen :)
| |
| 11:22 | makghosh has joined #ltsp | |
| 11:23 | spectra has joined #ltsp | |
| 11:23 | <ogra_cmpc> muh2000, gnome-power-manager (gpm has no dawshes ;) )
| |
| 11:23 | *dashes
| |
| 11:24 | i was referring to the bug report above
| |
| 11:26 | mccann has quit IRC | |
| 11:26 | muh2000 has quit IRC | |
| 11:32 | mccann has joined #ltsp | |
| 11:38 | <johnny> ogra_cmpc, hey :)
| |
| 11:42 | i have an ldm login screen, but it never logs in
| |
| 11:42 | just spits me back
| |
| 11:43 | <cliebow> got a homedir?
| |
| 11:43 | <johnny> yes
| |
| 11:43 | <cliebow> or ltsp-update-sshkeys
| |
| 11:43 | <ogra_cmpc> check .xsession-errors
| |
| 11:43 | <johnny> it's cannot open display
| |
| 11:44 | <ogra_cmpc> cliebow, since gutsy ldm shows an error if the keys are out of sync
| |
| 11:44 | <johnny> in .xsession-errors
| |
| 11:44 | <cliebow> ahhh..thank you
| |
| 11:44 | <johnny> ogra_cmpc, !
| |
| 11:44 | <ogra_cmpc> (instead of quietly dropping you back)
| |
| 11:44 | <johnny> welcome!
| |
| 11:51 | ogra_cmpc, so.. any word on "Cannot open display"
| |
| 11:52 | <cliebow> .xsession-errors tell you anything?
| |
| 11:52 | <johnny> that's what it told me :)
| |
| 11:52 | <cliebow> heh
| |
| 11:53 | ;-]
| |
| 11:53 | <ogra_cmpc> johnny, not atm, sorry, busy with release stuff
| |
| 11:53 | <johnny> yeah i figured
| |
| 11:53 | good luck with that buddy :)
| |
| 12:01 | <Gadi> johnny: using DIRECTX?
| |
| 12:03 | <johnny> no
| |
| 12:03 | even has joined #ltsp | |
| 12:05 | <Gadi> have you added any scripts to /etc/X11/Xsession.d/ on the server?
| |
| 12:05 | mccann has quit IRC | |
| 12:05 | <johnny> for example?
| |
| 12:08 | <Gadi> ill take that as a no
| |
| 12:08 | :)
| |
| 12:08 | best bet is to log into the thin client's console (tty1) and take a look at the ldm.log
| |
| 12:09 | deavid has quit IRC | |
| 12:09 | <johnny> ls /etc/X11/Sessions/
| |
| 12:09 | Gnome Xsession
| |
| 12:09 | is that what you mean?
| |
| 12:09 | <Gadi> no no
| |
| 12:09 | I mean the Xsession.d dir
| |
| 12:09 | <johnny> i dont' have one
| |
| 12:09 | this ian't ubuntu/debian
| |
| 12:09 | <Gadi> what distro?
| |
| 12:09 | <johnny> gentoo
| |
| 12:09 | <Gadi> ah
| |
| 12:10 | check that your ssh server allows X forwarding
| |
| 12:11 | as a test, you can run "ssh -X user@server xterm" from another computer
| |
| 12:11 | you may need to run an "xhost +" before that
| |
| 12:11 | to allow xterm access to your display
| |
| 12:13 | mccann has joined #ltsp | |
| 12:13 | <johnny> i win!
| |
| 12:13 | muh2000 has joined #ltsp | |
| 12:13 | <johnny> ltsp is boot
| |
| 12:13 | <cliebow> and you did??
| |
| 12:13 | <johnny> it was the ssh forwarding
| |
| 12:13 | from boot to login == success
| |
| 12:16 | stest1 has joined #ltsp | |
| 12:16 | <stest1> hi :)
| |
| 12:16 | <Gadi> see what happens when you build from source :P
| |
| 12:16 | stest1 has quit IRC | |
| 12:16 | <even> anyone uses rdesktop to use windows applications?!
| |
| 12:17 | <johnny> even, i assume so :)
| |
| 12:17 | but not me
| |
| 12:17 | <Gadi> plenty of people
| |
| 12:17 | <johnny> no windows around here
| |
| 12:17 | Gadi, what do you mean
| |
| 12:17 | that stest1 was me :)
| |
| 12:17 | i just wanted to test xchat
| |
| 12:17 | <Gadi> johnny: about what?
| |
| 12:17 | <johnny> something about building from src
| |
| 12:18 | <Gadi> ah, ssh's defaults when built from source are to be the tightest security - aka no X forwarding
| |
| 12:18 | :)
| |
| 12:18 | <johnny> yes.. gentoo adheres to upstream in most cases
| |
| 12:19 | ok.. so now i just have to fix some things
| |
| 12:19 | <even> i'm trying use ltsp to replace one of the 6 windows server. but we need open some windows apps with rdesktop. anyone uses that!?
| |
| 12:19 | <johnny> even, you're doing it in a way that won't break any licenses right?
| |
| 12:20 | <even> johnny: no, we have all licences. we are a education instution
| |
| 12:20 | <Gadi> even: are you having trouble?
| |
| 12:20 | whats the question?
| |
| 12:21 | <even> Gadi: yes, when i close the app, the rdesktop windows keeps open a grey window. that is sucks for a newbie user
| |
| 12:21 | <Gadi> there's a flag for that
| |
| 12:21 | there is also seemless windows
| |
| 12:22 | <gvy> ogra_cmpc, warren, brought up 23 clients yesterday -- 23 kde3s, 23 firefoxen and 1 oowriter would eat up 1200M RAM on the server
| |
| 12:22 | and just cooked an iso which would boot on 16m out-of-box :) (5.1.2-based, already a microfork again)
| |
| 12:22 | mean "which would boot 16m-clients"
| |
| 12:23 | <ogra_cmpc> nice, but distro specific i guess
| |
| 12:23 | (you wont boot a default ubuntu or debian kernel in 16M)
| |
| 12:23 | <gvy> ogra_cmpc, well i've told you the better half of the hints before mentioning compcache :)
| |
| 12:23 | i can put up an image if anyone cares (2G dvd iso)
| |
| 12:24 | <ogra_cmpc> right, compcache is on my list to diuscuss with our kernel team in inrepid (8.10)
| |
| 12:24 | <johnny> ogra_cmpc, i'm booted up :)
| |
| 12:24 | <gvy> ogra_cmpc, btw it's automagically turned off in this case, too scarce memory
| |
| 12:24 | <ogra_cmpc> but that wont help the fact that the kernel alone needs 36M to uncompress
| |
| 12:24 | <gvy> and turned on when makes sense (e.g. i'm typing this on a 64m system)
| |
| 12:24 | <ogra_cmpc> or 32 ... i didnt measure for some time
| |
| 12:25 | <laga> ogra_cmpc: yay! great upload, thanks
| |
| 12:25 | <gvy> ogra_cmpc, how did they manage to build such a thoroughly broken kernel in the first place? :)
| |
| 12:25 | <ogra_cmpc> and i fear none of the distros apart from gentoo probably will have such a kernel
| |
| 12:25 | laga, thanks for all the patches :)
| |
| 12:26 | * gvy . o O ( altlinux != gentoo :) | |
| 12:26 | <johnny> patches ?
| |
| 12:26 | <ogra_cmpc> gvy, ubuntu kernels try to cover as much HW as they can ... we pay for that
| |
| 12:26 | <gvy> ogra_cmpc, but... initramfs?..
| |
| 12:26 | <ogra_cmpc> i'm planning for a lowram kernel setup for next release (even though not with focus on ltsp it will help here too)
| |
| 12:26 | indradg has joined #ltsp | |
| 12:26 | <ogra_cmpc> gvy, initramfs ??? what about it
| |
| 12:27 | <gvy> ogra_cmpc, well one can take modules along with the kernel
| |
| 12:27 | and stick only minor part of those in initrd/initramfs still
| |
| 12:27 | <ogra_cmpc> sure, but no autodetection etc
| |
| 12:27 | <gvy> and then mount the rest over nfs. :)
| |
| 12:27 | <ogra_cmpc> i'm not willing to go back to ltsp 4.x and maintain a monolithic unupdateable blob
| |
| 12:27 | <gvy> or probably nbd, didn't yet get there
| |
| 12:28 | ogra_cmpc, well we don't maintain mono blobck :)
| |
| 12:28 | <ogra_cmpc> the focus of ltsp5 is that the distros care for the packages
| |
| 12:28 | <gvy> well i saw you were busy, so probably son't take up time
| |
| 12:28 | <ogra_cmpc> i.e. kernel updates and security fixes have to come from the sdistro
| |
| 12:28 | <gvy> ogra_cmpc, *yeah*, and thanks for that :-)
| |
| 12:29 | i think we better will gather time for next merge-up so to speak :)
| |
| 12:29 | i was presented with somewhat more tight timeframe wish on a school project
| |
| 12:29 | <ogra_cmpc> that means the debian, ubuntu and fedora kernel teams would have to agree to support a lowram kernel package we can use
| |
| 12:29 | <gvy> say twice as short as i thought..
| |
| 12:29 | <dberkholz> ogra_cmpc: ubuntu doesn't do module hotplugging only for available hw? that´s interesting
| |
| 12:29 | well, coldplugging i guess
| |
| 12:29 | <ogra_cmpc> dberkholz, ubuntu uses whatever udev and the kernel detect
| |
| 12:29 | <gvy> ogra_cmpc, well there's already altlinux one published (but it's i586)
| |
| 12:30 | http://sisyphus.ru/srpm/Sisyphus/kernel-image-led-tc/get [ru]
| |
| 12:30 | <dberkholz> ogra_cmpc: ok. so you´re not loading say, 20 different sata chipset modules, you´re just loading the one detected on that system?
| |
| 12:31 | <ogra_cmpc> gvy, that doesnt really help the ltsp5 spirit it will break in horroid ways to use such a kernel on ubuntu or debian for example
| |
| 12:31 | <johnny> dberkholz, why should we do that? :)
| |
| 12:31 | <ogra_cmpc> for such an image we need to get the respective meintainers to agree
| |
| 12:31 | <laga> dberkholz: at least in the initramfs, only network related modules are included AFAIK. haven't verified though :)
| |
| 12:31 | <gvy> ogra_cmpc, i think that so far only patches and .config would be more or less universally useful :-(
| |
| 12:31 | <ogra_cmpc> dberkholz, xactly
| |
| 12:32 | <johnny> dberkholz, so.. i'm in :)
| |
| 12:33 | <dberkholz> johnny: awesome!
| |
| 12:33 | <ogra_cmpc> dberkholz, as laga said, the netboot initramfs only ships modules for NICs so there it cant detect anything else (udev does that later if run from its initscript)
| |
| 12:34 | <gvy> and probably shouldn't
| |
| 12:34 | <ogra_cmpc> ??
| |
| 12:35 | you mean udev ?
| |
| 12:35 | <johnny> dberkholz, we still have to worry about the kernel
| |
| 12:35 | bootup will take forever with all the modules genkernel adds
| |
| 12:36 | <ogra_cmpc> johnny, cant you hand over a list with modules to that tool ?
| |
| 12:36 | <johnny> not yet
| |
| 12:36 | <gvy> "anything else", yes
| |
| 12:36 | <dberkholz> johnny: oh, we´ll figure something out. we should make genkernel be smart about hotplugging if it´s not yet.
| |
| 12:37 | <ogra_cmpc> gvy, how will users get their HW working then ?
| |
| 12:37 | <dberkholz> johnny: if you post your stuff, i could try it out and take a closer look
| |
| 12:37 | <johnny> i'll be posting it as soon as i can
| |
| 12:37 | today/tonight
| |
| 12:37 | <gvy> ogra_cmpc, well, ours do somehow... i didn't look in our mkinitrd for quite a time, led did
| |
| 12:37 | <johnny> i need a patch in baselayout i think
| |
| 12:37 | or rather openrc
| |
| 12:37 | <gvy> but generally it's surprisingly ribust
| |
| 12:37 | <dberkholz> johnny: for?
| |
| 12:37 | <gvy> *robust
| |
| 12:38 | <johnny> or maybe i don't need it now, i thought i did, during the testing phase
| |
| 12:38 | * gvy runs home: don't break what's working :-) | |
| 12:38 | <johnny> gvy you should try to be more inclusive
| |
| 12:38 | <gvy> johnny, i try...
| |
| 12:38 | * johnny thinks ogra is doing a great job at that | |
| 12:38 | <ogra_cmpc> gvy, i want that exotic soundcard that is supported by any ubuntu system to work as well as the sata controller that has the ide cdrom attached ....
| |
| 12:39 | * johnny thanks ogra | |
| 12:39 | <ogra_cmpc> thats not just solved through hardcoding a set of modules in the intrd
| |
| 12:39 | <gvy> ogra_cmpc, we do udev with minimal ruleset off initramfs and another one from startup scripts
| |
| 12:39 | the latter can have way more modules and rules too
| |
| 12:39 | <ogra_cmpc> right, the same as we do
| |
| 12:39 | <gvy> and can be shot down to save ram for X
| |
| 12:40 | ogra_cmpc, well then i don't really know the difference :-) set the image to push out of the office...
| |
| 12:40 | <ogra_cmpc> initramfs simply has a limited set of modules available ... so it doesnt matter which rules yu have
| |
| 12:40 | that saves us from fiddling with distro rules
| |
| 12:40 | <johnny> dberkholz, take a look at bootmisc
| |
| 12:41 | deavid has joined #ltsp | |
| 12:41 | <johnny> down past # Make sure our X11 stuff have the correct permissions
| |
| 12:41 | <ogra_cmpc> for the one run from initscripts we use the default ubuntu setup simply to be sure to have the same level of HW support and to not have to fiddle with distro rules
| |
| 12:41 | <dberkholz> looking
| |
| 12:41 | <johnny> do we need that stuff? it seems like we do to me
| |
| 12:41 | i could be totally wrong tho
| |
| 12:42 | <dberkholz> johnny: which part? there´s 3 things after that comment
| |
| 12:42 | <johnny> all of it
| |
| 12:42 | <gvy> sorry, gotta run, should pick up mommy on the way
| |
| 12:42 | <johnny> any of it
| |
| 12:42 | <dberkholz> johnny: yeah we want that
| |
| 12:42 | <johnny> then we have to patch the beginning of start()
| |
| 12:43 | <gvy> ogra_cmpc, if you'd have some merging time any day, ping me please?
| |
| 12:43 | <dberkholz> hmm... yeah that test is inappropriate
| |
| 12:43 | <johnny> it should check /var/run /mtp directly
| |
| 12:43 | <dberkholz> it should test the places it wants to write
| |
| 12:43 | <johnny> /tmp*
| |
| 12:43 | yes
| |
| 12:43 | <gvy> bb folks!
| |
| 12:43 | * gvy is away: ~ | |
| 12:52 | <johnny> dberkholz, quickstart needs one more patch, to use the /etc/timezone file instead if /etc/conf.d/clock , which isn't in openrc anyways
| |
| 12:53 | <dberkholz> johnny: /etc/timezone doesn´t exist on my system
| |
| 12:53 | johnny: did you mean /etc/localtime?
| |
| 12:53 | <johnny> it falls back on /etc/conf.d/clock
| |
| 12:53 | no
| |
| 12:53 | take a look at pkg_config in timezone-data
| |
| 12:54 | that discussion just took place on #gentoo-desktop, scroll up a bit
| |
| 12:56 | <dberkholz> i´m not convinced of that answer
| |
| 12:56 | <johnny> well pkg_config shows it plainly
| |
| 12:56 | <dberkholz> it doesn´t have any comment indicating when it´s expected to be used
| |
| 12:57 | <johnny> /etc/init.d/clock is gone from openrc
| |
| 12:57 | only hwclock is there, and it has no TIMEZONE
| |
| 12:57 | tux_440volt has joined #ltsp | |
| 12:58 | <dberkholz> hmm. that was just added a couple of weeks ago..
| |
| 12:59 | <johnny> dberkholz, it seems to be added for openrc specifically
| |
| 12:59 | i've seen other such things in ebuilds
| |
| 13:01 | chup has joined #ltsp | |
| 13:01 | <dberkholz> openrc doesn´t do anything with it, though...
| |
| 13:01 | <johnny> i think it was left out on purpose, i don't know why
| |
| 13:03 | <dberkholz> ah, found it.
| |
| 13:03 | <johnny> oh?
| |
| 13:03 | <dberkholz> in the openrc ebuild
| |
| 13:03 | chupa has quit IRC | |
| 13:03 | <dberkholz> it grabs the timezone from conf.d/clock, if it exists, and echoes it to /etc/timezone
| |
| 13:03 | ogra_cmpc has quit IRC | |
| 13:03 | even has quit IRC | |
| 13:04 | <johnny> ok.. i'll hold off on my quickstart patch then
| |
| 13:04 | until openrc is in a stage3
| |
| 13:13 | ogra_cmpc has joined #ltsp | |
| 13:24 | mhterres has quit IRC | |
| 13:25 | mhterres has joined #ltsp | |
| 13:33 | Skarmeth has joined #ltsp | |
| 13:36 | elisboa has joined #ltsp | |
| 13:38 | Q-FUNK has joined #ltsp | |
| 13:46 | viking-ice has joined #ltsp | |
| 13:46 | jammcq has quit IRC | |
| 13:48 | tux_440volt has quit IRC | |
| 13:57 | gvy_ has joined #ltsp | |
| 14:05 | makghosh has quit IRC | |
| 14:10 | indradg is now known as idg|playing_mahj | |
| 14:12 | generalsnus has joined #ltsp | |
| 14:13 | idg|playing_mahj is now known as indradg | |
| 14:13 | plamengr has quit IRC | |
| 14:14 | makghosh has joined #ltsp | |
| 14:16 | Q-FUNK has quit IRC | |
| 14:17 | gvy has quit IRC | |
| 14:27 | <muh2000> VOLUME = 100 still an accepted value in lts.conf?
| |
| 14:31 | Q-FUNK has joined #ltsp | |
| 14:33 | <generalsnus> Ok, ive managed to get our edubuntu clients authenticate against our w2k3 server(Joined domain). But is it possible to have roaming profiles or somthing similar? lot of users are going to use the klients and we want a roaming profile.. or atlest a folder where they can save in, we would also need to deny them saving anything on the client then
| |
| 14:41 | chupa has joined #ltsp | |
| 14:43 | chup has quit IRC | |
| 14:45 | Skarmeth has left #ltsp | |
| 14:47 | <warren> Q-FUNK: blacklist patch rejected? did they state why?
| |
| 14:47 | <Q-FUNK> warren: that the PCI parsing patch ought to be good enough
| |
| 14:48 | ajax ventures that with PCI parsing being fixed, the blacklist is superfluous and not so well written anyhow, in his opinion
| |
| 14:49 | Egyptian[Home1 has joined #ltsp | |
| 14:49 | <Q-FUNK> he doesn't reject the idea of a blacklist, just the proposed implementation
| |
| 14:49 | indradg is now known as haumaukhau | |
| 14:49 | * Gadi coughs - seems it was the emulator that was poorly written *cough* | |
| 14:50 | <Q-FUNK> Gadi: *cough* you figured that out already? *cough*
| |
| 14:50 | :-P
| |
| 14:50 | <Gadi> :)
| |
| 14:50 | <Q-FUNK> well, yes, x86emu was not thorough or good
| |
| 14:50 | but they decided that the best way to expose bugs would be to release it, then brace for the flood of bug reports
| |
| 14:51 | haumaukhau is now known as indradg | |
| 14:56 | Egyptian[Home] has quit IRC | |
| 14:57 | cliebow has quit IRC | |
| 15:05 | indradg is now known as idg|dinner | |
| 15:18 | idg|dinner is now known as indradg | |
| 15:19 | spectra2 has joined #ltsp | |
| 15:19 | spectra2 has quit IRC | |
| 15:27 | K_O-Gnom has quit IRC | |
| 15:29 | TelnetManta has quit IRC | |
| 15:32 | Q-FUNK has quit IRC | |
| 15:40 | indradg is now known as idg|Zzzzz | |
| 15:43 | makghosh is now known as makghosh|afk | |
| 15:54 | johnny_ has joined #ltsp | |
| 16:03 | <wwx> with xorg possibe to inverse mouse cursor moving direction with xrandr or something like that?
| |
| 16:05 | generalsnus has quit IRC | |
| 16:05 | mhterres has quit IRC | |
| 16:07 | generalsnus has joined #ltsp | |
| 16:07 | deavid has quit IRC | |
| 16:09 | chup has joined #ltsp | |
| 16:09 | chupa has quit IRC | |
| 16:11 | <sutula> wwx: Some input drivers allow you to do that
| |
| 16:12 | May be other ways I don't know of
| |
| 16:12 | <wwx> mhm, thanks, going to investigate
| |
| 16:14 | <sutula> wwx: I got this working for an Elo touchscreen, had to do that with that particular input driver
| |
| 16:14 | <wwx> but with input_mouse?
| |
| 16:16 | * sutula looks | |
| 16:18 | <sutula> wwx: That particular InputDevice (from the xorg.conf file) was Driver "elographics"
| |
| 16:18 | It allowed me to set MinX, MaxX, ...
| |
| 16:18 | Reversing them reversed the mouse (so that it worked right for my display unit
| |
| 16:19 | <wwx> thanks for hint
| |
| 16:19 | viking-ice has quit IRC | |
| 16:20 | viking-ice has joined #ltsp | |
| 16:25 | bobby_C has joined #ltsp | |
| 16:30 | muh2000 has quit IRC | |
| 16:42 | Guaraldo has joined #ltsp | |
| 16:44 | Guaraldo has left #ltsp | |
| 16:46 | <klausade> generalsnus: got any further with ldap and Skolelinux?
| |
| 16:48 | <generalsnus> we decided to auth against w2k3 AD instead
| |
| 16:49 | was easy with "Likewise"
| |
| 17:00 | <johnny_> dberkholz, you interested in working on the patch for openrc?
| |
| 17:00 | <dberkholz> johnny_: oh, i'll bug him about it now
| |
| 17:00 | <johnny_> he told me what to he wanted
| |
| 17:01 | http://rafb.net/p/2kE4zP58.html
| |
| 17:01 | that's the simple version
| |
| 17:01 | <Uber> johnny_: what you need is to make that test into a function
| |
| 17:01 | <Uber> and then split the script into chunks where each chunk deals with one bit
| |
| 17:01 | <Uber> so you can do something like
| |
| 17:01 | <Uber> if mount_write /var/run; then clean_var_run; fi
| |
| 17:04 | <dberkholz> oh sure, shouldn't take a minute
| |
| 17:12 | <johnny_> i'll get the rest of the stuff uploaded tonight when i get a minute
| |
| 17:12 | after work
| |
| 17:13 | <dberkholz> johnny_: try something like http://dev.gentoo.org/~dberkholz/ltsp/bootmisc
| |
| 17:13 | untested
| |
| 17:15 | muh2000 has joined #ltsp | |
| 17:24 | generalsnus has quit IRC | |
| 17:42 | J45p3r has joined #ltsp | |
| 17:45 | chup has quit IRC | |
| 17:58 | slidesinger has quit IRC | |
| 18:03 | bobby_C has quit IRC | |
| 18:05 | xwind has joined #ltsp | |
| 18:09 | xwind has quit IRC | |
| 18:27 | muh2000 has quit IRC | |
| 18:33 | <johnny_> dberkholz, http://rafb.net/p/7NuUZW18.html
| |
| 18:33 | one by uberlord himself
| |
| 18:36 | there won't be a release until .2.3 tho
| |
| 18:37 | or whatever version is next
| |
| 18:38 | muh2000 has joined #ltsp | |
| 18:43 | <dberkholz> johnny_: doesn't account for /etc not being writable for /etc/nologin
| |
| 18:49 | mccann has quit IRC | |
| 18:51 | staffencasa has quit IRC | |
| 18:51 | <johnny_> why don't you join #gentoo-base
| |
| 18:51 | and talk to uberlord yourself :)
| |
| 18:51 | since i'm at work
| |
| 18:59 | TelnetManta has joined #ltsp | |
| 18:59 | Gadi has left #ltsp | |
| 19:04 | Dalran has joined #ltsp | |
| 19:13 | ^^Kilroy has joined #LTSP | |
| 19:40 | J45p3r has quit IRC | |
| 19:40 | makghosh|afk has quit IRC | |
| 21:52 | mccann has joined #ltsp | |
| 21:58 | jammcq has joined #ltsp | |
| 22:01 | pleaseandthankyo has joined #ltsp | |
| 22:01 | johnny_ has quit IRC | |
| 22:02 | <pleaseandthankyo> is there a good diet softwares? like for a diabetes guy or a healthy living diet software for person who has heart d eases?
| |
| 22:17 | pleaseandthankyo has quit IRC | |
| 22:42 | mccann has quit IRC | |
| 22:42 | <johnny> too bad he left
| |
| 22:42 | cuz i just saw something for diebetics
| |
| 22:42 | diabetics*
| |
| 22:46 | mccann has joined #ltsp | |
| 22:47 | mccann has joined #ltsp | |
| 23:08 | mccann_ has joined #ltsp | |
| 23:22 | mccann has quit IRC | |
| 23:42 | idg|Zzzzz is now known as indradg | |