UbuntuIRC / 2008 /02 /09 /#edubuntu.txt
niansa
Initial commit
4aa5fce
[11:49] <stgraber> ogra: We may have found the bug for iTalc :) (at least Tobias found a QRgb mistake in his code)
[11:50] <laga> ogra: ping
[13:09] <stgraber> RichEd-1: italc now works
[13:09] <stgraber> RichEd-1: upstream tells me he'll release a new version later in the day and I hope to have it ready for inclusion by Monday
[13:10] * RichEd-1 cheers for stgraber and Tobias for identifying & fixing
[13:10] <stgraber> RichEd-1: so oliver will have till Thursday to fill the MIR and have it moved to main (if he doesn't want to fill a FF exception too)
[13:10] * RichEd-1 cheers twice & jumps up & down at the news that it's on the inclusion path
[13:10] <RichEd-1> w00t
[13:11] <RichEd-1> stgraber: where can I get a featutre list for the version you are working on ?
[13:11] <RichEd-1> and some high level info on how it works ...
[13:11] <RichEd-1> i.e. teacher loads app on normal PC desktop
[13:11] <stgraber> RichEd-1: http://italc.sourceforge.net/ seems to be a good start
[13:12] <stgraber> RichEd-1: the client part will be starting by default on the clients and the server part will be available to the member of the admin group
[13:12] <stgraber> RichEd-1: starting it will show the classroom UI with all the computers and the possible actions at the top of the window
[13:12] <RichEd-1> and version # of the code you are working on so i can make sure i have the correct documentation feature set ?
[13:13] <stgraber> 1.0.6 which will be released today, functions are identitical as 1.0.3, 1.0.4 and 1.0.5 as those were only bugfixes
[13:13] <RichEd-1> (go on .. that level of detail is perfect ... it's for the suits in Intel who will never touch iTalc but want to tell people that "teacher admin" is available
[13:13] <RichEd-1> also, w.r.t. the component that runs on the client:
[13:13] <RichEd-1> #1 does it do any display on the local client
[13:14] <RichEd-1> i.e. will the smaller screen be any problem ?
[13:14] <stgraber> no, it's completely hidden so no problem client side
[13:14] <stgraber> the only problem is if one wants to run the teacher part on a CMPC
[13:14] <RichEd-1> #2 any other potential issues with runnning this on the classmate
[13:14] <RichEd-1> i think ogra should test it asap on his notebook, taking over his cmpc
[13:15] <RichEd-1> stgraber: nah ... the teacher needs a workstation end of story
[13:15] <RichEd-1> the teacher is an adult
[13:15] <stgraber> well, I haven't a CMPC or any of those small screen device around (still waiting for my EEE) but it shouldn't cause any problem as a client
[13:15] <RichEd-1> make a note to remind ogra to test when you "release the code to him"
[13:15] <RichEd-1> i'll mail him as well
=== RichEd-1 is now known as RichEd
=== highvoltage is now known as highvoltage-bot
=== highvoltage-bot is now known as highvoltage
[17:47] <achandrashekar> Hello.I am setting up two edubuntu servers (Athlon X2 2.2GHz's,2GB of mem,450GB HDD) to power a 60 person lab type environment. Id like more information how to cluster the boxes to power up the lab. Is there any type of how to for this?
[17:51] <johnny> anybody here testing out hardy?
[17:57] <achandrashekar> anybody on the clustering?
[18:12] <achandrashekar> rephrase of my original question anyone familiar with openmosix framework for clustering and how it works with edubuntu?
[18:13] <johnny> i've never heard anybody in this channel mention openmosix
[18:13] <johnny> is that software still in devel?
[18:14] <Paladine> yodles
[18:15] <achandrashekar> its a mechanism for clustering for ltsp and supposedly you need a recompilation of kernel to get it going. So I figured that "might" be what is being considered in terms of clustering
[18:15] <Paladine> I have been assigned the task of setting up an Open Learning environment so I was looking for some advice
[18:15] <achandrashekar> but
[18:15] <achandrashekar> Id like to get a set up going such that I CAN cluster two boxes, and/or use the idling clients to work in a powerful cluster
[18:17] <johnny> it might be too advanced for most of the folks in this channel
[18:17] <johnny> try #ltsp
[18:17] <johnny> as well
[18:17] <achandrashekar> hmm..is there a development thread of edubuntu channel??
[18:17] <johnny> and/or #ubuntu-server .. if there is one
[18:17] <johnny> your question isn't development atm
[18:17] <achandrashekar> ie..i have read something about a development group working on this.
[18:17] <johnny> oh
[18:18] <johnny> well, email those folks to see where they hang out i guess
[18:18] <achandrashekar> okay..will have to join "yet another" list...lol
[18:18] <achandrashekar> ;)
[18:18] <Paladine> is there an edubuntu mailing list? it might be easier for me to write everything in an email than to try and do it over irc
[18:19] <achandrashekar> i "might" just get busy with trying to set it up myself and see what happens.
[18:19] <achandrashekar> and then seeing if there is any interest.
[18:19] <stgraber> Paladine: http://lists.ubuntu.com => edubuntu-user or edubuntu-devel
[18:19] <Paladine> thanks stgraber
[18:20] <stgraber> achandrashekar: I never heard of someone doing clustering with LTSP, some have multiple LTSP servers and do load balacing (that's the official and supported way)
[18:22] <achandrashekar> stgraber: http://openmosix.sourceforge.net/ltsp-omr4-1.html
[18:23] <stgraber> achandrashekar: is MOSIX working with 2.6 ? last I heard of it it was only for 2.4
[18:24] <achandrashekar> yep...that seemed to be the issue. That was going to be the next lead in question..to see "how far" it had come..but it appears...it has been a slow road.
[18:25] <stgraber> I never saw openmosix running on 2.6, that will be your major issue with Ubuntu/Edubuntu as I doubt you can run a 2.4 kernel on Ubuntu (mainly due to udev)
[18:27] <johnny> if openmosix hasn't advanced to a 2.6 kernel by now
[18:27] <johnny> it sounds like a dead project
[18:28] <stgraber> lastest release is from December 9, 2004
[18:28] <johnny> yep.. dead
[18:28] <johnny> achandrashekar, i wouldn't put much effort into that method then
[18:28] <johnny> there must be a reason nobody has picked it up in so long
[18:29] <stgraber> johnny: it simulates a big SMP computer, so if your software aren't threaded it's useless (it was the main problem with MOSIX and that kind of clustering)
[18:30] <achandrashekar> I see..hmm..i guess the question what the best methodology is to use the "combined" horespower of two systems to power up an ltsp env, And if there is a how to for that for edubuntu.. I guess your previous post about where to look is best?
[18:30] <achandrashekar> best question to ask about methodology i mean...
[18:30] <achandrashekar> sorry poor grammar
[18:31] <achandrashekar> i "could" power up 1/3 of the lab with one, and the other, and yet another..
[18:32] <achandrashekar> but i figured id never have ALL nodes up at the same time..so why not come up with a solution to aggregrate the power..i suppose its a bit of wishful thinking.
[18:32] <achandrashekar> nodes used that is..not "up" ...my bad.
[18:49] <Paladine> k email sent to the list hopefully there will be hundreds of people all wanting to share their experiences with me :)
[19:00] <pem725> I have a quick question about root logins to the F1 terminal
[19:01] <pem725> I followed the directions detailed in the handbook but when I try to login - even after rebooting the server and my thin client - the root login and password are not accepted
[19:01] <pem725> any ideas why this might be happening?
[19:04] <stgraber> pem725: what exactly did you do ?
[19:04] <pem725> I did sudo chroot /opt/ltsp/i386
[19:04] <stgraber> pem725: I don't have the handbook in mind and it's not always up to date (depending on where you are reading it from)
[19:04] <pem725> and then I issued a passwd command
[19:04] <pem725> entered a password and confirmed it
[19:04] <johnny> and then you rebuild the image?
[19:04] <pem725> I am assuming I did this for the root user
[19:05] <pem725> yes
[19:05] <johnny> that's all i had to do
[19:05] <pem725> hmmm
[19:05] <johnny> was set it and rebuild the image
[19:05] <pem725> I might need to step through the process slower
[19:05] <pem725> when you login, do you login as root?
[19:06] <johnny> login to the thinclient?
[19:06] <johnny> yes
[19:06] <pem725> yes
[19:06] <pem725> great
[19:06] <pem725> k
[19:06] <johnny> but only from the terminal
[19:06] <pem725> I will try it again
[19:06] <pem725> yes, thanks.
[19:06] <pem725> the reason I need to do this is to confirm that my via thin client is using the via xorg driver
[19:06] <pem725> my video performance is so bad on my thin client
[19:06] <pem725> horrible in fact
[19:07] <stgraber> pem725: so there is an easier way to do that than setting a root pw
[19:07] <pem725> excellent
[19:07] <pem725> I'm all eyes
[19:07] <stgraber> pem725: create a lts.conf file in /var/lib/tftpboot/ltsp/i386/
[19:08] <pem725> ok
[19:08] <pem725> I have one for my via client already
[19:08] <stgraber> ok, so you already have a general or default section at the top ?
[19:08] <pem725> yes
[19:09] <pem725> is there something I can add to the already existing default section?
[19:09] <stgraber> just add : SCREEN_01=shell
[19:09] <pem725> great
[19:10] <pem725> will that allow me to login as my regular user and then sudo su?
[19:10] <johnny> no
[19:10] <stgraber> no, it'll open a root shell
[19:10] <pem725> ok
[19:10] <pem725> I guess I am confused here.
[19:11] <pem725> so the shell will already have root logged in
[19:11] <pem725> correct?
[19:11] <stgraber> yes
[19:11] <pem725> wonderful
[19:11] <pem725> thank you very much. Once I get this done, I will tackle sound and local devices
[19:11] <pem725> thanks for your help.
[19:13] <stgraber> np
[19:14] <stgraber> pem725: here it doesn't seem to work with _01, I'm trying with _02 now
[19:17] <stgraber> pem725: ok, it works fine with _02, it'll open a root shell on tty2 (F2)
[19:20] <johnny> stgraber, do you have a hardy install?
[19:21] <stgraber> johnny: yes running in Xen
[19:21] <stgraber> and my lappy on Hardy (but Ubuntu, not edubuntu)
[19:21] <johnny> can you test something for me? it should only take 2 minutes
[19:21] <stgraber> sure
[19:21] <johnny> that's fine..doesn't matter if it's edubuntu or whatever afaik
[19:21] <johnny> what version of gnome-settings-daemon does it currenlty include?
[19:22] <stgraber> stgraber@castiana:~$ dpkg -l | grep gnome-settings-daemon
[19:22] <stgraber> ii gnome-settings-daemon 2.21.90.2-0ubuntu2
[19:22] <johnny> aha.. perfect
[19:22] <johnny> do you have xnest?
[19:22] <stgraber> yes
[19:22] <johnny> ok.. su to some user and then run
[19:23] <johnny> some other user not currently running
[19:23] <johnny> and then
[19:23] <johnny> Xnest -ac :1
[19:23] <johnny> err
[19:23] <johnny> Xnest -ac :1 &
[19:23] <johnny> and then DISPLAY="0:1"/usr/libexec/gnome-settings-daemon
[19:23] <johnny> that's all
[19:24] <stgraber> Xnest doesn't work as my test user doesn't seem to have the right to access my X instance
[19:26] <johnny> thus -ac ?
[19:27] <johnny> that's what i had to do
[19:27] <stgraber> test@castiana:~$ Xnest -ac :1
[19:27] <stgraber> No protocol specified
[19:27] <stgraber> Fatal server error:
[19:27] <stgraber> Unable to open display ":0.0".
[19:27] <johnny> that makes no sense..
[19:28] <stgraber> it makes sense, a software (xnest), tries to access my X server without my tester user having the right to do so
[19:28] <stgraber> -ac is valid for everything running in the Xnest, not on the main X server
[19:28] <johnny> hmm.. i didn't do anything special on my server to make it work
[19:29] <stgraber> if you do :
[19:29] <stgraber> su yourtestuser
[19:29] <stgraber> DISPLAY=:0 xeyes
[19:29] <stgraber> does it work ?
[19:29] <stgraber> (it shouldn't)
[19:29] <johnny> su -
[19:29] <johnny> is what i did
[19:29] <johnny> sure.. but that's why i do 0:1
[19:29] <johnny> which i seem to have rights to do :)
[19:31] <stgraber> hmm, in fact I was wrong, :1 is my current X server (or at least I have a :1 server running)
[19:31] <stgraber> though trying with :2 doesn't work either
[19:31] <stgraber> test@castiana:/home/stgraber$ Xnest -ac :2
[19:31] <stgraber> Fatal server error:
[19:31] <stgraber> Could not create server lock file: /tmp/.X2-lock
[19:32] <stgraber> hmm, let me clean a bit my /tmp
[19:32] <johnny> i know in the dbus xorg file i only have display0 and display1
[19:32] <johnny> so i'm sticking with :0 and :1 atm