IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 14 May 2007 (all times are UTC)
| 00:10 | bgomes has joined #ltsp | |
| 00:41 | bgomes has quit IRC | |
| 01:13 | MoL0ToV has joined #ltsp | |
| 01:59 | Appiah has quit IRC | |
| 02:28 | plamengr has joined #ltsp | |
| 02:44 | vanya has joined #ltsp | |
| 02:56 | MoL0ToV has quit IRC | |
| 03:03 | MoL0ToV has joined #ltsp | |
| 03:58 | pere_away is now known as pere | |
| 04:18 | jsgotangco has joined #ltsp | |
| 04:31 | cyberorg has joined #ltsp | |
| 04:34 | str4nd_ has quit IRC | |
| 04:35 | muh2000 has quit IRC | |
| 04:36 | Jenna has joined #ltsp | |
| 04:40 | muh2000 has joined #ltsp | |
| 04:41 | jsgotangco has quit IRC | |
| 04:56 | bobby_C has joined #ltsp | |
| 05:01 | plamengr has quit IRC | |
| 05:02 | plamengr has joined #ltsp | |
| 05:35 | ogra has joined #ltsp | |
| 05:45 | bronze has joined #ltsp | |
| 05:48 | freet15 has quit IRC | |
| 05:49 | Jenna has quit IRC | |
| 05:52 | Jenna has joined #ltsp | |
| 05:55 | federicoco has joined #ltsp | |
| 05:55 | <federicoco> hi all....have you rdesktop screen.d script for ltsp5 ?
| |
| 05:58 | <ogra> federicoco, i think Gadi has them, wait for him to show up
| |
| 05:59 | <cyberorg> hi ogra
| |
| 06:00 | <federicoco> ok thank u ogra
| |
| 06:01 | <cyberorg> ogra: would it be possible to incorporate couple of changes to ldm that are required for suse?
| |
| 06:02 | <ogra> sure
| |
| 06:02 | as long as they are writte in a way that they dont break for the other distros
| |
| 06:02 | *written
| |
| 06:03 | <cyberorg> greeter = subprocess.Popen(['/usr/lib/ltsp/greeters/gtk'], suse 64 bit lib path is /usr/lib64
| |
| 06:03 | <ogra> so you want to support 64bit thin clients ?
| |
| 06:04 | <cyberorg> yeah
| |
| 06:04 | <ogra> thats a very very rare case ...
| |
| 06:04 | <cyberorg> people are requesting PPC support too :)
| |
| 06:04 | <ogra> can you write a proper patch that checks if you are in a suse chroot and changes that value dynamically ?
| |
| 06:05 | <cyberorg> for lib64?
| |
| 06:05 | <ogra> you will have to redo half the package for using luib64 ...
| |
| 06:05 | dynamic installation path setting during package build for different arches etd
| |
| 06:05 | *etc
| |
| 06:06 | <cyberorg> yeah, that we have we can specify --libdir=%{_libdir} macro during rpm build time
| |
| 06:06 | <ogra> right
| |
| 06:07 | so add a patch to the python code to do something similar based on $VENDOR ...
| |
| 06:07 | and give me the branch url
| |
| 06:07 | <cyberorg> ogra: only if i could code :) let me look up what i tried
| |
| 06:08 | <ogra> ($VERNDOR should be filled with the Distributor ID output from lsb_release)
| |
| 06:08 | <cyberorg> greeter = subprocess.Popen(['/usr/lib/ltsp/greeters/gtk /usr/lib64/ltsp/greeters/gtk'] i tried something like that
| |
| 06:09 | <ogra> thats ugly
| |
| 06:09 | create a self.greeter variable and put it in there instead ... fill that variable based on the lsb_release output
| |
| 06:10 | <cyberorg> ok, i'll try doing that, if not will get someone who can code python, will send you patch as soon as i get em
| |
| 06:11 | another change required is Xsession location, it is in /usr/X11/xdm/Xsession not /usr/X11/Xsession which is hardcoded in ldm too
| |
| 06:11 | i've done the same ugly bit there too :)
| |
| 06:11 | <ogra> same thing as above
| |
| 06:12 | <cyberorg> so far that is all i have changed in ltsp code, ltspfs works out of box :)
| |
| 06:12 | <ogra> you should add a check_vendor() function that sets such variables
| |
| 06:13 | <cyberorg> we can have more generic solution with pkgconfig checks?
| |
| 06:13 | <ogra> others might need different paths as well so we can add all distros in one place and dont have to change it inside the rest of the code
| |
| 06:14 | <cyberorg> pkg-config --variable=libdir somepackage
| |
| 06:15 | <ogra> we use lsb_relese everywhere and i'm not really keen on introducing autotools in ldm ...
| |
| 06:16 | <cyberorg> ah, ok
| |
| 06:18 | LSB Version: core-2.0-noarch:core-3.0-noarch:core-2.0-ia32:core-3.0-ia32 desktop-3.1-ia32 desktop-3.1-noarch:graphics-2.0-ia32:graphics-2.0-noarch:graphics-3.1-ia32:graphics-3.1-noarch
| |
| 06:18 | this is what i get from lsb_release
| |
| 06:19 | <ogra> lsb_release -a|grep ID
| |
| 06:19 | <cyberorg> Distributor ID: SUSE LINUX
| |
| 06:19 | <ogra> right
| |
| 06:20 | match on that
| |
| 06:20 | <cyberorg> ok, will try something with that
| |
| 06:22 | can you tell me why you use delayed_mounter instead of mounting /media as soon as user logs in?
| |
| 06:23 | <ogra> delayed_mounter is used for devices that exist before a logging in
| |
| 06:23 | floppies etc
| |
| 06:23 | *a user is loggin in
| |
| 06:24 | or local HDs
| |
| 06:24 | <cyberorg> mounting /media will give everything including all the devices mounted before or after user loggs in
| |
| 06:24 | <ogra> how do you mount the ones that exist before and get no udev trigger ?
| |
| 06:25 | udev handles such devices *before* the ssh tunnel and socket exist
| |
| 06:25 | <cyberorg> all the devices auto mounted shows up in /media/* before or after login
| |
| 06:25 | <ogra> yes
| |
| 06:25 | but the initital trigger to mount them is handled by udev
| |
| 06:26 | for devices that are detected during boot you wont get such a trigger
| |
| 06:26 | <cyberorg> yes, but they will still be in /media
| |
| 06:26 | <ogra> delayed_mounter queues them up on boot and sends the trigger during login
| |
| 06:26 | <cyberorg> we just mount ltspfs clientip:/media /mountpoint
| |
| 06:26 | <ogra> we're talking about two different things here ...
| |
| 06:27 | i think
| |
| 06:27 | <cyberorg> thing is delayed_mounter mounts each devices as they become available, mounting /media will mount everything regardless
| |
| 06:27 | <ogra> no it wont
| |
| 06:28 | where would the device mounts on the client come from ?
| |
| 06:29 | <cyberorg> does all the automounts show up in /media on ubuntu?
| |
| 06:30 | <ogra> in /media/$USER/<devicename>
| |
| 06:30 | <cyberorg> that is what is different on suse everything is mounted in /media/devicename
| |
| 06:31 | <ogra> dont do that
| |
| 06:31 | that will give you hundrets of devices on the desktops of all users
| |
| 06:31 | (if you have hundrets of users)
| |
| 06:31 | <cyberorg> you could try mounting /media/$USER and see if everything under that shows up
| |
| 06:31 | 100 user on a client? nope
| |
| 06:31 | <ogra> but that part will get redone completely soon anyway
| |
| 06:32 | 100 users on a *server*
| |
| 06:32 | <cyberorg> cleint sees only his local devices
| |
| 06:32 | <ogra> we have way bigger setups out ther
| |
| 06:32 | so you have patched gnome/kde already ?
| |
| 06:33 | they usually show *everything* mounted in media
| |
| 06:34 | <cyberorg> we have been seeing everything mounted in /media for a long time
| |
| 06:34 | <ogra> https://blueprints.launchpad.net/ubuntu/+spec/ltspfs-virtual-hal-devices will change everything on the server side
| |
| 06:34 | right not only the devices the user owns
| |
| 06:35 | waht we want is that we only see the devices belonging to one user ... i.e. the ones we can access
| |
| 06:35 | <cyberorg> for ltsp i think we should mount localmedia in /home/$USER/ltspmounts to keep stuff separate from the server mounts
| |
| 06:35 | <ogra> no
| |
| 06:36 | we wont touch the users home in any case for system level stuff
| |
| 06:36 | thats the most evil thing you can do
| |
| 06:36 | the system (even suse i guess) has functionallity to handle thst, use it
| |
| 06:36 | *that
| |
| 06:37 | <cyberorg> wouldn't putting in /media be a nightmare with 100s of devices in /media
| |
| 06:37 | <ogra> not if you have per user subdirs in there
| |
| 06:37 | <cyberorg> we dont
| |
| 06:38 | <ogra> then your ltspfs implementation is broken ... ltspfsmounter should create and dlete them
| |
| 06:39 | but as i said, all that is obsolete with the above spec
| |
| 06:39 | <cyberorg> ogra: yeah, i have not used ltspfsmouunter, just tried manual ltspfs cleint:/media /home/$USER/ltspmounts
| |
| 06:39 | <ogra> hal will take care of it in the next release
| |
| 06:39 | <cyberorg> next release, when? :)
| |
| 06:39 | <ogra> oct
| |
| 06:40 | the release for which we disussed implementaion specs the last ten days :)
| |
| 06:40 | <cyberorg> hmm, so can't meet openSUSE 10.3 release, will be around that time
| |
| 06:43 | let me see if i can get suse/novell developers interested in this then we can have something to test soon
| |
| 06:43 | <ogra> make them take a look at the spec
| |
| 06:44 | <cyberorg> yeah, couple of guys working on thin client have said they will work with us :)
| |
| 07:14 | jsgotangco has joined #ltsp | |
| 07:26 | jsgotangco has quit IRC | |
| 07:26 | MM2_ has joined #ltsp | |
| 07:27 | * kaminski-ltsp-br is back | |
| 07:27 | <kaminski-ltsp-br> ;)
| |
| 07:28 | joebake3 has joined #ltsp | |
| 07:28 | joebake1 has joined #ltsp | |
| 07:29 | joebake2 has quit IRC | |
| 07:29 | joebaker has quit IRC | |
| 07:45 | Blinny has joined #ltsp | |
| 07:45 | J45p3r_ has joined #ltsp | |
| 07:49 | gepatino has joined #ltsp | |
| 07:52 | <Blinny> http://wiki.ltsp.org/twiki/bin/view/Ltsp/Sound seems to be a bit outdated.. is there a newer source for sound info?
| |
| 07:55 | <ogra> jammcq, around ?
| |
| 07:56 | Blinny has quit IRC | |
| 08:02 | joebaker has joined #ltsp | |
| 08:04 | fxrsliberty has quit IRC | |
| 08:04 | <jammcq> ogra: ping
| |
| 08:05 | fxrsliberty has joined #ltsp | |
| 08:08 | <ogra> jammcq, i got udp running ...
| |
| 08:08 | <jammcq> oh?
| |
| 08:08 | and...... ?
| |
| 08:08 | <ogra> 9secs for our testfile on the inital run
| |
| 08:08 | <jammcq> whoa
| |
| 08:08 | <ogra> 0.522 on the second
| |
| 08:08 | <jammcq> no way
| |
| 08:08 | 100mb ?
| |
| 08:09 | which client?
| |
| 08:09 | <ogra> i have to solder a cable on the powerbrick first, so i did my testing with the classmate for now ...
| |
| 08:09 | <jammcq> oh
| |
| 08:09 | that's why it's so fast
| |
| 08:09 | it's no the 200mhz unit
| |
| 08:09 | * jammcq was excited for a few seconds | |
| 08:10 | <jammcq> did you try the classmate with tcp ?
| |
| 08:10 | federicoco has quit IRC | |
| 08:10 | <ogra> not yet
| |
| 08:10 | <jammcq> so we have nothing to compare that 9 secs to
| |
| 08:10 | <ogra> i'm not really sure why my chroot works and highvoltages didnt
| |
| 08:11 | i'll build a new chroot during the day with the defaults ...
| |
| 08:11 | <jammcq> something was messed up on his chroot
| |
| 08:11 | <ogra> the one i'm testing with is my classmate flashing root i wont modify anyway
| |
| 08:11 | <jammcq> it got too late in the day to look at it
| |
| 08:11 | Blinny has joined #ltsp | |
| 08:11 | <ogra> yeah
| |
| 08:11 | but it seems we'Re on the right track
| |
| 08:12 | <jammcq> I sure hope so
| |
| 08:12 | <ogra> iirc i tested with the classmate before and got something around 23secs
| |
| 08:12 | but i might remember that wrongly
| |
| 08:12 | <highvoltage> perhaps there was something funky installed on my laptop
| |
| 08:12 | (although, afaik it's all pretty much standard)
| |
| 08:14 | <jammcq> hmm
| |
| 08:14 | cliebow has joined #ltsp | |
| 08:16 | Jenna has quit IRC | |
| 08:17 | MM2 has joined #ltsp | |
| 08:22 | <Blinny> I've recently upgraded from K12v5 to K12v6. Sound has become broken in the process for multiple terminals.. I think the cards (pci) are still being found, though. I've tried Gideon's recommendations (http://wiki.ltsp.org/twiki/bin/view/Ltsp/WorkInProgress#esd_ALSA_sound_on_LTSP_4_2) but still no love. Is there a step-by-step method of checking where I'm going wrong?
| |
| 08:22 | gepatino has quit IRC | |
| 08:22 | * jammcq will be back online when he gets to the office | |
| 08:22 | jammcq has quit IRC | |
| 08:25 | <Blinny> I have two testing terminals. One is a SBLive! that is found correctly (start=shell and lspci finds it), the other is a Ensoniq ES1371 that also is found the same way.
| |
| 08:28 | The sound servers on both terminals report listening on port 16001
| |
| 08:33 | MM2_ has quit IRC | |
| 08:39 | <Blinny> Damnit.
| |
| 08:39 | Blinny has quit IRC | |
| 08:44 | irule has joined #ltsp | |
| 08:45 | Blinny has joined #ltsp | |
| 08:45 | <Blinny> Okay, good. Now I have flash sound working, but not xmms/totem
| |
| 08:47 | gstreamer-properties outputs file when I switch to esd
| |
| 08:49 | It looks like xmms is configured for ALSA
| |
| 08:53 | Okay, xmms-esd solved that.
| |
| 09:02 | gnome-sound-properties is set up to use esd
| |
| 09:02 | Why on earth is totem soundless??
| |
| 09:09 | <ogra> are you using the xine backend for it by chance ?
| |
| 09:09 | jammcq has joined #ltsp | |
| 09:09 | <Blinny> Si.
| |
| 09:10 | Do I need to gstreamer it instead?
| |
| 09:10 | <ogra> indeed or make xine use esd
| |
| 09:11 | <Blinny> Thank you ogra I'll work along this route.
| |
| 09:12 | <fxrsliberty> hello all
| |
| 09:13 | could some one help me recompile our initrd to work better with the LTSP 150's
| |
| 09:13 | <jammcq> why does the initrd need to be recompiled ?
| |
| 09:15 | <ogra> udp 10.9M/s , real: 9.420s
| |
| 09:15 | tcp 11.6M/s , real: 8.887s
| |
| 09:15 | jammcq, ^^^
| |
| 09:15 | :(((
| |
| 09:15 | <jammcq> ogra: hmm, not much better
| |
| 09:15 | <ogra> no, rather the opposite
| |
| 09:15 | <jammcq> actually, a bit worse
| |
| 09:15 | which doesn't make sense
| |
| 09:16 | <ogra> i'll do tests on the 200Mhz thingie ...
| |
| 09:16 | <jammcq> I talked to mdz on friday, and he believed that udp would speed things up alot
| |
| 09:16 | <ogra> but need to solder around a bit first
| |
| 09:16 | well, mount shows proto=tcp/udp as expected
| |
| 09:16 | <fxrsliberty> it doesn't seem to be able to load all the way , but yours works and ogra
| |
| 09:16 | <ogra> i havent had wireshark running yet
| |
| 09:16 | <jammcq> fxrsliberty: lots of people are using T-1540'
| |
| 09:17 | err
| |
| 09:17 | Lots of people are using T-150's and LTSP-5
| |
| 09:17 | <fxrsliberty> the initrd built with kiwi seems to be different
| |
| 09:17 | <jammcq> fxrsliberty: how big is it?
| |
| 09:18 | <fxrsliberty> I have 60 of them and ltsp 4.2 and ubuntu's is great the new one for opensuse is not so good yet
| |
| 09:18 | <jammcq> so you need kiwi help, not ltsp help
| |
| 09:19 | <fxrsliberty> i need to no what goes on inside yours that ours is missing
| |
| 09:19 | it's 17.9 meg
| |
| 09:19 | <jammcq> wow
| |
| 09:19 | <ogra> oof
| |
| 09:19 | <jammcq> what the hell is in that thing.... the kitchen sink?
| |
| 09:19 | <ogra> tcos ?
| |
| 09:19 | <fxrsliberty> i don't no
| |
| 09:19 | prolly some default
| |
| 09:20 | <jammcq> either way, the T-150 should be able to load it
| |
| 09:20 | <ogra> even though it might take ages to unpack
| |
| 09:20 | and tftp ...
| |
| 09:20 | <fxrsliberty> i've left it over night
| |
| 09:20 | <jammcq> I can't imagine it taking more than 5 seconds to load
| |
| 09:20 | sbalneav has joined #ltsp | |
| 09:20 | <jammcq> 17mb isn't all that big
| |
| 09:20 | !s
| |
| 09:20 | <ltspbot> jammcq: "s" is Scotty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
| |
| 09:20 | <sbalneav> Morning all!
| |
| 09:21 | <ogra> !s
| |
| 09:21 | <ltspbot> ogra: "s" is Scotty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
| |
| 09:21 | <sbalneav> !monkeys
| |
| 09:21 | <ltspbot> sbalneav: "monkeys" is SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS
| |
| 09:21 | <ogra> heh
| |
| 09:21 | <sbalneav> Everybody make it back home safe and sound?
| |
| 09:21 | <ogra> fsvo safe
| |
| 09:21 | i had a horrible trip
| |
| 09:21 | <sbalneav> What happened?
| |
| 09:22 | <cyberorg> jammcq: hi, this is what is in the suse netboot initrd http://pastebin.ca/487480
| |
| 09:22 | <ogra> it started with me having typoed my reservation ... so i was booked on the 11th
| |
| 09:22 | and had to buy a new ticket for the 12th ...
| |
| 09:22 | <jammcq> ogra: whoa
| |
| 09:22 | <ogra> then both of my flight were delayed
| |
| 09:23 | <cyberorg> is there a list of things that goes in ltsp initrd somewhere that we can have a look?
| |
| 09:23 | <ogra> so i missed the last tram in hannover to get to the train station ... weher i in turn missed the train then
| |
| 09:23 | <sbalneav> ugh
| |
| 09:23 | <ogra> and for extra bonus my luggage got lost
| |
| 09:23 | <sbalneav> Suckage!
| |
| 09:23 | <jammcq> meanwhile, sbalneav and I had a great trip back
| |
| 09:23 | <ogra> luggage will arrive tomorrow ... and i'm sitting in my living room now ... so all fine in the end
| |
| 09:24 | <jammcq> well.... at least as far as Detroit
| |
| 09:24 | <sbalneav> The rest was good too.
| |
| 09:24 | Exit row on DTW->MSP
| |
| 09:24 | 1st Cl. MSP->YGW
| |
| 09:25 | <fxrsliberty> hey guys, what package on the 4.2 iso holds the initrd?
| |
| 09:25 | <jammcq> kiwi includes reiserfs !!!! ummm
| |
| 09:25 | <ogra> ouch
| |
| 09:25 | <jammcq> fxrsliberty: 4.2 iso ?
| |
| 09:25 | <fxrsliberty> ltsp 4.2
| |
| 09:25 | <ogra> i bet it includes *everything* with thst huge size
| |
| 09:25 | <jammcq> iso ?
| |
| 09:25 | <fxrsliberty> yup
| |
| 09:25 | <jammcq> fxrsliberty: the kernel package includes the iso
| |
| 09:26 | <fxrsliberty> was thinking of unpacking your initrd and looking around
| |
| 09:26 | <jammcq> it's got a lot less than the kiwi has
| |
| 09:26 | <fxrsliberty> ok so let me ss in yours and make ours better
| |
| 09:26 | <jammcq> ss ?
| |
| 09:26 | <ogra> ubuntus only has the NIC and keyboard drivers
| |
| 09:27 | <fxrsliberty> see
| |
| 09:27 | do you load a tmpfs?
| |
| 09:27 | <jammcq> fxrsliberty: first off, we don't include glibc
| |
| 09:27 | yeah
| |
| 09:27 | <ogra> we do in ubuntu
| |
| 09:27 | <jammcq> glibc is HUGE for an initramfs
| |
| 09:27 | ogra: you include glibc ?
| |
| 09:27 | <ogra> but only the libc6.so file
| |
| 09:27 | yes
| |
| 09:27 | <jammcq> ah
| |
| 09:28 | <ogra> all stuff is linked against klibc anyway i think libc6 is for compatiblity reasons ...
| |
| 09:28 | just sitting here
| |
| 09:28 | <jammcq> looks like kiwi is using all the BIG stuff, rather than the slimmed down klibc and busybox stuff
| |
| 09:28 | <ogra> or probably udev makes use of it ... i didnt bother looking yet
| |
| 09:29 | but even includion everything that shouldnt prevent you from booting ... i guess the problem lies elsewhere
| |
| 09:29 | *including
| |
| 09:29 | <jammcq> PXES was using a much larger initrd without problems
| |
| 09:30 | <ogra> yeah
| |
| 09:31 | <cyberorg> we want to figure exactly that out, what is missing in spite of kitchen sink being in there :)
| |
| 09:32 | <ogra> see above you only need the NIC modules ...
| |
| 09:33 | and if you want input support in busybox the keyboard drivers
| |
| 09:33 | federicoco has joined #ltsp | |
| 09:33 | <ogra> oh207, and nfs indeed
| |
| 09:33 | meh
| |
| 09:33 | and nfs indeed
| |
| 09:33 | <cyberorg> we got "drivers/net/*"
| |
| 09:34 | "fs/nfs/*"
| |
| 09:34 | <ogra> right that should suffice
| |
| 09:34 | just dont put all the other stuff in
| |
| 09:34 | <cyberorg> ok, will try that
| |
| 09:34 | <fxrsliberty> what about the tmpfs
| |
| 09:37 | <cyberorg> fxrsliberty: try taking sfuff out one by one and see what works ;)
| |
| 09:37 | Blinny has quit IRC | |
| 09:38 | <ogra> we mount the nfsroot in /root and change to that at the end of the initramfs booting
| |
| 09:38 | in ubuntu/debian with run-init ... 4.2 used a binary called switchroot ...
| |
| 09:39 | tmpfs isnt used in ubuntu/debian ... 4.2 twiddles with that though
| |
| 09:39 | (isnt used == isnt used for /, udev uses a tmpfs indeed)
| |
| 09:41 | <fxrsliberty> that what i think the problem is > udev can't find a place to work
| |
| 09:41 | pardon my noob speak
| |
| 09:41 | <ogra> udev should care for that itself ...
| |
| 09:42 | our udev initscript creates the necessary tmpfs mounts
| |
| 09:43 | <fxrsliberty> does this init script do something that needs attention > http://pastebin.ca/487516
| |
| 09:43 | <ogra> mount -n -t tmpfs -o mode=0755 udev /dev
| |
| 09:43 | mkdir -m 0700 -p /dev/.static/dev
| |
| 09:43 | mount -n --move /etc/udev /dev/.static/dev
| |
| 09:43 | thats what we use
| |
| 09:45 | <fxrsliberty> sop this needs changing ? #======================================
| |
| 09:45 | # 1) Mounting local file systems
| |
| 09:45 | #--------------------------------------
| |
| 09:45 | mount -t proc proc /proc
| |
| 09:45 | mount -t sysfs sysfs /sys
| |
| 09:45 | mount -t devpts devpts /dev/pts
| |
| 09:45 | closeKernelConsole
| |
| 09:46 | <ogra> well that depends on your udev implementation ... i'D talk to the udev maintainers in suse
| |
| 09:51 | vanya has quit IRC | |
| 10:00 | efra has joined #ltsp | |
| 10:04 | <joebaker> I had another XCMCP shutdown of the server on Friday. To temporarily remedy this, I went into the Login preferences and blanked out the commands for shutdown, restart, suspend & hibernate. Now when they click the gnome-logout button it immediately shuts down the user's session.
| |
| 10:04 | ogra: sorry to hear of your travel difficulties.
| |
| 10:05 | <ogra> joebaker, i'm used to that :)
| |
| 10:06 | http://www.grawert.net/weblog.cgi/2006/11/25/2006-11-25_15:51:10_Zen_and_the_art_of_travelling_to_eastern_europe.html
| |
| 10:06 | thats a typical trip of mine ...
| |
| 10:14 | Avatara has quit IRC | |
| 10:14 | rafaelmf has joined #ltsp | |
| 10:14 | Gadi has joined #ltsp | |
| 10:21 | Blinny has joined #ltsp | |
| 10:26 | nicoAMG has joined #ltsp | |
| 10:28 | Blinny has quit IRC | |
| 10:35 | Blinny has joined #ltsp | |
| 10:40 | joebake3 has quit IRC | |
| 10:40 | joebake1 has quit IRC | |
| 10:47 | sepski has joined #ltsp | |
| 10:50 | <Blinny> ogra: Hrm. in xine, 'esd' is not an option for sound output. I see alsa, oss, file, but no esd.
| |
| 10:56 | Blinny has quit IRC | |
| 11:00 | tiagovaz has joined #ltsp | |
| 11:00 | pere is now known as pere_away | |
| 11:02 | muh2000 has quit IRC | |
| 11:08 | Guaraldo has joined #ltsp | |
| 11:10 | MoL0ToV has quit IRC | |
| 11:13 | rafaelmf has quit IRC | |
| 11:15 | rafaelmf has joined #ltsp | |
| 11:16 | spectra has joined #ltsp | |
| 11:19 | federicoco has quit IRC | |
| 11:19 | tiagovaz has quit IRC | |
| 11:26 | rafaelmf has quit IRC | |
| 11:29 | marek82 has joined #ltsp | |
| 11:44 | marek82 has quit IRC | |
| 11:44 | marek82 has joined #ltsp | |
| 11:46 | avlis has joined #ltsp | |
| 11:47 | cyberorg has left #ltsp | |
| 12:01 | rjune has left #ltsp | |
| 12:03 | plamengr has quit IRC | |
| 12:20 | muh2000 has joined #ltsp | |
| 12:24 | Blinny has joined #ltsp | |
| 12:26 | SBNet has joined #ltsp | |
| 12:38 | gonzaloaf_work has joined #ltsp | |
| 12:39 | <gonzaloaf_work> hi, why when I use the w command to see what users I have logged I just see root and one or two ltsp users? why I dont see all of them or none of them, why just one? I use ltsp4.2
| |
| 12:42 | avlis has quit IRC | |
| 12:45 | <Blinny> gonzaloaf_work: Try this: last |grep still
| |
| 12:46 | <gonzaloaf_work> Blinny, why does the command w is useless in this case?
| |
| 12:53 | <Blinny> gonzaloaf_work: I don't know exactly why; I recall seeing this on the LTSP mailing list recently.
| |
| 12:54 | <gonzaloaf_work> Blinny, let me check
| |
| 12:58 | mistik1 has quit IRC | |
| 13:01 | * kaminski-ltsp-br is away: tchau | |
| 13:10 | <dan__t> 'morning.
| |
| 13:10 | avlis has joined #ltsp | |
| 13:21 | mistik1 has joined #ltsp | |
| 13:23 | marek82 has quit IRC | |
| 13:25 | <ogra> jammcq, measuring in virtualbox proves the former values right ...
| |
| 13:25 | udp 2.4M/s , real: 42.763s
| |
| 13:25 | tcp 3.6M/s , real: 28.656s
| |
| 13:26 | intrestingly i seem to get no nfs caching in the virtual machine ... the second and third run dont differ much
| |
| 13:37 | fernando1 has quit IRC | |
| 13:38 | <jammcq> ogra: are you getting no caching on both tcp AND udb ?
| |
| 13:38 | s/udb/udp/
| |
| 13:38 | mhterres has joined #ltsp | |
| 13:39 | <mhterres> afternoon
| |
| 13:39 | <ogra> yep
| |
| 13:39 | <jammcq> hmm
| |
| 13:39 | <ogra> i think thats a vm thing
| |
| 13:39 | <jammcq> ogra: I'm assuming thats on the e2300 ?
| |
| 13:39 | <ogra> nope
| |
| 13:39 | virtualbox
| |
| 13:39 | <jammcq> oh
| |
| 13:39 | have you tried the 2300 yet?
| |
| 13:39 | <ogra> i havent digged up my soldering iron yet to have a proper cable on the power brick
| |
| 13:40 | <jammcq> why do you need to solder something?
| |
| 13:40 | don't you have an adapter?
| |
| 13:40 | <ogra> not one euro->us ... only us->euro
| |
| 13:40 | <jammcq> bummer
| |
| 13:41 | <ogra> i rarely use us wallplugs over here :)
| |
| 13:42 | <jammcq> yeah, I rarely us euro wall plugs on this side of the pond
| |
| 13:42 | <ogra> indeed :)
| |
| 13:44 | Q-FUNK has joined #ltsp | |
| 13:50 | <ogra> jammcq, which libc do you use in the 4.2 initramfs ?
| |
| 13:50 | kaminski-ltsp-br has quit IRC | |
| 13:50 | <jammcq> uclibc
| |
| 13:50 | I think
| |
| 13:50 | <ogra> dietlibc ? klibc ?
| |
| 13:50 | ah
| |
| 13:50 | <jammcq> hmmmmmm
| |
| 13:51 | kaminski-ltsp-br has joined #ltsp | |
| 13:51 | <jammcq> yeah, it's uclibc
| |
| 13:51 | * kaminski-ltsp-br is away: tchau | |
| 13:51 | <ogra> i wonder how i can get that into the 5 one without rebuilding all binaries ...
| |
| 13:52 | <jammcq> well, I don't see how it could matter for performance. possibly before the switchroot is done, it could matter, but that's fairly early in the booting
| |
| 13:53 | <ogra> right
| |
| 13:53 | but nothing we tired yet gained us anything
| |
| 13:53 | *tried
| |
| 13:53 | i'm slowly running out of ideas where to look
| |
| 13:54 | <jammcq> yeah, it worries me
| |
| 13:54 | <ogra> me too
| |
| 13:54 | <jammcq> how about that message about the cpu might be 'unstable'
| |
| 13:54 | I don't think we see that with ltsp-4.2
| |
| 13:54 | <ogra> you are to tied to the e2300 :)
| |
| 13:55 | <jammcq> it's not just me. soon, lots of people will be tied to it
| |
| 13:55 | <ogra> that bug is there generally ... its just exposed more on the 2300
| |
| 13:55 | <jammcq> but, does it put the kernel into a degraded mode or something?
| |
| 13:55 | <ogra> might be
| |
| 13:55 | but didnt you show that to BenC ?
| |
| 13:56 | <jammcq> no, I don't think we did
| |
| 13:56 | <ogra> in your first sessions
| |
| 13:56 | oh
| |
| 13:56 | <jammcq> lemme fire up my laptop, and try booting ltsp-4.2 on the e2300, and see if the message appears
| |
| 13:56 | <ogra> well, lets go to #ubuntu-kernel show them the exact message and see what they think :)
| |
| 13:57 | andrea has joined #ltsp | |
| 13:57 | <jammcq> good idea
| |
| 13:57 | bengoa has joined #ltsp | |
| 13:58 | <ogra> do you have the exact wording there ?
| |
| 13:58 | <andrea> hi, I install an ltsp server in edubuntu server 7.4; I create few user with the same name of the hostname of the client. I need to autologin them, can somebody help me? Please
| |
| 13:59 | mEDUXa has joined #ltsp | |
| 13:59 | <ogra> jammcq, hmm, looks like BenC took the day off or something ... i know i saw him around this morning
| |
| 14:00 | andrea, ldm cant do autologin yet ... you will need to resort to gdm which means local devices wont work
| |
| 14:00 | mEDUXa has quit IRC | |
| 14:01 | mEDUXa has joined #ltsp | |
| 14:02 | <andrea> i know nothing about perl, but i have a perl script that give me the correct username.
| |
| 14:02 | privet has quit IRC | |
| 14:02 | <ogra> well, look up how to set up autologin with gdm ...
| |
| 14:02 | <andrea> Is really hard to integrate this script in ldm?
| |
| 14:03 | <ogra> to switch edubuntus ltsp to XDMCP with gdm, you need to enable it in the gdmsetup tool ... and add the line "SCREEN_07=startx" to /opt/ltsp/i386/etc/lts.conf
| |
| 14:04 | yes, there is no way yet to do it in ldm
| |
| 14:04 | you need to fiddle a lot with ssh keys etc ...
| |
| 14:04 | mEDUXa has quit IRC | |
| 14:04 | <ogra> ldm is ssh based
| |
| 14:05 | sdm might be an option to do it via ssh, but i dont think thats documented anywhere you would have to study sdm a bit ...
| |
| 14:05 | fernando1 has joined #ltsp | |
| 14:05 | <ogra> and replace ldm with sdm in the client envoronemnt
| |
| 14:05 | *anvironment
| |
| 14:05 | gah
| |
| 14:05 | *environment
| |
| 14:05 | privet has joined #ltsp | |
| 14:06 | <jammcq> ogra: ltsp-4.2 also gives the message:
| |
| 14:06 | <ogra> aha
| |
| 14:06 | <jammcq> CPU: Vendor unknown, using generic init
| |
| 14:06 | <ogra> then its not it
| |
| 14:06 | <jammcq> CPU: Your system may be unstable
| |
| 14:07 | <ogra> i thought you said something like "4.2 has it too" when i asked first
| |
| 14:07 | <jammcq> that was too long ago for me to remember
| |
| 14:07 | <ogra> so thats not it either
| |
| 14:08 | <jammcq> welp, it appears that's not the problem
| |
| 14:09 | <ogra> i'll get a new usbdisk these days and check how slow a local boot from usb is ... you guys only tested the installer, right ?
| |
| 14:09 | <jammcq> right
| |
| 14:09 | <ogra> ok, i'll test with a copied chroot
| |
| 14:09 | <jammcq> k
| |
| 14:17 | SBNet has quit IRC | |
| 14:21 | Blinny has quit IRC | |
| 14:25 | andrea has quit IRC | |
| 14:27 | hansatuofd has joined #ltsp | |
| 14:28 | <privet> hi all
| |
| 14:29 | I need some tips/suggestions on fixing a problem I have with a specific terminal
| |
| 14:29 | running ltsp4.2 on fc4
| |
| 14:29 | X runs nicely
| |
| 14:29 | but I am unable to get a normal terminal to work...
| |
| 14:30 | the screen just goes into a mode where it looks like the resultion/vsync is to high
| |
| 14:30 | but it is not, since X is running fine
| |
| 14:30 | any suggestions/tips?
| |
| 14:31 | <jammcq> what do you mean by "normal terminal" ?
| |
| 14:32 | <privet> bash terminal
| |
| 14:32 | <jammcq> like xterm or gnome-terminal?
| |
| 14:32 | <privet> sorry: "shell"
| |
| 14:32 | <jammcq> or are you talking about setting 'SCREEN_01 = shell' ?
| |
| 14:33 | <privet> *exactly*
| |
| 14:33 | I have SCREEN_01 = startx
| |
| 14:33 | <jammcq> so, if you set SCREEN_01 = startx, everything works just fine. but if you set SCREEN_01 = shell, you have problems ?
| |
| 14:33 | <privet> and screen_02 shell
| |
| 14:33 | <jammcq> and when do you see the problem?
| |
| 14:34 | <privet> yes- startx is fine, but doing ALT-F2 to switch to the shell screen
| |
| 14:34 | <jammcq> ah
| |
| 14:34 | what kind of video card do you have?
| |
| 14:35 | * privet double checking that... | |
| 14:35 | Q-FUNK has quit IRC | |
| 14:36 | Q-FUNK has joined #ltsp | |
| 14:36 | <privet> jammcq: "nx cx 5530A chipset"
| |
| 14:36 | <jammcq> ah
| |
| 14:37 | too bad
| |
| 14:37 | <privet> the "pc" is a IB300
| |
| 14:37 | too bad? yes?
| |
| 14:37 | <jammcq> some video chipsets, including the geode 5530 have problems switching from graphics mode to character mode
| |
| 14:37 | <privet> hhhmmmmm
| |
| 14:37 | too bad (for me) then
| |
| 14:37 | <jammcq> yeah, unfortunately
| |
| 14:38 | <privet> I have the same issue when not running startx at all
| |
| 14:38 | when only running screen_01=shell
| |
| 14:38 | I can also boot slackware (from a local compactflash) and works just fine
| |
| 14:39 | <jammcq> what video driver is it using?
| |
| 14:39 | hmm, SCREEN_01 = shell is also giving you trouble?
| |
| 14:39 | that's odd
| |
| 14:39 | <privet> you see: without a local shell, how do I check what driver is it loading?
| |
| 14:40 | a5benwillis has joined #ltsp | |
| 14:41 | <moquist> jammcq: hey there
| |
| 14:41 | <jammcq> hey moquist
| |
| 14:41 | <cliebow> moquist:hey there
| |
| 14:41 | <moquist> jammcq: I didn't realize it last week, but it turns out that thin client boot time is very important to my schools. (Apparently we have low-end hardware I didn't know about.)
| |
| 14:42 | So I'm trying to figure out how to speed things up now. :)
| |
| 14:42 | <jammcq> moquist: using ltsp-4.2 or ltsp-5 ?
| |
| 14:43 | <cliebow> 4.2 Rocks
| |
| 14:43 | <moquist> jammcq: edubuntu feisty and edgy are both slow.
| |
| 14:43 | <jammcq> yeah, we spent all last week on this issue
| |
| 14:43 | still working on it now
| |
| 14:43 | <moquist> the TC I'm using right now takes 4.5 min to get LDM up and going.
| |
| 14:43 | <jammcq> :)
| |
| 14:43 | <moquist> jammcq: OK.
| |
| 14:43 | <jammcq> what kind of thin client?
| |
| 14:43 | <moquist> jammcq: yeah, I was going to get you those specs. hang on.
| |
| 14:44 | celeron 650Mhz
| |
| 14:44 | 128MB
| |
| 14:44 | <ogra> woah
| |
| 14:44 | <moquist> I think it's shared video memory
| |
| 14:44 | <jammcq> whoa, you are seeing that kind of problem on a 650 ?
| |
| 14:45 | <ogra> thats extremly slow for such HW
| |
| 14:45 | <moquist> yep
| |
| 14:45 | no kidding...so maybe we have something else slowing us down, if this is abnormal.
| |
| 14:45 | <ogra> for me usually everything >500Mhz with 128M boots around 90sec or less
| |
| 14:45 | <moquist> Our purchased, production TCs take 1-2 min to boot.
| |
| 14:46 | ogra: Hmmm.
| |
| 14:46 | <ogra> feisty on a 900Mhz system even around 60sec
| |
| 14:46 | <moquist> k
| |
| 14:46 | <jammcq> ltsp-4.2 on a 200Mhz system, around 41 secs :)
| |
| 14:46 | <ogra> (which is still to slow imho, but not really 4.5min)
| |
| 14:46 | <moquist> ogra: How can I watch the log during boot?
| |
| 14:46 | <jammcq> sorry, had to throw that in
| |
| 14:46 | <moquist> (so I can see what's taking so long)
| |
| 14:47 | <ogra> moquist, drop splash and quiet from the pxeconfig line
| |
| 14:47 | <moquist> ogra: k; will do
| |
| 14:47 | <cliebow> moquist:wanna take quiet splash out of pxelinux.xfg
| |
| 14:47 | hreh
| |
| 14:47 | <ogra> i assume its some hw they have builtin
| |
| 14:47 | are these refurbished desktops ?
| |
| 14:47 | <moquist> ogra: I don't think so. Just old.
| |
| 14:49 | <ogra> i have one compaq workstation with onboard scsi controller and some old ... that takes minutes but also on non ltsp boots
| |
| 14:50 | *old disks
| |
| 14:51 | <moquist> k. rebooting TC now. I only have one network jack here, so I'll be gone for a bit. :p
| |
| 14:51 | bellis has joined #ltsp | |
| 14:51 | <moquist> screw this. I'm going downstairs where I can have TWO network jacks. :)
| |
| 14:51 | hansatuofd has quit IRC | |
| 14:55 | <bellis> Anybody using ltsp5 with the jammin 125 from Diskless Workstations?
| |
| 14:56 | <moquist> starting kernel event manager took a while
| |
| 14:56 | starting the LTSP client is taking quite a while
| |
| 14:56 | <ogra> moquist, thats edgy, right ?
| |
| 14:56 | <moquist> ogra: yes
| |
| 14:56 | ogra: I can test with feisty instead. I'll go back to the office.
| |
| 14:56 | <ogra> remove the xfonts* packages from the chroot
| |
| 14:56 | <moquist> ogra: k
| |
| 14:57 | <ogra> leave only -base in place
| |
| 14:57 | that should speed ldm up
| |
| 14:57 | <moquist> xfontsel?
| |
| 14:57 | <ogra> for the "setting up ltsp client" step it helps to have a static xorg.conf for these clients, that skips the autodetection
| |
| 14:57 | <moquist> xfonts-utils?
| |
| 14:58 | <ogra> no -75dpi, 100dpi, scalable, misc etc etc
| |
| 14:58 | * moquist nods | |
| 14:58 | <ogra> x loads all of them to ram ... i dropped these packages completely from feisty
| |
| 14:58 | * moquist nods | |
| 14:58 | <moquist> er, --nodeps or something? it wants to remove xorg, which I want to keep...
| |
| 14:58 | <jammcq> bellis: for j-125's, it's still preferred to use LTSP-4.2
| |
| 14:58 | * moquist reads the man page | |
| 14:59 | <jammcq> there's been some reports of problems with ltsp-5 and J-125's
| |
| 14:59 | <ogra> moquist, you dont want to keep xorg
| |
| 14:59 | <bellis> jammcq, I figured, I was finally able to get it boot by running mknbi to generate a new netboot file
| |
| 14:59 | <jammcq> is it working ok for you?
| |
| 14:59 | <bellis> jammcq, but now getting the X to work correctly is being a bit difficult
| |
| 14:59 | <ogra> moquist, its just a metapackage
| |
| 14:59 | <bellis> jammcq, yeah its working
| |
| 15:00 | <moquist> ogra: I'm going to stop doing this on our edgy production server and go back to the feisty testing server. :)
| |
| 15:00 | <bellis> jammcq, if you go into /var/lib/tftpboot/ltsp/i386 (on fiesty)
| |
| 15:00 | <ogra> moquist, well, feisty has these changes
| |
| 15:00 | <moquist> ogra: It's feisty we really want to work with these clients, so I should just stop screwing around here.
| |
| 15:00 | <ogra> ah
| |
| 15:00 | <moquist> I just wanted to see how slow edgy was in comparison.
| |
| 15:00 | <ogra> well, static xorg will help a lot
| |
| 15:01 | <bellis> and run mknbi-linux --output=blah.nbi --rotdir="/dev/ram0" --append="rw" vmlinuz initrd.img
| |
| 15:01 | <ogra> and dropping the fonts as well
| |
| 15:01 | <moquist> ogra: K. back online in a bit...
| |
| 15:01 | <bellis> it boots each time
| |
| 15:01 | <ogra> bellis, the shipped nbi.img didnt work ?
| |
| 15:01 | <bellis> ogra, no it wouldnt boot it gave me a segment in reserved area error
| |
| 15:01 | <jammcq> ogra: I heard somebody else have a problem with the shipped nbi last week
| |
| 15:02 | <bellis> ogra, when I checked, the working ltsp 4 images were NBI format and the new images are ELF format
| |
| 15:02 | ogra, but when I created a new one using NBI it worked properly
| |
| 15:02 | <ogra> bellis, could you file a bug about that in launchpad and attach the steps you needed to it ?
| |
| 15:02 | i'll make sure it ges into gutsy
| |
| 15:02 | *goes
| |
| 15:02 | <bellis> ogra, sure thing, what is the link to launchpad?
| |
| 15:03 | <jammcq> https://launchpad.net/ubuntu/+source/ltsp/+filebug
| |
| 15:03 | <ogra> https://bugs.launchpad.net/ubuntu/+source/ltsp/+filebug
| |
| 15:03 | gah, jammcq bets me
| |
| 15:03 | *beats
| |
| 15:03 | :)
| |
| 15:03 | <jammcq> heh
| |
| 15:04 | xlyz has joined #ltsp | |
| 15:04 | gentgeen__ has joined #ltsp | |
| 15:09 | J45p3r_ has quit IRC | |
| 15:11 | ahz has joined #ltsp | |
| 15:12 | <moquist> ogra: in feisty, hardware detection took a long time, and setting up swapspace was slow
| |
| 15:12 | <ahz> I am trying to play videos via mplayer. When I do -ao null -vo x11 it works or -ao esd -vo null it works but -ao esd -vo x11 freezes on a blue screen.
| |
| 15:13 | <moquist> 15 seconds between the screen blanking (X starting) and seeing the "X" cursor
| |
| 15:14 | ogra: 60 seconds between seeing the "X" cursor and getting the LDM screen
| |
| 15:20 | cliebow has quit IRC | |
| 15:21 | mhterres has quit IRC | |
| 15:22 | <bellis> During client bootup, what process is generating the "0 value set 0 true..." messages?
| |
| 15:22 | mhterres has joined #ltsp | |
| 15:23 | privet has quit IRC | |
| 15:27 | ahz has quit IRC | |
| 15:29 | <bellis> Ahh, that seems to be when its reading the lts.conf file (I added some entries and noticed it added additional lines)
| |
| 15:33 | nf1 has joined #ltsp | |
| 15:34 | a5benwillis has quit IRC | |
| 15:34 | Q-FUNK has quit IRC | |
| 15:36 | <bellis> Yay! I have a working LTSP5 install with graphics
| |
| 15:37 | * moquist high-fives bellis | |
| 15:37 | <bellis> jammcq, in case you run into it, I had to set most of the X options in the lts.conf, it was not friendly to scanning
| |
| 15:38 | the best thing is that when you get it running right, you don't have to mess with it again for so long
| |
| 15:39 | we have a manufacturing facility where they use these, those terminals have been running non stop for 1-2 years except for power outages and no maint on them
| |
| 15:39 | <Gadi> hey, gang - if an iface in Ubuntu comes up as eth1_ifre what does that mean/
| |
| 15:40 | <jammcq> hmm
| |
| 15:40 | neat-o
| |
| 15:40 | bellis: can you show me what the settings are that you use?
| |
| 15:40 | efra has quit IRC | |
| 15:41 | SBNet has joined #ltsp | |
| 15:41 | <bellis> jammcq, this was with a Sony Multiscan 200sx as the monitor
| |
| 15:41 | privet has joined #ltsp | |
| 15:41 | <bellis> What is a good site to link text?
| |
| 15:42 | <jammcq> !pastebot
| |
| 15:42 | <ltspbot> jammcq: "pastebot" is The LTSP pastebot is at http://pastebot.ltsp.org. Please paste all text longer than a line or two to the pastebot, as it helps to reduce traffic in the channel. A link to the content will be pasted in the channel.
| |
| 15:42 | <jammcq> actually, it would be great if you'd put it on the wiki
| |
| 15:42 | on the 'clients' page
| |
| 15:42 | http://wiki.ltsp.org/twiki/bin/view/Ltsp/Clients
| |
| 15:42 | http://wiki.ltsp.org/twiki/bin/view/Ltsp/Clients#Jammin_125
| |
| 15:43 | <ltsppbot> "bellis" pasted "lts.conf settings for ltsp5 and jammin 125" (13 lines) at http://pastebot.ltsp.org/133
| |
| 15:43 | SBNet has quit IRC | |
| 15:45 | <bellis> jammcq, okay I will edit the wiki once the lock opens up
| |
| 15:45 | <jammcq> bellis: should be open now, I just released it
| |
| 15:52 | * jammcq is heading home for the day. | |
| 15:53 | jammcq has quit IRC | |
| 15:57 | sbalneav has quit IRC | |
| 15:59 | bellis has quit IRC | |
| 16:02 | Guaraldo has left #ltsp | |
| 16:06 | Q-FUNK has joined #ltsp | |
| 16:07 | mhterres has quit IRC | |
| 16:07 | J45p3r has joined #ltsp | |
| 16:14 | bengoa has quit IRC | |
| 16:34 | sepski has quit IRC | |
| 16:36 | <privet> when you specify a custom xorg.conf file using XF86CONFIG_FILE: is the value the absolute path to that file?
| |
| 16:36 | or should it be in a specific directory? (using ltsp4.2)
| |
| 16:56 | Gadi has quit IRC | |
| 16:58 | Q-FUNK has quit IRC | |
| 16:59 | muh2000 has quit IRC | |
| 17:02 | xlyz has left #ltsp | |
| 17:28 | nicoAMG has quit IRC | |
| 17:41 | alfred_ has joined #ltsp | |
| 17:41 | <alfred_> is there local device support in ltsp 5 on the clients?
| |
| 17:43 | <moquist> ogra: I'm building a kernel with Jim's config from here: http://librarian.launchpad.net/7039447/config_2.6.20-ltsp
| |
| 17:44 | ogra: Any hack I can put in place easily ("easy" is a checkbox requirement here) that gets the TCs booting more quickly will at least buy me time.
| |
| 17:45 | If I have a kernel .deb that I can point to ("Just install this in the chroot"), that'll be good enough.
| |
| 18:02 | bobby_C has quit IRC | |
| 18:22 | jammcq has joined #ltsp | |
| 18:30 | irule has quit IRC | |
| 18:31 | bobby_C has joined #ltsp | |
| 18:34 | bobby_C has quit IRC | |
| 18:34 | muh2000 has joined #ltsp | |
| 18:37 | cliebow has joined #ltsp | |
| 18:44 | a5benwillis has joined #ltsp | |
| 18:55 | davidj has joined #ltsp | |
| 18:55 | <davidj> jammcq: Ping
| |
| 18:56 | Hello, everyone. What's new?
| |
| 18:58 | <jammcq> hey
| |
| 18:58 | how you doin?
| |
| 19:00 | cliebow has quit IRC | |
| 19:10 | <moquist> jammcq: hey! you're back.
| |
| 19:10 | * moquist is still at work | |
| 19:10 | <moquist> jammcq: I'm trying out your kernel config from http://librarian.launchpad.net/7039447/config_2.6.20-ltsp
| |
| 19:11 | but I only gets a kernel panic so far :(
| |
| 19:11 | privet has quit IRC | |
| 19:18 | <jammcq> moquist: did you build a new initramfs to go with it?
| |
| 19:19 | <moquist> jammcq: I used make-kpkg with --initrd, so yes.
| |
| 19:19 | <jammcq> umm, no
| |
| 19:20 | umm
| |
| 19:20 | no
| |
| 19:20 | <moquist> eh? ok...
| |
| 19:20 | <jammcq> you need to do 'update-initramfs' inside the chroot
| |
| 19:20 | I think, anyway
| |
| 19:20 | <moquist> hmm.
| |
| 19:22 | jammcq: not enough to chroot and dpkg -i the .deb, and then ltsp-update-kernels from outside the chroot?
| |
| 19:22 | /var/lib/tftpboot/ltsp/i386/ has new initrd.img, nbi.img, and vmlinuz that I built.
| |
| 19:22 | <jammcq> umm, I'm an ltsp-4.2 guy, I really don't know what all you need to do for ltsp-5
| |
| 19:23 | <moquist> np; I don't really know, either. :)
| |
| 19:23 | it's a HUGE initrd - 22125683
| |
| 19:23 | * moquist isn't sure why | |
| 19:23 | <jammcq> heh
| |
| 19:23 | yer doing something wrong
| |
| 19:24 | <moquist> I took stock 2.6.20 (the linux-source-2.6.20 package), dropped your config in, did 'make oldconfig', and then the stuff I said above.
| |
| 19:24 | <jammcq> you need ogra
| |
| 19:24 | * moquist nods | |
| 19:32 | <moquist> jammcq: I have no idea what happened with the first one, but in a second attempt I used the same config (yours) and then modified it further myself (I took out wireless, etc.). That one is booting just fine.
| |
| 19:32 | jammcq: also, the initrd this time is 3044618 big instead of 22125683.
| |
| 19:37 | comfrey has joined #ltsp | |
| 19:37 | <comfrey> hey all...
| |
| 19:38 | for some reason, my boot on a 4.2 client is halting at "Enter runlevel:
| |
| 19:38 | i will add shell to screen_01
| |
| 19:38 | and see
| |
| 19:38 | <jammcq> sounds like you are using the wrong kernel
| |
| 19:39 | <comfrey> ah, good point.
| |
| 19:39 | <jammcq> like a ltsp-4.1 kernel with ltsp-4.2
| |
| 19:39 | <comfrey> i downgraded from 5
| |
| 19:39 | <moquist> jammcq: my kernel boots now, but it's slower than the first one 8\
| |
| 19:39 | <jammcq> comfrey: or a ltsp-5 kernel with ltsp-4.2
| |
| 19:39 | <comfrey> exactly
| |
| 19:40 | what is the kernel name for 4.2?
| |
| 19:40 | <jammcq> vmlinuz-2.6.17.8.ltsp-3
| |
| 19:41 | or something close to that
| |
| 19:41 | <comfrey> ok, i will check the dhcp.conf
| |
| 19:45 | ok, i think this will be better
| |
| 19:46 | ok, file not found...
| |
| 19:46 | a tftp thing
| |
| 19:46 | * comfrey checks | |
| 19:48 | <moquist> jammcq: do you know how much I would have to do to switch to XDMCP (instead of LDM+ssh) and get sound and USB drives working?
| |
| 19:48 | <jammcq> moquist: sorry, that's not yet supported
| |
| 19:48 | <moquist> jammcq: :)
| |
| 19:50 | jammcq: I didn't figure it was "supported", but I figured it wouldn't be *too* bad to copy a bunch of 4.2-ish things in to set env vars and the like, and to use X properties to pass usb mounts, etc. But maybe it *would* be too bad. :\
| |
| 19:50 | <jammcq> well, when you say it like that, it sounds easy. sadly, there's alot of plumbing to be done, to actually make it work. If it was easy, we would have already done it
| |
| 19:50 | <moquist> the greeter ('gtk') pegs the CPU on this laptop for 1:20 before I can log in.
| |
| 19:51 | jammcq: right.
| |
| 19:52 | ah, well.
| |
| 19:52 | * moquist heads upstairs with a K12LTSP Cd | |
| 19:52 | <jammcq> moquist:
| |
| 19:52 | you could just install ltsp-4.2 on ubuntu
| |
| 19:52 | <moquist> hmm.
| |
| 19:53 | I could install ltsp-4.2 on this very same edubuntu server.
| |
| 19:53 | <jammcq> that's what I'm saying
| |
| 19:53 | * moquist nods | |
| 19:54 | <comfrey> grrr. file not found
| |
| 19:57 | <moquist> jammcq: is grabbing the "Ubuntu Breezy" package for local dev support the right thing to do?
| |
| 19:57 | petre has joined #ltsp | |
| 19:58 | <jammcq> umm
| |
| 19:58 | <petre> evening all
| |
| 19:58 | <jammcq> sbalneav is your friend on that one
| |
| 19:58 | * moquist nods | |
| 20:00 | <moquist> jammcq: Ooooo. ltspadmin says "grep: /etc/inittab: No such file or directory" on feisty.
| |
| 20:00 | * moquist doesn't yet grok upstart (or whatever the SysV init replacement is called) | |
| 20:00 | <jammcq> moquist: yeah, ltspcfg doesn't work so well on feisty. So, you just need to manually configure a few things. Changes are most of it is already configured
| |
| 20:00 | <moquist> I think that's it...
| |
| 20:01 | jammcq: Yeah, that's what I figure.
| |
| 20:02 | <comfrey> ok sweet
| |
| 20:02 | just an inet.d thing
| |
| 20:03 | <moquist> jammcq: is ltsp-update-kernels an Ubuntu thing, or is it the right thing for me now?
| |
| 20:03 | <jammcq> it's an ubuntu thing
| |
| 20:03 | <moquist> k
| |
| 20:03 | <jammcq> now, one thing you'll find is ltspadmin installed the kernel in /tftpboot. you'll need to move it to /var/lib/tftpboot
| |
| 20:04 | <moquist> ahh
| |
| 20:04 | so I'll just change the pxe config and not mess with any update-y kernels things
| |
| 20:05 | (I just put lts in /var/lib/tftpboot
| |
| 20:05 | )
| |
| 20:07 | * moquist is booting ltsp 4.2 | |
| 20:09 | <comfrey> anyone know of cursor issues in rdesktop with via chipset?
| |
| 20:09 | <moquist> bah! couldn't mount / via NFS. how could I screw that up?
| |
| 20:10 | <comfrey> moquist: check your exports
| |
| 20:10 | <jammcq> moquist: did you edit your /etc/exports file?
| |
| 20:10 | try: showmount -e
| |
| 20:10 | <moquist> yes
| |
| 20:10 | spdc-ltsp /var/lib/tftpboot # showmount -e
| |
| 20:10 | Export list for spdc-ltsp:
| |
| 20:10 | /opt/ltsp *
| |
| 20:10 | /opt/ltsp-4.2 *
| |
| 20:10 | <jammcq> did you change your root-path in /etc/ltsp/dhcpd.conf ?
| |
| 20:10 | <moquist> I just successfully mounted it from another system.
| |
| 20:10 | jammcq: yeah. :(
| |
| 20:11 | <jammcq> what does it look like?
| |
| 20:11 | <moquist> option root-path "/opt/ltsp-4.2";
| |
| 20:11 | <jammcq> ah, that ain't right
| |
| 20:11 | <moquist> yay!
| |
| 20:11 | <jammcq> should be /opt/ltsp-4.2/i386
| |
| 20:12 | also, for ltsp-4.2, you need the IP address in front of it
| |
| 20:12 | so, it should be like:
| |
| 20:12 | "192.168.0.254:/opt/ltsp-4.2/i386";
| |
| 20:12 | <moquist> 10.10.1.101:/opt/ltsp-4.2/i386
| |
| 20:12 | * moquist nods | |
| 20:12 | <moquist> ok, trying again.
| |
| 20:13 | my boss is going to think I'm obsessing over this when he finds out how late I'm staying to do this.
| |
| 20:14 | "could not find lts.conf file"
| |
| 20:15 | nope. it's definitely not there.
| |
| 20:15 | <comfrey> vnc is supported on a screen, yes?
| |
| 20:15 | <moquist> comfrey: I think so. But I'm hardly an authority on that.
| |
| 20:15 | <comfrey> just SCREEN_XX = vncviewer (options)
| |
| 20:15 | i assume
| |
| 20:15 | ok, will tryu
| |
| 20:20 | kaminski-ltsp-br has quit IRC | |
| 20:20 | gonzaloaf_work has quit IRC | |
| 20:20 | whiprush has quit IRC | |
| 20:20 | vmlintu has quit IRC | |
| 20:20 | pere_away has quit IRC | |
| 20:20 | jbrett has quit IRC | |
| 20:20 | ufo has quit IRC | |
| 20:20 | Legenda has quit IRC | |
| 20:20 | sep has quit IRC | |
| 20:20 | petre has quit IRC | |
| 20:20 | comfrey has quit IRC | |
| 20:20 | ogra has quit IRC | |
| 20:20 | psyco-obiwan has quit IRC | |
| 20:20 | gentgeen__ has quit IRC | |
| 20:20 | mistik1 has quit IRC | |
| 20:20 | kuzew has quit IRC | |
| 20:20 | twinprism has quit IRC | |
| 20:20 | jeremydyoung has quit IRC | |
| 20:20 | stillflame has quit IRC | |
| 20:20 | ccherrett has quit IRC | |
| 20:20 | gu has quit IRC | |
| 20:20 | milosz has quit IRC | |
| 20:20 | Lumiere has quit IRC | |
| 20:21 | alfred_ has quit IRC | |
| 20:22 | <moquist> jammcq: login screen in 38 seconds
| |
| 20:22 | <jammcq> THAT'S what I'm talkin about
| |
| 20:22 | <moquist> now I just need sound and USB
| |
| 20:23 | * moquist high-fives jammcq | |
| 20:30 | <moquist> jammcq: following the breezy instructions worked like a charm
| |
| 20:38 | jammcq: is it even worth thinking about pulseaudio in 4.2?
| |
| 20:38 | <jammcq> hmm, no clue what it would take to make that work
| |
| 20:38 | <moquist> (I mean for me, right now. Not for everyone in general.)
| |
| 20:38 | * moquist nods | |
| 20:39 | <moquist> This is very fast, but very silent.
| |
| 20:41 | Lumiere has joined #ltsp | |
| 20:41 | milosz has joined #ltsp | |
| 20:42 | gentgeen__ has joined #ltsp | |
| 20:42 | mistik1 has joined #ltsp | |
| 20:42 | jeremydyoung has joined #ltsp | |
| 20:42 | stillflame has joined #ltsp | |
| 20:42 | ccherrett has joined #ltsp | |
| 20:42 | twinprism has joined #ltsp | |
| 20:42 | kuzew has joined #ltsp | |
| 20:42 | psyco-obiwan has joined #ltsp | |
| 20:43 | petre has joined #ltsp | |
| 20:43 | ogra has joined #ltsp | |
| 20:43 | kaminski-ltsp-br has joined #ltsp | |
| 20:43 | gonzaloaf_work has joined #ltsp | |
| 20:43 | pere_away has joined #ltsp | |
| 20:43 | vmlintu has joined #ltsp | |
| 20:43 | jbrett has joined #ltsp | |
| 20:43 | whiprush has joined #ltsp | |
| 20:43 | ufo has joined #ltsp | |
| 20:43 | sep has joined #ltsp | |
| 20:43 | Legenda has joined #ltsp | |
| 20:48 | gu has joined #ltsp | |
| 21:05 | J45p3r has quit IRC | |
| 21:06 | bgomes has joined #ltsp | |
| 21:19 | ogra has quit IRC | |
| 21:20 | ogra has joined #ltsp | |
| 21:20 | <petre> ogra: isn't it the middle of the night for you?
| |
| 21:33 | Riddell has quit IRC | |
| 21:34 | Riddell has joined #ltsp | |
| 21:54 | dida has joined #ltsp | |
| 21:57 | petre has quit IRC | |
| 22:08 | dida has quit IRC | |
| 22:27 | freet15 has joined #ltsp | |
| 23:16 | ChinaForge has joined #ltsp | |
| 23:19 | <ChinaForge> Hello,Who have rdesktop screen script for ltsp 5.0?
| |
| 23:21 | <bgomes> ChinaForge: I think dont have... what root-tree u using?
| |
| 23:22 | <ChinaForge> bgomes,What is "root-tree"?
| |
| 23:23 | /opt/ltsp/i386?
| |
| 23:25 | <bgomes> ChinaForge: what dist u using on your ltsp5 root tree ?
| |
| 23:25 | ubuntu/debian ? then made a apt-get install rdesktop with chroot no your root-tree of ltsp5
| |
| 23:26 | <ChinaForge> Ubunt
| |
| 23:26 | Ubunt 7.04
| |
| 23:27 | Ubuntu 7.04 include rdesktop,
| |
| 23:27 | but i hope like ldm/gdm/telnet,
| |
| 23:28 | <bgomes> u can use sdm
| |
| 23:29 | with sdm u can run rdesktop on ltsp5-client
| |
| 23:29 | <ChinaForge> but need login to ubuntu first.
| |
| 23:29 | I hope run rdesktop immediacy.
| |
| 23:30 | <bgomes> why? if u make the key rsa for you ssh-server you dont need login first
| |
| 23:30 | *your
| |
| 23:31 | <ChinaForge> Too difficulty for me.
| |
| 23:32 | ltsp 4.2 have rdesktop screen script,
| |
| 23:32 | <bgomes> ChinaForge: is easy learn about sdm in wki page of ltsp and you will use this easy
| |
| 23:32 | <ChinaForge> I need set lts.conf SCREEN01=rdesktop xxx.xxx.xxx.xx only.
| |
| 23:33 | <bgomes> ChinaForge: it dont working?
| |
| 23:34 | <ChinaForge> ltsp 5.0 no rdesktop screen script.
| |
| 23:34 | <bgomes> I am dont have ltsp now for testing for u, but use this and check if exit rdesktop binary on ltsp-client if not exist use apt-get install rdesktop on root-tree
| |
| 23:35 | or learn abou sdm it's easy
| |
| 23:36 | <ChinaForge> ubuntu 7.04 include rdesktop.
| |
| 23:37 | cyberorg has joined #ltsp | |
| 23:38 | <ChinaForge> I try sdm,Thank you very much.
| |
| 23:39 | <bgomes> ChinaForge: good luck
| |