IRC chat logs for #ltsp on irc.freenode.net (webchat)


Channel log from 22 May 2009   (all times are UTC)

00:29alkisg has joined #ltsp
00:50alkisg has quit IRC
01:03alkisg has joined #ltsp
01:06The_Code has quit IRC
01:36shogunx has quit IRC
01:36shogunx has joined #ltsp
01:42vagrantc has quit IRC
01:49hanthana has quit IRC
01:54alkisg has quit IRC
01:57Egyptian[Home]1 has joined #ltsp
01:57Egyptian[Home] has quit IRC
02:08hanthana has joined #ltsp
02:12alkisg has joined #ltsp
02:16alkisg1 has joined #ltsp
02:21alkisg has quit IRC
02:31syn has quit IRC
02:44intelliant has joined #ltsp
02:54nubae has joined #ltsp
02:58
<alkisg1>
Argh... I just tried spawning some shells from within ltsp-setup to debug https://bugs.launchpad.net/ubuntu/+source/ltsp/+bug/348305 and it works fine... I hate race conditions
03:21alkisg1 has quit IRC
03:49alkisg has joined #ltsp
03:50hanthana has quit IRC
04:00pmatulis has joined #ltsp
05:35intelliant has quit IRC
05:39
<nubae>
alkisg: might be something for you: http://blog.lxde.org/?p=336
05:40
<alkisg>
ty, looking...
05:50DarKnesS_WolF has joined #ltsp
06:03
<stgraber>
alkisg: what would be interesting to know is where that race condition actually is (as I never was able to reproduce the issue ...), then we can just put some code waiting for whatever is causing the issue to finish, then continue the boot process
06:03
<alkisg>
stgraber: I think I'm pinpointing it, I'll post the results in a few minutes
06:03
<stgraber>
alkisg: seeing that kind of error message and the comments, that nbd reconnection code seems to be the issue though it was already fixed to avoid a race condition :)
06:04
(and I'm not going to put a "sleep 2s" after it ;))
06:04
<alkisg>
When I get the error, I also get a SQUASHFS error... ok gimme a few minutes and we'll talk then
06:10
stgraber: could you explain to me the logic in configure_nbd ?
06:13
<dmarkey>
has any changes been done to the ltspfs loopback stuff?
06:14mikkel has joined #ltsp
06:14_Shane has joined #ltsp
06:16
<alkisg>
stgraber: I put a shell right before and after configure_nbd. If I just "exit" from the shell, I get the error. If I do "ls -la /etc/resolv.conf" before exiting the shell, I don't get the error! Argh...
06:17
The SQUASHFS error (SQUASHFS error: sb_bread failed reading block 0x44781) is random, sometimes I get it sometimes I don't, independendly from if the resolv.conf bug appears or not.
06:24
A `touch /etc/resolv.conf` just before configure_nbd seems to fix things...
06:44_Shane has left #ltsp
06:47Shane_ has joined #ltsp
06:48
<alkisg>
stgraber: verified: a `touch /etc/resolv.conf` as the first line of configure_nbd fixes the bug for me. A `sleep 10` there instead of `touch` *doesn't* fix it.
07:00jammcq has quit IRC
07:23pmatulis has quit IRC
07:34
<Shane_>
Can anyone help me work through a problem getting localdev / ltspfs working.On a client shell when I type the following command:ssh -X -S /var/run/ldm_socket_5179_172.16.1.253 172.16.1.253 "/usr/sbin/ltspfsmounter /tmp add"I get the following errors:Couldn't read LTSPFS_TOKEN atom.Error: /tmp/.shane-ltspfs/tmp is not mounted
07:34Ahmuck_ has joined #ltsp
07:35
<Ahmuck_>
hi. i have RPL ROM boot error messages, something to do with novell ... does ltsp boot this type of ROM?
07:35
or only PXE
07:36
ROM Firmware 1994
07:42
ok, it's Novell RPL RPM (boot) for the nic, and here is no way to change it to pxe. does ltsp have support for novell rpl boot rom ?
07:50Faithful has joined #ltsp
08:08
<alkisg>
Ahmuck_: http://www.ltsp.org/twiki/bin/view/Ltsp/RPL
08:11alkisg has quit IRC
08:14alkisg has joined #ltsp
08:19Gadi has joined #ltsp
08:47litlebuda has joined #ltsp
08:55
<stgraber>
alkisg: ok, so let's try something :)
08:56
<alkisg>
stgraber: I'm here
08:57
<stgraber>
alkisg: in ltsp-client-setup at the end of the file, move configure_resolver before configure_nbd
08:57
<alkisg>
Hmmm sounds good
08:57twinprism has quit IRC
08:58
<alkisg>
(updating image.... I wonder why nbd needs restarting, though)
08:59
<stgraber>
ok, so I'll try to explain that quickly :)
08:59
basically nbd is started from the initrd with -persistent
08:59
one would think it'd be persistent :)
08:59
but it's not really as when it dies the userspace process dies too and then the *kernel* tries to respawn the userspace process to initialize the reconnection
09:00
then isssue being that the *kernel* only knows the location of the userspace binary in the initrd
09:00
and the initrd is no longer there because it's wiped as soon as the system boots
09:01
so configure_nbd disconnects and reconnects using the userspace binary from the NBD itself, that way when the kernel respawns the userspace binary it'll work (as it'll be cached and from an existing location)
09:02
<alkisg>
Thanks, I got it.
09:03
TADA: moving configure_resolver before configure_nbd worked fine for me!
09:03
<stgraber>
btw, for these of you using Jaunty and tired of gimp, inkscape and firefox being extremely slow (right click in firefox and startup for the two others), I've a patched libxcb1 in my PPA solving the issue
09:03
<alkisg>
YEAAAH!!!!
09:04
woot woot!
09:04
<stgraber>
the guys at Revolution Linux office say it works just perfectly, so I'm now nagging Bryce to get that in ubuntu
09:04
(and I'll be able to nag him in person next week in Barcelona)
09:05
<alkisg>
Really good!!! Ah, question, will you put udhcpc as the default client in 9.10?
09:05
<stgraber>
that's the plan yes
09:05
we have a session about that
09:05
<alkisg>
Could you support IPAPPEND 3?
09:05
<nubae>
how does it speed up ubuntu?
09:06
<stgraber>
it doesn't speed it up a lot but it gives you lease renewal so you won't get your ip assigned to another thin client after a while
09:06
<alkisg>
I mean, it's supported by the ipconfig command line, but not with udhcpc, so a little scripting will be needed. I can send you the needed lines if you want
09:07
<stgraber>
alkisg: you can have a look at the udhcpc script I'm using in Jaunty and give me a patch based on that ?
09:07
<alkisg>
Sure, will do.
09:07
Thanks
09:07
<stgraber>
that's in /usr/share/initramfs-tools/scripts/init-premount/udhcp in the thin client chroot
09:09Faithful has quit IRC
09:11fgiraldeau_ has joined #ltsp
09:12
<fgiraldeau_>
hello, does someone can remind me why a ltspfsd mount-bind directory?
09:15twinprism has joined #ltsp
09:19twinprism_ has joined #ltsp
09:19twinprism has quit IRC
09:22twinprism has joined #ltsp
09:22twinprism_ has quit IRC
09:32
<stgraber>
Gadi: ^ (we have some issues with ltspfs and rdesktop at Revolution Linux if you have a minute)
09:32Shane_ has left #ltsp
09:36
<Gadi>
stgraber: you have of my undivided attention ;)
09:36
*you have half
09:42pmatulis has joined #ltsp
09:58
<stgraber>
fgiraldeau_: ^
09:58* stgraber tries to do some coordination ... :) (I'm actually on holiday so ...)
09:59
<stgraber>
doh, and now francis isn't answering is phone ...
10:01
<fgiraldeau_>
stgraber, hey! you shouldn't be connected :)
10:01
<Gadi>
well, I am on no such holiday ;)
10:01
<stgraber>
fgiraldeau_: and you should answer your phone :)
10:01* Gadi shakes the chain attached to his leg
10:01
<fgiraldeau_>
Gadi, hello
10:01* Gadi waves to fgiraldeau_ and asks cliebow to hug him for me
10:01
<Gadi>
:)
10:01litlebuda has quit IRC
10:02
<Gadi>
so, anybody bring snacks?
10:04
<fgiraldeau_>
Gadi, why lbmount is doing also a mount bind in /tmp ?
10:05
<stgraber>
fgiraldeau_: maybe start with the general issue no ? (just to put a bit more pressure on Gadi ;))
10:05
<Gadi>
security - afaik
10:06
:)
10:06
<fgiraldeau_>
Gadi, I saw that ltspfsmounter is now also called on the thin-client itself?
10:06* Gadi needs more pressure - he wants to become a panini
10:06
<Gadi>
fgiraldeau_ yeah, I saw that too
10:06
:)
10:06
<fgiraldeau_>
stgraber, you're right... let's step back
10:07
there is a hang when removing a usb key while not using ldm
10:08
seems that two ltspfs_entry process tries to unmount the device
10:08
<stgraber>
(in our case, using rdesktop to export /media to the Windows terminal server)
10:11
<fgiraldeau_>
the script ltspfs_entry launch a local process of ltspfsmounter, but the xauth file is not in /tmp anymore... should be updated. But I don't think this is related to the problem.
10:15hanthana has joined #ltsp
10:15litlebuda has joined #ltsp
10:15
<fgiraldeau_>
in screen.d/ldm : XAUTHORITY=`mktemp -u /var/run/ldm-xauth-XXXXXXXXX`
10:16
<stgraber>
btw: I'm going to tag ltsp-trunk now unless someone is working on it.
10:16
<Gadi>
fgiraldeau_: not sure why there would be 2 ltspfs_entry processes
10:16
<fgiraldeau_>
in xinitrc.d/I00-xauth : XAUTHORITY=`mktemp -u /var/run/xauth-$DISPLAY-XXXXXXXXX`
10:17
<Gadi>
but, with ltspfs_entry, there are two calls - one to ssh server ltspfsmounter and one to local ltspfsmounter
10:17
<fgiraldeau_>
Gadi, that's because a ltspfs triggers a mount
10:17
<Gadi>
so?
10:17
<fgiraldeau_>
strange, but in the mount procedure, if the mount fail, it calls ltspfs_entry to remove the entry
10:18
<Gadi>
only if it is not a fail because of being "already mounted"
10:18
<fgiraldeau_>
but when the usb key is removed, there is also a udev event that triggers ltspfs_entry remove
10:18
<Gadi>
(unless that part is in upstream and not jaunty)
10:19
<fgiraldeau_>
it fail because the usb has been removed ;)
10:19
<Gadi>
ok
10:19twinprism has quit IRC
10:19
<Gadi>
so, then it should call ltspfs_entry remove
10:19
I see
10:19
and thats the second ltspfs_entry process
10:19
which hangs
10:20
what if you change the unmount script to do a lazy unmount?
10:20
does it fix things?
10:20
<fgiraldeau_>
I will try that
10:20staffencasa has joined #ltsp
10:21
<fgiraldeau_>
... and about ltspfsmounter in ltspfs_entry : XAUTHORITY=$(find /tmp -maxdepth 1 -mindepth 1 -type f -name xauth-$DISPLAY-*)
10:22
Should the ltspfsmounter be called locally with LDM, or only with rdesktop?
10:22
<Gadi>
fgiraldeau_: I don't think the XAUTHORITY is even necessary
10:22
because we patched ltspfsd to not require auth when the connection is from localhost
10:23
<fgiraldeau_>
ok
10:23
the user variable is defined only if the user has it's home dir mounted by sshfs
10:23
which is not the case with RDP, isn't?
10:23
<Gadi>
while ur testing, if you test with removing that and it works, lets take it out of upstream code
10:24
<fgiraldeau_>
ok
10:24
<Gadi>
right, some of this was an attempt to satisfy localapps, as well
10:24
so, for localapps, it uses the USER
10:24
if USER is empty, it uses root
10:25
which is why we redirect /media/root in RDP
10:25
because lbmount requires the use of USER hardcoded
10:25
<fgiraldeau_>
good, everything fine then
10:25
<Gadi>
which is annoying
10:25
:)
10:25fgiraldeau_ has left #ltsp
10:25
<Gadi>
but, I didnt want to touch lbmount, because it would mean a security audit
10:26fgiraldeau_ has joined #ltsp
10:26
<fgiraldeau_>
oups
10:26yanqui` has joined #ltsp
10:27yanqui` is now known as syn
10:27twinprism has joined #ltsp
10:29
<fgiraldeau_>
I wonder what happens for a ltspfs client to trigger some activity on the mount point while removing the device
10:30
I don't know how to find it ;)
10:33
thanks for your input, will figure it out and keep you in touch
10:35dmaran has joined #ltsp
10:45CAN-o-SPAM has joined #ltsp
10:46
<stgraber>
alkisg and others: I have all of Karmic ltsp backported in my PPA, haven't tested them yet though
10:47
<alkisg>
stgraber: I'll do some extensive testing in the weekend, along with the backports, thanks
10:47
<stgraber>
np
11:01nubae has quit IRC
11:02hanthana has quit IRC
11:02hanthana has joined #ltsp
11:02
<dmaran>
I am putting together a in house thin client and had some issues with a D945GCLF2. Quick documentation in here so it gets logged. Seems the HT technology throws Ubuntu 8.04's NBD server for a loop so that gets disabled in the bios. Once that is done you can then almost get to the ldm, but you need to specify the intel grfx in the lts.conf "XSERVER = intel" first. Testing it more after lunch.
11:03
I saw some info from Asmo about this, anyone know of a wiki entry for it?
11:13nubae has joined #ltsp
11:15litlebuda has quit IRC
11:18
<Ahmuck_>
stgraber: kermick ltsp ?
11:20
<nubae>
Ahmuck_: your nick always makes me think of the saying "how much wood would a woodchuck chuck
11:20
:-)
11:20
<Ahmuck_>
heh heh
11:20
when 911 happened i used to get harassed over it
11:22
<cyberorg>
Ahmuck_, to me it reminds me of this http://www.youtube.com/watch?v=1uwOL4rB-go
11:22
<Ahmuck_>
see what i mean
11:23* Ahmuck_ grumbles
11:23
<cyberorg>
Ahmuck_, seen that?
11:23
that is probably one of the most memorable youtube video i've seen ;)
11:26litlebuda has joined #ltsp
11:30
<loather>
SILENCE!
11:30
i *KILL* you.
11:30
<nubae>
heh
11:31
jeez I've been reading your nick as ahmchuck till now... I must be visually dislexic or something
11:36litlebuda has quit IRC
11:40alkisg has quit IRC
12:04etyack has joined #ltsp
12:05Lns has joined #ltsp
12:05etyack has quit IRC
12:11dmaran has quit IRC
12:23
<syn>
nubae: what are you using to get the logs http://www.nubae.com/logs/
12:23GodFather has joined #ltsp
12:27
<nubae>
ltsplogbot does that and he's an eggdrop
12:28
I use a script on the backend to take the logs and convert to html
12:28
on my site, the procedure is described and documented
12:28
www.nubae.com
12:28
<syn>
cool thanks
12:54pmatulis has quit IRC
12:55alkisg has joined #ltsp
13:14litlebuda has joined #ltsp
13:18davidj has quit IRC
13:19twinprism_ has joined #ltsp
13:21spectra has joined #ltsp
13:23spectra has quit IRC
13:23hanthana has quit IRC
13:24twinprism_ has quit IRC
13:24spectra has joined #ltsp
13:29din_os has joined #ltsp
13:29vagrantc has joined #ltsp
13:36twinprism has quit IRC
13:55twinprism has joined #ltsp
14:07mikkel has quit IRC
14:20intelliant has joined #ltsp
14:34
<fgiraldeau_>
Gadi, I confirm that the bug can be avoided by making sure that the mount is not triggered while umounting
14:34
patch here : https://bugs.edge.launchpad.net/ubuntu/+source/ltspfsd/+bug/378495
14:37johnny has left #ltsp
14:50
<vagrantc>
fgiraldeau_: seems like a reasonable fix.
14:51
still some potential for a race condition, but less
14:52
<fgiraldeau_>
vagrantc, you're right... BTW, I do have few questions on ltspfsd
14:52
why is there a mount --move on the mount point?
14:52
<vagrantc>
fgiraldeau_: as opposed to --bind?
14:52
fgiraldeau_: or mounting directly?
14:52
<fgiraldeau_>
and in ltspfsmounter, there is a lbmount --umount and fusermount -u, they seems to do the same thing...
14:53
Why not mount directly, yeah?
14:53
<vagrantc>
i brought up that question years ago, too ... if i remember the rationale...
14:54
basically, it allows for /media to only be writeable by root, so you do the mount in /tmp as the user, and then the setuid binary relocates the mount...
14:54
<fgiraldeau_>
ok, fine
14:54
I was wondering why, that make sens
14:54
<vagrantc>
it used to use --bind mounts, but i switched it to --move at some point
14:55
not sure on the fusermount -u vs. lbmount --umount issue...
14:55
<fgiraldeau_>
I found another bug while inspecting the code... let me a minute to document it
14:57
<vagrantc>
ah, in scripts/ltspfsmounter, lbmount --umount only handles mounts in /media, whereas fusermount is used if for some reason there's a leftover mount in /tmp
14:59tbb\fbr_ has joined #ltsp
15:00 i was given this channel by the people at freegeek in portland oregon and am hoping to get some assistance with an ltsp5/ubuntu 8.10 setup i have going which is going me video issues.
15:00
<vagrantc>
fgiraldeau_: maybe not the most clean way to handle it, but there's some difference to how each is used.
15:01
tbb\fbr_: welcome ... i'm not the most savvy with ubuntu, but there's a number of other people here who are
15:03 when the thin client is started, it goes through the boot procedure and then gets to a point where an xauth message is printed in the upper line of the screen. the screen continually blinks, and never starts. if i unplug the video cable from the monitor, and wait a few miniutes, and then plug the video cable back in, i then see the ubuntu login screen... anyone familiar with this? 1 point to note. this is a widescreen display.
15:04
<fgiraldeau_>
vagrantc, https://bugs.edge.launchpad.net/ubuntu/+source/ltspfsd/+bug/379525
15:05
vagrantc, about lbmount : in remove_ltspfsmount
15:05
lbumount_command=['lbmount', '--umount', dev]
15:05
is equivalent to
15:05
ltspfs_umount=['fusermount', '-uzq', root+'/'+dev]
15:07
Hummm... Does a simple user can unmount with fusermount in /media?
15:07
<vagrantc>
fgiraldeau_: it doesn't clean up the directory and such, which the user won't have permission to do
15:07
<fgiraldeau_>
you're right
15:07
<vagrantc>
that's probably why there's the difference
15:08
<fgiraldeau_>
that's fine
15:08
there is a reason to everything ;)
15:09 i just attached a normal size lcd to confirm that this issue is only affecting wide screen at the moment. i think this is only happening when the wide screen is attached, but i will know in a few minutes.
15:12 confirmed. the issue only presents itself when a widescreen is attached. if the thin client is booted with a widescreen attached, it will display an xauth message in the upper line of the screen and *blink* continuously non stop forever until the video cable is pulled for a couple of minutes.
15:13
<alkisg>
tbb\fbr_: try to force a specific X_HORZSYNC and X_VERTREFRESH in lts.conf.
15:14
Look up the specific values for your monitor in the manual.
15:15
<vagrantc>
fgiraldeau_: hmm... the second one is odd. i'm surprised LDM is using a different xauth file.
15:15
fgiraldeau_: we should standardize that.
15:15
unless there's a reason for that, too.
15:15
alkisg: or does hardy support the XRANDR stuff?
15:15nubae1 has joined #ltsp
15:15
<alkisg>
tbb\fbr_: for 8.10, you'll also need to put CONFIGURE_X=true
15:15
vagrantc: I think hardy doesn't
15:16
But I haven't used in for ages :(
15:16nubae1 is now known as Nubae1
15:16nubae has quit IRC
15:16Nubae1 is now known as nubae
15:17intelliant has quit IRC
15:19
<vagrantc>
oh, this isn't hardy, either.
15:19
like i said, ubuntu isn't my strength.
15:21 alkis: okay. i am going to try that. i just hooked up a regular 19" monitor, and everything boots fine, but I am stuck at a 950x600 resolution which it will not let me change. I am going to do those lts.conf changes now. i'm assuming the lts.conf in tftp root is the lts.conf i will be changing and NOT the lts.conf in /opt/ltsp. right?
15:22
<vagrantc>
yeah, in /var/lib/tftpboot
15:22
i know that much :)
15:25
<alkisg>
tbb\fbr_: yeah, create a file named /var/lib/tftpboot/ltsp/i386/lts.conf containing: [Default] CONFIGURE_X=true / X_HORIZSYNC=30.0-90.0 / X_VERTREFRESH=60.0 ==> but put the correct timings for your monitor.
15:26 me too... and the correct timings you mean in placeof the 30-90 and 60 above, right? i can get the timings from the monitor manual or manufacturers web site, that much i can do...
15:27
<alkisg>
Yup
15:35 and to be absolutely sure i'm putting the lts.conf in the right place (it doesn't exist yet) it's going to go in my tftproot (/var/lib/tftpboot/ltsp/i386). doing it now.
15:35warren has joined #ltsp
15:35
<warren>
With Ubuntu's initramfs, if something goes wrong during boot does it drop to a shell or just fail?
15:36CAN-o-SPAM has quit IRC
15:39
<alkisg>
(until someone more knowledgeable speaks) I know that when ipconfig fails, it drops to busybox. I've also seen it dropping to busybox in non-ltsp-pcs with hw problems. I don't know if it does that for all kinds of problems.
15:44 okay, manufacturer lists Max Sync Rate a (VxH)= 76hz x 81.1KHz so X=HORIZSYNC=81.1 and X_VERTREFRESH=76. Look right?
15:45 X_HORIZSYNC=81.1; X_VERTREFRESH=76;CONFIGURE_X=true is what I'm going to stick in lts.conf 1 item per line...
15:47 okay, changes made, rebooting...
15:47
<alkisg>
tbb\fbr_: try with vert refresh = 60
15:47
Ah ok if you already done it.. :)
15:47 i can change it after this time. yeah, it's a 19" lcd so it should be 60
15:48 76 got crt monitors i would guess (right?)
15:48
<alkisg>
Because 76 might be the maximum (for e.g. 800x600), not valid for 1280 or similar
15:48 gotcha
15:48
<alkisg>
TFT can also go up to 75, but not for all resolutions
15:48
But 81 seems too little to me
15:48
(horizontal)
15:50
<fgiraldeau_>
vagrantc, about xauth: you're right, we should use the same file path, it doesn't matter which DM is in use...
15:51
I will do a bundle patch to fix that
15:51
vagrantc, thanks and have a nice day!
15:51fgiraldeau_ has quit IRC
15:52 okay, i put those settings in but nothing has changed
15:53
<alkisg>
Try with vertical=60
15:53 i'm stuck at a really ugly 960x600 and cant even see the ok and apply buttons at the bottom of screen resolution change utility. gonna try 60 now
15:53
<Gadi>
use ranges
15:54Shingoshi has quit IRC
15:54
<alkisg>
tbb\fbr_: wait a minute
15:54
<Gadi>
otherwise you are shooting flies with a shotgun
15:54
<alkisg>
"nothing changed" => the screen still flashes like before?
15:54
Or "still 960x600" ?
15:54 the thin client does boot. it already boots fully and i get to a login screen, i can log in, it's just i can't change the resolution
15:54Lns has quit IRC
15:55
<alkisg>
tbb\fbr_: then it's the 81 horizontal
15:55 the screen flashing thing was with widescreen. i've eliminated that and moved to regular sized 19" monitopr
15:55
<alkisg>
With about 87, you get 1024
15:55 now i'm just trying to get the ability to change resolutions.
15:56
<alkisg>
Try with horizontal=90, you'll get the 1024 resolution
15:56
Your manual doesn't have the correct timings.
15:56
:)
15:56 ok
15:56 will try now
15:56
<nubae>
alkisg: so now that we've got approval to make a livedvd/usb, how do u want to proceed?
15:57
<alkisg>
nubae: I can only start working a month from now, so I'm not sure.
15:57 my lts.conf in /var/lib/tftpboot/ltsp/i386 now has: CONFIGURE_X=true; X_HORIZSYNC=90; X_VERTREFRESH=60
15:57 rebooting
15:57
<nubae>
alkisg: yeah me too
15:58
<alkisg>
I think Svenstaro has free time now, so maybe he could look into the infrastructure?
15:58
tbb\fbr_: X_HORIZSYNC=30.0-90.0
15:59warren has left #ltsp
15:59 oh
16:00 i'll change it on the next go
16:00 right now i set it to 90 i think
16:00 let me see what happens on this boot
16:02 no changhhe
16:03 i'll try the range now
16:03 for some reason, i don't this is it... could have something to do with the thin client hardware i have?
16:04
<alkisg>
boot with X_HORISZYNC=30-90 and if you don't get 1024x768 we'll look at xorg.7.log
16:05 i have a little thin client box (5"X5"X1"). it's some special thin client with 128mb RAM and I don't even know what kind of video it has do i need to load a special driver for this thin client box/.
16:05 ?
16:05
<alkisg>
Do you have [Default] as the first line of lts.conf ?
16:05 no
16:06 '[Default]'
16:06 without quotes?
16:06
<alkisg>
Without quotes
16:06 ok
16:06
<alkisg>
(11:24:33 μμ) alkisg: tbb\fbr_: yeah, create a file named /var/lib/tftpboot/ltsp/i386/lts.conf containing: [Default] CONFIGURE_X=true / X_HORIZSYNC=30.0-90.0 / X_VERTREFRESH=60.0 ==> but put the correct timings for your monitor.
16:06 missed that, making change now
16:06class1 has joined #ltsp
16:08 okay, it's booting now... thin clients sure do take a long time to boot... i might have to get this customer to just go with real computers...
16:08 but i'm going to eat about $350
16:08spectra has quit IRC
16:08
<class1>
in ubuntu 8.10 and 9.04 system messages are being broadcasted to every client. this is an issue, as if teacher a is printing something up on disciplinary measures on studen a and stundent b sees the title, it comes across on every computer
16:09twinprism has quit IRC
16:11 i just saw a Ioctl failure error which i never saw before just flash on the screen.
16:11
<alkisg>
Ignore it :)
16:11 it's on the login screen now though...
16:12 it's logging me in...
16:13 right now it's in 960x540. the highest option is 960x600, but can't choose that because i can't hit the ok or apply button because they are off the screen
16:13 nothing has changed
16:14
<alkisg>
OK, let's try debugging.
16:14
Append the following lines in lts.conf:
16:14
SCREEN_02=shell
16:14
SCREEN_07=ldm
16:14
What do you have now in lts.conf?
16:14
!pastebot
16:14
<ltspbot>
alkisg: "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.
16:16 [Default]
16:16 CONFIGURE_X=true
16:16 X_HORIZSYNC=30-90
16:16 X_VERTREFRESH=60
16:16 SCREEN_02=shell
16:16 SCREEN_07=ldm
16:16GodFather has quit IRC
16:16
<alkisg>
Hmmm maybe replace the SCREEN_0x lines with 'LDM_DEBUG_TERMINAL=yes' (without quotes) it'll make it easier
16:16 and this is in lts.conf on /var/lib/tftpboot/ltsp/i386/lts.conf
16:17 ok
16:17
<alkisg>
Reboot the client and login, a local xterm should pop up
16:17Shingoshi has joined #ltsp
16:17 edits made, rebooting client
16:17Lns has joined #ltsp
16:18
<alkisg>
tbb\fbr_: wait!!!
16:18
Typo!!!
16:18
It's X_HORZSYNC
16:18 !!
16:18
<ltspbot>
tbb\fbr_: Error: "!" is not a valid command.
16:18 okay, fixing now
16:19 [Default]
16:19 CONFIGURE_X=true
16:19 X_HORZSYNC=30-90
16:19 X_VERTREFRESH=60
16:19 LDM_DEBUG_TERMINAL=yes
16:19
<class1>
u know what irc channels need ... bot with bug filing capablities
16:19
<alkisg>
ok
16:19 turning thin client on now...
16:19
<class1>
!bug -- system wide messages on every client
16:19
<ltspbot>
class1: Error: "bug" is not a valid command.
16:19
<class1>
lol
16:20 i really appreciate this...
16:21 doing this in real time... well... it rox...
16:21 i used the 8.10 alternate cd for the ltsp install (hit f4)...
16:21 ubuntu
16:21
<alkisg>
9.04 boots a little faster
16:22
And LDM_DIRECTX=true also helps...
16:22artista_frustrad has quit IRC
16:22 i had 9.04 but could not get ltsp-server-standalone to work at all... but i am new to all of this... i have been doing unix/linux for 20 years so my skills arent lacking there...
16:23 it's sitting on the ubuntu login screen now...
16:24 nothing has changed so far...
16:24 logging in
16:24
<alkisg>
Wow, it takes a lot of time to boot :)
16:24 xterm up
16:27 this environment i am looking at now is the chroot environment under /opt/ltsp/i386 right?
16:28
<alkisg>
No, it's the DE of the server
16:28
The xterm is local
16:29
tbb\fbr_: so, do you have support for 1024x768 now?
16:29 ctrl-alt-+ just made the fonts super duper small... how do i make them bigger (i thought i was making them bigger.
16:29 no
16:29 smae thing
16:30
<alkisg>
OK, in the xterm there's a file /var/run/Xorg.7.log. Upload it to pastebot:
16:30
!paste
16:30
<ltspbot>
alkisg: "paste" is try !pastebot
16:30
<alkisg>
!pastebot
16:30
<ltspbot>
alkisg: "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.
16:30 the xterm is now unreadable as i've made the fonts too damn small... i'm really starting to think it's the hardware on the thin client i have. i have loaded no special drivers for it...
16:31din_os has quit IRC
16:31 i'm going to have to reboot because i apparently cant make the fonts bigger in the xterm with
16:31
<alkisg>
Xorg.log contains the info for your hardware...
16:31 oops
16:31 there we go
16:31 got it
16:32 where is the Xorg.log
16:32
<alkisg>
/var/log/Xorg.7.log
16:33
Don't paste it here
16:33 that's where i was...
16:33 i wont
16:33 no such file or directory
16:34 i don't have an Xorg.log file... running a find now...
16:34 doesn't exist
16:34
<alkisg>
ls -la /var/log/Xorg*
16:35twinprism has joined #ltsp
16:35 nothing
16:35 how about dmesg
16:35
<alkisg>
That's in the local xterm?
16:35 yes, thats in the local xterm
16:36
<alkisg>
OK. Do: lspci|grep VGA
16:36 SiS 550 PCI/AGP VGA Display Adapter
16:37 so it looks like it did find the vga
16:37 am i having driver issue?
16:44
<alkisg>
You can see what driver is loaded with lsmod
16:47Ahmuck_ has quit IRC
16:47class1 has quit IRC
16:50tbb\fbr_ has quit IRC
16:51DarKnesS_WolF has quit IRC
16:54vagrantc has quit IRC
17:07alkisg has quit IRC
17:24Gadi has left #ltsp
17:44Mip5 has quit IRC
18:16dan_young has joined #ltsp
18:24Sarten-X has joined #ltsp
18:29dan_young has quit IRC
18:32twinprism has quit IRC
18:40staffencasa has quit IRC
19:07elisboa has quit IRC
19:08jammcq has joined #ltsp
19:08
<jammcq>
hello friends
19:12nubae has quit IRC
19:13nubae has joined #ltsp
19:14twinprism has joined #ltsp
19:18
<Lns>
hey jammcq !
19:26vagrantc has joined #ltsp
19:36Lns has quit IRC
19:45twinprism has quit IRC
20:02Egyptian[Home]1 has quit IRC
20:02Egyptian[Home] has joined #ltsp
20:24try2free has joined #ltsp
20:24twinprism has joined #ltsp
20:38vagrantc has quit IRC
21:29
<chupacabra>
[C[C[C[C[C[Cjim
21:29
jim
21:32
<epsas>
win 2
22:04try2free has left #ltsp
22:12litlebuda has quit IRC
22:13topslakr_ has joined #ltsp
22:20topslakr has quit IRC
22:22hanthana has joined #ltsp
23:40alkisg has joined #ltsp