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


Channel log from 15 February 2010   (all times are UTC)

00:11jammcq has quit IRC
01:07klausade has joined #ltsp
01:21alkisg has joined #ltsp
01:39shogunx has joined #ltsp
01:51vmlintu has joined #ltsp
01:52HardDisk has quit IRC
01:59frederickjh has joined #ltsp
02:38pts has joined #ltsp
02:49alkisg has quit IRC
02:57mikkel has joined #ltsp
03:03F-GT has quit IRC
03:06F-GT has joined #ltsp
03:11F-GT has quit IRC
03:33ogra has quit IRC
03:33ogra has joined #ltsp
03:46F-GT has joined #ltsp
04:02ogra has quit IRC
04:02ogra has joined #ltsp
04:12leio has joined #ltsp
05:20alkisg has joined #ltsp
05:39artista_frustrad has quit IRC
05:42otavio has joined #ltsp
05:51artista_frustrad has joined #ltsp
06:11lucascoala has joined #ltsp
06:27nobo has joined #ltsp
06:48lucascoala has quit IRC
06:50
<alkisg>
Gadi_eeepc: ping?
06:51HardDisk has joined #ltsp
07:04etyack has joined #ltsp
07:16mikkel has quit IRC
07:27pmatulis has joined #ltsp
07:33pmatulis has quit IRC
07:33pmatulis has joined #ltsp
07:35pmatulis has quit IRC
07:35pmatulis has joined #ltsp
07:38Faithful has joined #ltsp
07:46sene has joined #ltsp
07:54mikkel has joined #ltsp
08:15F-GT has quit IRC
08:16F-GT has joined #ltsp
08:24bobby_C has joined #ltsp
08:43pts has quit IRC
08:52Gadi has joined #ltsp
08:53
<Gadi>
alkisg: pong
08:54
<alkisg>
Gadi, please can we talk later? we're on a debug session right now... :)
08:56
<Gadi>
np
09:01
<alkisg>
Gadi, about $CHROOT/etc/ltsp/update-kernels.conf. I was under the impression that this would be a used-edited file.
09:02
Now, ltsp-update-image doesn't preserve that file's contents...
09:04
<johnny>
s/used/user/
09:04
<alkisg>
right, thanks johnny
09:05
So e.g. if I put BOOTPROMPT_OPTS='nbdroot=10.160.31.10:2000", it would stay there
09:05shawnp0wers has joined #ltsp
09:05
<Gadi>
ur right - now, you would have to call: ltsp-update-image -S 10.160.31.10
09:07
I suppose we could have a flag that preserves the file and does not write to it, but at that point ur saying, "let's have a tool that can figure out the ports and do the right thing AND let people screw it up"
09:07
:)
09:08
<alkisg>
Gadi, why would it rewrite the file instead of just update the bits it needs?
09:08
E.g. NBDPORT=xxx <=update that line,
09:08
<Gadi>
because that's a great way to produce more errors
09:09
if you like, you should edit ltsp=update-image's config file
09:09
so, for your example, simply put:
09:10
NBD_SERVER=10.160.31.10
09:10
in /etc/ltsp/ltsp-update-image.conf
09:10
so you don't have to pass a commandline argument to it next time
09:10
<alkisg>
Got it... thinking...
09:11
<Gadi>
I don't think users should ever need to chroot and run update-kernels by hand
09:11
or chroot and run much of anything
09:11
the server-side tools should do the dirty work
09:11
<alkisg>
Sure, I was just thinking that it should read any info that ltsp-build-client put to update-kernels.conf
09:12
Let me think of any use cases I may have, and how those would be done with the new way... (thanks btw :))
09:12
<Gadi>
thats also why I have the -f flag
09:12
to force regeneration of kernels and configs
09:13F-GT has quit IRC
09:13CAN-o-SPAM has joined #ltsp
09:13F-GT has joined #ltsp
09:13
<alkisg>
Gadi, also, ltsp-update-kernels now overwrites pxelinux.cfg/default with the chroot one
09:14
<CAN-o-SPAM>
morning all
09:14
<alkisg>
Did that change?
09:14
<Gadi>
alkisg: no, that's the same as it was
09:14
alkisg: I just have a failsafe in the l-u-i to catch if none was copied for some reason
09:15
but, it should pull in the one from the chroot
09:15
<alkisg>
So the user will still need to edit $ROOT/etc/ltsp/update-kernels.conf, chroot, create a proper pxelinux.cfg/default, then run ltsp-update-kernels,
09:15
<Gadi>
it should also be smart enough to use the same ports as were used previously (unless there are 2 chroots with same port)
09:15
alkisg: no no
09:15
l-u-i takes care of everything
09:16
l-u-i writes to $ROOT/etc/ltsp/update-kernels.conf, updates the kernels, etc
09:16
<alkisg>
Gadi, and where could I put an extra param? E.g. autoconf=dhcp ?
09:16
<Gadi>
user just calls l-u-i
09:17
l-u-i -o "autoconf=dhcp"
09:17
<alkisg>
Got it. Thank you - I'll look into all that...
09:17
<Gadi>
or l-u-i "quiet splash autoconf=dhcp"
09:17
oops -o
09:17
but, u get the idea
09:18
<alkisg>
Or BOOTPROMPT_OPTIONS in the config file, yup
09:18
<Gadi>
OR, you could set: BOOTPROMPT_OPTIONS="quiet splash autoconf=dhcp" in /etc/ltsp/ltsp-update-image.conf
09:18
right
09:19
I also added a -T for timeout and -I for ipappend values
09:21etyack has quit IRC
09:22etyack has joined #ltsp
09:30
<stgraber>
alkisg, Gadi: Has one of you tested the package currently in my PPA ?
09:30
at home I wasn't able to login with LDM after updating to that
09:31
<alkisg>
stgraber: I'm testing right now, but I'm not even able to boot yet :D
09:31
<stgraber>
haven't tested at the office yet, so it may be that my appserv is broken somehow
09:31
<alkisg>
(some pxelinux.cfg/default changes...)
09:33
stgraber: does tftpd-hpa start on boot for you? It doesn't for me..
09:33
(Lucid)
09:33
<stgraber>
I didn't try to reboot ;)
09:34
it's a VZ container I'm using for it
09:34
<ogra>
lol
09:34
<alkisg>
:)
09:34
<alincoln>
i had to manually add the -s switch to /etc/default/tftpd-hpa this morning after updating my lucid server...
09:34
<ogra>
alkisg, what does your /etc/default/tftpd-hpa say ?
09:35
<stgraber>
it's supposed to start at S20 here but it's just looking at the init scripts
09:35
and I know that /etc/init.d/tftpd-hpa start works
09:35
<alincoln>
stgraber: i just built upstream trunk uing your packaging branch, and i'm not able to log in with ldm.
09:36
so whereas it's not your PPA, it might be close...?
09:37
<alkisg>
http://paste.ubuntu.com/376917/
09:37
ogra ^
09:39
<ogra>
looks different to what i got with todays upgrade
09:39
directory definately points to /srv
09:40
<alkisg>
ogra: it was decided that we should keep using /var/lib/tftpboot - when did you get /srv ?
09:41
Can you purge that file and reinstall to check what you get?
09:42
<ogra>
alkisg, i'm doing a daily update, so i guess it was changed automatically when the debian sync came in... but its definately something to look out for wrt alpha testers
09:43
people that have gotten the plain debian package might have it wrong
09:43
<alkisg>
Yeah, anyone who installed that week would have problems
09:43
<ogra>
right
09:43
<alkisg>
But it gave a visible prompt, so I wouldn't worry too much
09:44
<Gadi>
stgraber: I dont have a lucid build to test with atm, only karmic
09:44
<ogra>
i didnt have a visible prompt
09:44
<alkisg>
(the debconf question visibility was high)
09:44
<ogra>
i know, i still didnt have one
09:46
<alkisg>
daemon.log.1:Feb 15 00:32:00 alkis in.tftpd[1355]: cannot resolve local IPv4 bind address: 0.0.0.0, Name or service not known
09:46
Maybe tftpd starts before networking, and that's why it doesn't work for me?
09:46
(it works fine if I start it later on...)
09:46
<alincoln>
stgraber: well, i can authenticate just fine, but gnome doesn't start and my session ends immediately.
09:46
is this what you see?
09:46
<ogra>
alkisg, someone should turn the initscript into an upstart job so that cant happen ;)
09:47
<stgraber>
alincoln: yeah
09:47
alincoln: and looking at the log it shows SSH worked correctly
09:47
<alkisg>
ogra, heh, what about the server team? :D
09:47* alkisg has no clue about upstart jobs :)
09:47
<ogra>
well :)
09:48
<stgraber>
I don't feel like doing the upstart job, at least not this week ;)
09:48
<ogra>
alkisg, its not that hard, look at /etc/init/ssh.conf
09:50
<alkisg>
Damn I have the old packages :( upgrading...
09:50
<alincoln>
stgraber: i can try to look into what's causing it. open to suggestions if you have any
09:51
<stgraber>
alincoln: I'm not sure, maybe have a look to see if somethin changed in the rc.d scripts and check the environment to see if something is wrong there
09:51
<alkisg>
ogra, that would be a packaging thing, which can be done after FF, right?
09:51
<stgraber>
alkisg: yeah, it can be done after FF
09:51
<alkisg>
OK, no hurry then
09:51
<ogra>
alkisg, i think it requires a bug to be filed though
09:52* alkisg might even ditch tftpd-hpa completely since he's already using dnsmasq for proxydhcp...
09:52
<alkisg>
*and dns...
09:55
Gadi, so if I'm understanding this correctly, if I have "BOOTPROMPT_OPTIONS='autoconf=dhcp'" in /etc/ltsp/ltsp-update-image.conf *before* I call ltsp-build-client, pxelinux.cfg/default will contain it, right?
09:59sbalneav has joined #ltsp
09:59
<Gadi>
it should, tho honestly I tested the command-line args more
10:00
<alkisg>
Thanks Gadi, building...
10:01
<Gadi>
alkisg: btw, I noticed yesterday that, by default, alsa-utils runs from udev on boot to try and set volume levels from /var/lib/alsa/asound.state
10:02
<alkisg>
Gadi yeah I saw it the other day right after you left irc :)
10:02
<Gadi>
got me wondering whether we should simply write that file in an earlier udev rule
10:02
or some such
10:02
(tho, not sure how ubuntu-specific that is)
10:02
<alkisg>
Gadi, the whole thing is to set the volumes specified by lts.conf, right?
10:03
<Gadi>
right
10:03
<alkisg>
What if this was done on the user session ?
10:03
I think there's even a bug filed in ubuntu-ltsp for that...
10:03
<Gadi>
well, you cannot assume that everyone uses ldm
10:03
;)
10:03
<alkisg>
True .... hm...
10:03
<Gadi>
of course, you *should* be able to adjust through ldm in addition
10:04
<ogra>
some people might use the extended version :) (lxdm) :P
10:04
<Gadi>
difficult to restore from ldm, because you may be on a different thin client with different channels
10:06
<alkisg>
Gadi, alsa-utils has hardcoded values for volumes...
10:06
<Gadi>
right
10:06
if none are stored
10:06
<alkisg>
Ah.
10:06
But I wonder what happens if alsa-utils and ltsp tries to set the volumes at the same time
10:07
<Gadi>
yeah - we should prolly add an ubuntu plugin to kill alsa-utils
10:08litlebuda has joined #ltsp
10:13
<_UsUrPeR_>
hey all. I was working on a 9.10 LTSP test env I have, and noticed a lot of slowness on the server's response time from a client. I went to the server to see what was drawing processor power, and noticed that there are logged in sessions from last month still listen in a "ps aux". I assume this is from when a client was not "logged off" and instead the client was shut off. I could have sworn that after no response, the user would be automatically logged of
10:13
f after a period of inactivity. Is that not the case?
10:14Faithful has quit IRC
10:14Faithful has joined #ltsp
10:17jammcq has joined #ltsp
10:17
<jammcq>
bom dia
10:18
<_UsUrPeR_>
hey jamm
10:24otavio has quit IRC
10:24otavio has joined #ltsp
10:27staffencasa has joined #ltsp
10:27
<etyack>
jammcq: mornin
10:37
<jammcq>
etyack: howdie
10:37
_UsUrPeR_: hey
10:40
<_UsUrPeR_>
What function of LTSP in Ubuntu 9.10 logs a user off after inactivity in gnome?\
10:40Faithful has quit IRC
10:40evilx_ has joined #ltsp
10:40evilx has quit IRC
10:47
<alincoln>
stgraber: it's the ldm rev 1242 commit that's broken logging in
10:47
the new way to build the list of rc.d files doesn't work
10:48
<alkisg>
Ugh
10:48garymc has joined #ltsp
10:48
<garymc>
Anyone know how I remove a user?
10:48
Im using ubutu LTSP
10:48
what the sudo command is? in terminal
10:49
<alkisg>
alincoln: I have 2:2.0.55~bzr1242-0ubuntu1~ppa1~lucid1 and it works for me, is that the same that you have?
10:50
<alincoln>
alkisg: well, i packaged an upstream snapshot from this morning using stgraber's ubuntu packaging branch... but i'm at ldm rev 1242, yeah
10:50
alkisg: can you paste ldm-script from your version?
10:51
<alkisg>
Sure
10:52
http://paste.ubuntu.com/376962/
10:52
Ah, that has the old code, I see...
10:53
<alincoln>
alkisg: that's the new stuff, right? the old one used basename.
10:54
<alkisg>
alincoln: I'm not sure, let me check bzr...
10:59
<Gadi>
yup that syntax is flawed
10:59
we need a | tr ' ' '\n'
11:00* Gadi will update
11:01
<alincoln>
Gadi: thanks!
11:02
<Gadi>
try that
11:03
I should actually clean it up a bit more and eliminate one of the loops
11:04bobby_C has quit IRC
11:06
<Gadi>
Here's another that should work, as well
11:07Barbosa_ has quit IRC
11:20
<alincoln>
Gadi: still a no go. $SCRIPTS contains full pathnames, and the egrep line expects just filenames
11:20
so the egrep gets rid of everything
11:21
<Gadi>
ah
11:21
one sec
11:23garymc has quit IRC
11:27
<Gadi>
alincoln: I modified find to return just filenames to begin with
11:27
<alincoln>
Gadi: coolio
11:36
Gadi: yep, works, thanks
11:36
<Gadi>
excellent!
11:36
glad u caught that bug
11:36
<alincoln>
:)
11:44etyack has quit IRC
11:46Lns has joined #ltsp
11:58
<CAN-o-SPAM>
Gadi: socallinuxexpo? or no?
12:01
<Gadi>
California dreaming on such a winter's day
12:03
<CAN-o-SPAM>
heh
12:05
<Gadi>
OT: does anyone know of a zenity-like tool that would allow me to display the contents of a file in a scrollwindow with BOTH OK/Cancel buttons?
12:09
<alkisg>
Gadi, debconf?
12:11
(or a 5 line pygtk program? :D)
12:18
<Gadi>
yeah, yeah, yeah
12:18
:P
12:19
y'know, if everything's just a 5-line pygtk program, why even have zenity? :p
12:28
<Lns>
because not every ltsp dev likes python? ;)
12:39etyack has joined #ltsp
12:47litlebuda has quit IRC
12:55vvinet has joined #ltsp
13:14pmatulis has quit IRC
13:18pmatulis has joined #ltsp
13:22etyack1 has joined #ltsp
13:24etyack has quit IRC
13:33
<vvinet>
!seen sbalneav
13:33
<ltspbot>
vvinet: sbalneav was last seen in #ltsp 3 days, 4 hours, 28 minutes, and 24 seconds ago: <sbalneav> akSeya: https://help.ubuntu.com/community/UbuntuLTSP/ProxyDHCP
13:33
<stgraber>
Gadi, alkisg: Should I make new snapshots ?
13:33
!s
13:33
<ltspbot>
stgraber: "s" :: Scotty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
13:34dro has joined #ltsp
13:34
<vvinet>
thanks graber
13:37
<dro>
does the default pxelinux.cfg/default have 6 lines in it?
13:37Gadi_eeepc has quit IRC
13:37
<dro>
i just did a fresh install of ubuntu 9.10
13:37Kicer86 has joined #ltsp
13:37
<alkisg>
stgraber, I think so, alincoln said he wouldn't get to boot without Gadi's fix for ldm RCs...
13:38* alkisg booted fine, though
13:38
<dro>
alkisg: btw I'm having nbd issues with vbox and 9.10
13:39
<alkisg>
dro, issues like "I get some mountall warnings" or like "it doesn't boot at all"?
13:39
<dro>
alkisg: doesn't boot at all, says Error: failed to connect to NBD server
13:39
alkisg: i'm going to try the ipappend = 3 trick
13:40
<alkisg>
dro, do you have 2 dhcp servers?
13:40
<dro>
alkisg: this is on vbox host only adapter
13:41
<alkisg>
dro, do you see a line "udhcp: configured eth at <ip>:<gw>:<server>" etc line before the error?
13:41
If so, is the server IP there correct?
13:43
<stgraber>
new ltsp uploaded to ppa
13:43
next up is ldm
13:44
<dro>
alkisg; the server ip is wrong
13:44
<alkisg>
That's your problem then
13:44
<dro>
alkisg: but this shows
13:44
IP : IP : subnet
13:44
<alkisg>
Your dhcp server isn't providing the correct server ip
13:45
Yeah, my bad, here's the complete form: http://codtech.com/wiki/index.php/Ipconfig#interface_spec_long_form
13:45
<client-ip>:<server-ip>:<gw-ip>:<netmask>:<hostname>:<device>:<autoconf>
13:46
<stgraber>
ldm uploaded
13:47
<dro>
alkisg: cool modified the #next-server: option in dhcpd.conf and that did it, ty again!
13:55Gadi_eeepc has joined #ltsp
14:34GodFather has joined #ltsp
14:43shawnp0wers has quit IRC
14:48
<Gadi>
stgraber: I compiled your lucid source debs on karmic and installed them - will test this evening if they just work as trivial backport
14:50
<stgraber>
Gadi: ok
14:50pmatulis has quit IRC
14:58eyoop has joined #ltsp
14:59
<eyoop>
hellow all, anyone can help me about an error i get after loading initrd.img and decompress kernel?
15:00
<dro>
eyoop: whats the error
15:01
<eyoop>
init: line 174: cant't open /tmp/net-eth0.conf
15:01
and then kernel panic
15:03
I found some things googling about ... but i cant fix it
15:03
<Gadi>
eyoop: that is usually either because you have more than one network interface on the client or you lack the appropriate driver for the one you have
15:04
<eyoop>
well, i have only one network card
15:04
on the client... and the driver , downloaded from rom-o-matic was 3c90x
15:05
and the client is booted from cd
15:05
<Gadi>
that was for PXE
15:05
there is a problem with the driver in the initramfs
15:05
what distro are you using?
15:05
<eyoop>
debian lenny
15:06
<Gadi>
do you know what driver that card uses?
15:07
<Lns>
Gadi: any words of wisdom for me? rdesktop isn't seeing my win2k3 VM's sound device. I can log in through vmware and use it fine, but an 'rdesktop -r sound:local <ip_of_win_svr> doesn't work :(
15:07
<eyoop>
Gadi: no soryy but i will look for tha information
15:08
<Gadi>
eyoop: seems it might use the 3c59x driver
15:09
try adding the line "3c59x" to /opt/ltsp/i386/etc/modules and /opt/ltsp/i386/etc/initramfs-tools/modules
15:09
then: chroot /opt/ltsp/i386
15:09
update-initramfs -u
15:09
/usr/share/ltsp/update-kernels
15:09
exit
15:09Kicer86 has quit IRC
15:09
<Gadi>
ltsp-update-kernels
15:09
and see if that fixes it
15:10
<eyoop>
Gadi: ive already do this but with 3c90x xD
15:10
ill try with that driver
15:10
<Gadi>
ok
15:10
seems in kernel 2.6 or greater it may share that driver
15:10
and they dropped a separate 3c90x driver
15:10
<eyoop>
thanks you very much
15:10
<Gadi>
Lns: what do you mean by "not seeing"?
15:11
<eyoop>
ill try it and tell you the result ;)
15:12
<Lns>
Gadi: sorry :) Well I use a testsound.wav in wmp and it says "...cannot play the file because there is a problem with your sound device. There might not be a sound device installed on your computer, it might be in use by another program, or it might not be functioning properly."
15:13
that's in rdesktop (tried newest 1.6.0 too). VMWare console session plays it fine through server's sound card
15:13
I also tried -r sound:local:oss but no change
15:14
<Gadi>
Lns: when you go to control panel > Sound what does it have as the audio device?
15:15
<Lns>
Gadi: Microsoft RDP Audio Driver is only choice for default device.. there are no options for Recording Device dropdown, or midi music driver
15:15
<Gadi>
sounds about right
15:15
and what are you running rdesktop on?
15:15
<Lns>
ubu 8.04 *blush*
15:16
<Gadi>
on the server?
15:16
<Lns>
yep
15:16
oh
15:16
hehehe =) Win2k3 Server
15:16
<Gadi>
does: padsp rdesktop -r sound:local:oss .... work?
15:16
<Lns>
nope..sorry that's actually what i have in there
15:17
tried both with sound:local and sound:local:oss (w/padsp)
15:17
<Gadi>
did you uncheck Disable audio mapping in windows?
15:17
<Lns>
nope, where's that?
15:17
<Gadi>
TS Configuration
15:17
<Lns>
ah, lemme try :)
15:17
<Gadi>
right click on rdp, go to client settings and its in one of the tabs
15:17
audio is disabled on WTS by default
15:18
<Lns>
ok
15:18
<dro>
Lns: did you get your audio working with rdesktop?
15:19
<Gadi>
you have to logout/login for it to take effect
15:19
<Lns>
ok..gimme a sec
15:19
dro: that's what i'm working on =p
15:20
<dro>
Lns: what are you using for audio on the rdesktop command in lts.conf
15:20
rdesktop -r sound:local
15:20
found above
15:20
<Lns>
Gadi: don't see that in ts configuration
15:20
<dro>
Lns: did you try: -r sound:local:alsa
15:21
<Lns>
dro: yes
15:21
<dro>
lns: has it ever worked before?
15:21
<Lns>
dro: it seems to work fine at another site w/duplicate setup
15:22
<Gadi>
lns - right click on the RDP-tcp thing
15:22
go to properties
15:22
go to Client Settings tab
15:22
Disable the following:
15:22
make sure Audio is unchecked
15:22
<Lns>
OH gotcha, there it is =)
15:23
<dro>
Lns: :P
15:23
<Gadi>
was it checked?
15:23
<Lns>
Gadi: yep
15:23
unchecked, logged out, going back in
15:23
<dro>
Lns: is that sever part of a domain?
15:24
<Lns>
Gadi: bingo! =)
15:24
tytytytyty!!!
15:24
<dro>
Lns: is that server part of a domain?
15:24
<Gadi>
np - just means Ive been doing this too long
15:24
:P
15:24
<Lns>
Oh I was hoping I wouldn't have to deal with windows all day again today =)
15:25
Gadi: =p well you are a valuable resource for those who cannot wean themselves from windows all the way (yet)!
15:25
Tell your boss you need a raise!
15:25
<dro>
Lns: if that server is not part of a domain and you join it to the domain, there is a good chance it will break again. Even with that setting disabled, have to use the group policy to turn it off again
15:26
http://technet.microsoft.com/en-us/library/cc725887%28WS.10%29.aspx
15:26
<Lns>
dro: wow cool, ty. It is part of a domain actually
15:27
<dro>
Lns: then it will probably be fine, until you upgrade your dc or it to 2008r2
15:27
<Lns>
hopefully that isn't going to happen =p
15:28
<dro>
Lns: just out of curiosity, what all commands did you use on your rdesktop line?
15:28
"-a 16 -r sound:local:alsa -r disk:drives=/media/root -d crutcho -z -x l -P -u ''" is mine
15:29
<Lns>
dro: I just have 'padsp rdesktop -f -u "" -r sound:local <ip>' (heh i remembered that all...)
15:29
<eyoop>
Gadi: hello again but i do what you recomend me with that driver and didnt work
15:29
<dro>
Lns; -z enabled rdp compression, makes a difference
15:29
<eyoop>
when i type update-initramfs -u
15:29
this is the output
15:30
update-initramfs: Generating /boot/initrd.img-2.6.26-2-486
15:30
grep: /proc/modules: No existe el fichero o el directorio
15:30
grep: /proc/modules: No existe el fichero o el directorio
15:30
cryptsetup: WARNING: could not determine root device from /etc/fstab
15:30
<Gadi>
oh,
15:30
you need to mount the /proc directorio
15:30
<eyoop>
and then i lstp-update-kernels
15:30
<Gadi>
in the chroot
15:30
before running update-initramfs -u
15:31
<Lns>
dro: oh...wow that might just help! thx!
15:31
<eyoop>
yes firs i chroot
15:31
<dro>
Lns: no problem, unfortenly if you have windows rdp questions I can help, it's primarily what I have to deal with
15:31
<Lns>
dro: i'm sorry ;) j/k
15:32
<dro>
Lns: i work primarily at k12 schools, I had one district that requested me to be a board meeting
15:32
<Lns>
dro: wow that's great! i do the same actually =)
15:32
<dro>
they advised me that they didn't like linux and it was insecure and to stop using it and then asked what all it ran
15:32
<Lns>
i think we talked about this before ;)
15:32
<dro>
i explained to them everything that it did......then they said n/m
15:32
probably lool
15:33
Lns: you're in the UK right
15:33
<Lns>
heh!! Yeah, schools are a unique type of place to serve with open source
15:33
dro: nope, california, US
15:33
<dro>
thats right cool
15:33
do you work for a contractor or an actual school
15:34
<Lns>
dro: I am the contractor
15:34
<dro>
Lns: cool
15:34
<eyoop>
Gadi:any other suggestion?
15:37
<Gadi>
eyoop: you had an error
15:37dro has quit IRC
15:37
<Gadi>
you need to mount proc
15:38
you never created a new initramfs
15:39
<eyoop>
oooouch, you are right, im not very skilled sorry thanks ill try it :)
15:41vagrantc has joined #ltsp
15:49
<vagrantc>
Gadi: using a single for loop in ldm-script will sort based on the full path rather than matching filenames
15:49
Gadi: so a single for loop is actually broken.
15:50otavio has quit IRC
15:50
<vagrantc>
Gadi: i.e. it will sort all /etc/ldm/rc.d/ before /usr/share/ldm/rc.d, which doesn't allow for /etc/ldm/rc.d/foo to override /usr/share/ldm/rc.d/foo
15:51
<Gadi>
actually, it only sorts filenames currently
15:51
and the last loop will only run the first instance of a filename found in a dir
15:52
<vagrantc>
won't it run multiple times, then?
15:52
and in the wrong order?
15:52
<Gadi>
no
15:52
but, it will run differently than how you say
15:52
so, we should decide on what the desired intent is
15:52
of all these dirs
15:53
right now, it will run like this:
15:53
find all of the filenames in all of the dirs
15:53
sort them
15:53
make them unique
15:53
then source them from the dirs in the order of the loop
15:54
so, if you had, say...
15:54
/etc/ldm/rc.d/foo, /usr/share/ldm/rc.d/foo, /etc/ldm/rc.d/bar, and /usr/share/ldm/rc.d/baz
15:54
you should get:
15:55
/etc/ldm/rc.d/foo, then /etc/ldm/rc.d/bar, then /usr/share/ldm/rc.d/baz
15:55
(which may not be what you have inmind)
15:55
<vagrantc>
that's what i have in mind, but that doesn't look like what the code will do...
15:56vvinet has quit IRC
15:56
<vagrantc>
what does the find -printf do?
15:56
<Gadi>
it outputs just filename
15:56
"basename"
15:56
<vagrantc>
ah.
15:57
ah, well... then that should work fine :)
15:57
<Gadi>
so, it is basically saying: ok, I see foo bar and baz
15:57
now, let's execute them in the dirs
15:57
<vagrantc>
that's where i went wrong
15:57otavio has joined #ltsp
15:57
<Gadi>
oops, and my order was wrong
15:57
'f' comes after 'b'
15:58
:)
15:58
<vagrantc>
ah, sure.
15:58
<Gadi>
so, it'll go bar, baz foo
15:59
<vagrantc>
i didn't catch the -printf '%f\n'
15:59
slick. :)
15:59* Gadi always gets a headache with these "overrides" :)
16:00
<vagrantc>
overrides can be tricky... but they allow for a lot of flexibility.
16:00
<Gadi>
now, I have another question for you
16:00
why do we set: SCRIPTS="$SCRIPTS ...."
16:00
<vagrantc>
not sure.
16:00
<Gadi>
do we expect SCRIPTS to be predefined?
16:00* Gadi thinks that is scary
16:01
<Gadi>
because we never unset SCRIPTS
16:01
<vagrantc>
sounds potentially bad.
16:01
might have been cut-and-paste errors from other code.
16:01
<Gadi>
ah
16:01
I was afraid to change it
16:01
because I thought there was more to it
16:02
I guess we are just lucky that ldm-script is called from different threads in ldm
16:02
atm
16:02
each with its own env
16:02
<vagrantc>
i *think* it's a good idea to remove it...
16:04* Gadi lets vagrantc have the honors
16:04
<Gadi>
;)
16:04
<vagrantc>
heh
16:05
O K
16:06alkisg has quit IRC
16:06
<vagrantc>
we could probably drop @libexecdir@ too
16:06
that was a wild and crazy hack to follow fedora policy or something
16:06
at some point we were going to deprecate it
16:06
<Gadi>
I thought it was to handle backports to /usr/lib/ltsp
16:06
or some such
16:07
<vagrantc>
yeah
16:07
or in fedora's case, /usr/libexec/ltsp
16:07
<Gadi>
well, we have so much other hardcoded /usr/share/ltsp around
16:07
<vagrantc>
we eventually agreed that /usr/share was ok.
16:08
and to deprecate @libexecdir@ at some point
16:08
<Gadi>
Lns: ping
16:09
vagrantc: are you working with real thin clients these days?
16:09
or still VMs for testing?
16:09
<vagrantc>
most of my testing with VMs
16:09
but i occasionally dig out some real hardware
16:10
though i need another LCD monitor ... i've got a few laptops, but one has a broken ethernet port and the other has broken usb.
16:10
it's just enough hassle i don't test with real hardware usually.
16:10* Gadi needs someone to test his "menu" screen script to see whether there is ACTUALLY weird whiptail video ugliness when you log out from LDM and it respawns or if it is just the VM
16:10
<vagrantc>
whiptail is asking for troubles.
16:10
all the dialog implementations have ugly issues.
16:11
Gadi: might want to look at ssft
16:12
it will support whatever dialog frontend is available, falling back to a readline implementation.
16:12
i guess the recommends are a little annoying, though
16:13
<Gadi>
well, whiptail is usually just there
16:13
<vagrantc>
does it handle if it's not?
16:13
<Gadi>
the question is whether whiptail -> X -> whiptail causes badness
16:13
not at all
16:13
:)
16:14
<vagrantc>
in debian, it's listed as "important", which means it will usually be installed, but shouldn't be assumed it's there.
16:14
<Gadi>
nobody's forcing anyone to use the script
16:15
:)
16:15
just available in case
16:15GodFather has quit IRC
16:16
<vagrantc>
sure, but the code needs to handle it not being present unless you add a dependency.
16:16
at least in debian policy.
16:17
handling it could be as simple as saying "whiptail not installed. unable to FOO"
16:18
Gadi: for the record, i see no instances of the variable SCRIPTS in ltsp, ldm or ltspfs other than in ldm-script
16:19
the only use case i could think of would be through an lts.conf variable ... but then you'd want the script to be executed in all phases.
16:19
or, it would be ..
16:20
<Gadi>
ok, I will unable to FOO
16:20
:P
16:20
<Lns>
Gadi: pong
16:21
<Gadi>
Lns: dont know if you'll have a chance to test the "menu" screen script that is upstream
16:21
but if you do, let me know if the menu gets garbled if you log into ldm and logout and come back to the menu
16:21
<Lns>
Gadi: i wanted to say about that, when i exit my gnome session w/menu enabled, it goes back to ldm and not menu (not sure if that was intended)
16:22
<Gadi>
really?
16:22
<Lns>
have you made changes since we last troubleshot?
16:22
<Gadi>
yeah
16:22
I think so
16:22
<Lns>
lemme pull down the new one
16:24etyack1 has left #ltsp
16:31frederickjh has quit IRC
16:34eyoop has left #ltsp
16:43Gadi has left #ltsp
16:49mikkel has quit IRC
16:59
<Lns>
heh, not nice for the one i'm testing for to ditch me =p
17:06
Cheers all!
17:06Lns has quit IRC
17:10klausade has quit IRC
17:14jammcq has quit IRC
17:15vagrantc has quit IRC
17:19CAN-o-SPAM has quit IRC
17:35vvinet has joined #ltsp
17:56GodFather has joined #ltsp
18:22
<stgraber>
and a new ldm snapshot uploaded (rev 1246)
18:33pmatulis has joined #ltsp
18:47staffencasa has quit IRC
19:10slidesinger has joined #ltsp
19:36sene has quit IRC
19:43alexbodn has joined #ltsp
19:47
<alexbodn>
hello friends,
19:47
i have a ts with 2 stations, and it works great. but nfs caused some problems, so i wish to switch to nfs4. i know what to do on the server side, but where is the client defining the mount/boot device location?
19:49
<johnny>
alexbodn, not distro?
19:49
err not ubunu*
19:49
if you used ubuntu it would use nbd
19:49
never heard of anybody using nfs4
19:49
with ltsp
19:50
what problems with nfs3 exactly are you having?
19:50
<alexbodn>
hey johnny
19:51
i have general problems on the server. it serves the /home for other machines too, and file locks etc won't work.
19:52
<johnny>
ltsp doesn't use /home for nfs
19:52
<alexbodn>
nfs4 changed things dramatically, and the server performs much better
19:52
<johnny>
at least not by default
19:52
you would need to edit the initramfs code for nfs4
19:52
if it even includes the nfs4 module..
19:52
try asking on ltsp-discuss
19:52
the lists on sourceforge that is
19:52
!ltsp-discuss
19:52
<ltspbot>
johnny: "ltsp-discuss" :: at https://lists.sourceforge.net/lists/listinfo/ltsp-discuss
19:52
<johnny>
there ya go
19:53
<alexbodn>
thanks a lot, going there
19:56Damianos has joined #ltsp
19:57
<Damianos>
I think I hosed my server
19:57
Can somebody help me?
19:59
<johnny>
!ask
19:59
<ltspbot>
johnny: "ask" :: Don't ask to ask a question, simply ask it, and if someone knows the answer, they'll respond. Please hang around for at least 15 minutes after asking a question, as not everybody constantly monitors the channel.
20:00
<johnny>
i hope you learn a lesson.. never start a support request with that
20:00
just state your case
20:02
<Damianos>
My ubuntu won't boot...somebody mentioned booting from the alt CD and choosing rescue a broken system...won't that completely screw up my network settings and everything?
20:03
<johnny>
no.. it doesn't mean reformat
20:03
go to #ubuntu
20:04
if you need help with your ubuntu system
20:04
this is for supportint ltsp.. or thin client boot problems
20:04
not ubuntu
20:04
you'll find better help there anyways..
20:04
as this channel is dead atm.. i'm the only one here
20:04
and i'm about to go get some food
20:06
<alexbodn>
enjoy your meal :)
20:06
<Damianos>
kali orexi
20:07
<alexbodn>
;)
20:07
<Damianos>
I just came from the ubuntu channel and I started to follow the advice and I saw the installer reconfiguring network settings and all kinds of stuff. I just wanted to make sure it was kosher
20:08
I guess I'm on my own
20:08
...again
20:09Damianos has left #ltsp
20:09
<alexbodn>
always backup/write down your special settings :)
20:11
here, to know if something is kosher, you need a real rabbi ;)
20:45Gama1 has joined #ltsp
20:47Gama1 has quit IRC
21:19dro has joined #ltsp
21:48Gadi_eeepc has quit IRC
21:48Gadi_eeepc has joined #ltsp
21:53Gadi_eeepc has quit IRC
21:53Gadi_eeepc has joined #ltsp
21:59dro has quit IRC
22:02try2free has joined #ltsp
22:07try2free has left #ltsp
22:18GodFather has quit IRC
22:21pmatulis has quit IRC
22:25dro has joined #ltsp
22:27nobo has left #ltsp
22:45dro has quit IRC
23:43Damianos_ has joined #ltsp
23:46johnny has left #ltsp
23:47johnny has joined #ltsp