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


Channel log from 11 October 2008   (all times are UTC)

00:00alkisg has joined #ltsp
00:05
<johnny>
warren, .. so what you're saying.. is that we should have thin clients with lil flash disks inside?
00:05
<warren>
not really
00:05
<johnny>
for local firefox?
00:05
<warren>
the lil flash disks don't have enough space to allow write leveling to make it safe
00:06
larger flash disks is where write leveling makes it safe
00:12
<johnny>
hmm.. i was really hoping to run firefox and flash locally, how much ram is it supposed to take?
00:19* johnny loves finding music he hasn't listened to in a long time
00:19
<johnny>
/listens to Dance Hall Crashers - Enough
00:19
oops
00:19
not many ska bands with two female singers
00:22
1995.. that sure seems like a long time ago now .. uggh.. getting old..
00:23alkisg has left #ltsp
00:25alkisg has joined #ltsp
01:38nubae has joined #ltsp
01:39nubae1 has quit IRC
01:59alkisg has left #ltsp
02:26alkisg has joined #ltsp
03:15alkisg has quit IRC
03:28feinbein has joined #ltsp
03:29
<feinbein>
test
03:33
<laga_>
passed
03:34
<feinbein>
thanks
03:37feinbein has quit IRC
03:43alkisg has joined #ltsp
03:45Q-FUNK has joined #ltsp
03:49gate_keeper_ has joined #ltsp
04:13Pascal_1 has joined #ltsp
04:14K_O-Gnom has joined #ltsp
04:22Pascal_1 has left #ltsp
04:50nubae1 has joined #ltsp
04:51nubae has quit IRC
04:52darrylb_ has joined #ltsp
05:01
<darrylb_>
how can I enable local device support on the client for local apps?
05:30darrylb_ has quit IRC
05:33ogra has quit IRC
05:36ogra has joined #ltsp
06:00toscalix has joined #ltsp
06:19otavio has quit IRC
06:29otavio has joined #ltsp
06:39nubae1 is now known as nuabe
06:40nuabe is now known as nubae
06:46alkisg has quit IRC
07:03nubae has left #ltsp
07:10mikkel has joined #ltsp
07:11hersonls has joined #ltsp
07:22alkisg has joined #ltsp
07:23nubae has joined #ltsp
07:28alkisg has left #ltsp
07:34gate_keeper_ has quit IRC
08:03alkisg has joined #ltsp
08:06ltsplogbot has left #ltsp
08:06ltsplogbot has joined #ltsp
08:19feinbein has joined #ltsp
08:21gate_keeper_ has joined #ltsp
08:24Q-FUNK has quit IRC
08:25
<feinbein>
Hi there! Has anyone managed to get italc working on Ltsp5 ubuntu 8.04.1?
08:26
where do i have to put the line '/usr/local/bin/start_ica &' ?
08:26
I tried with session-manager but to no effect.
08:27gate_keeper__ has joined #ltsp
08:28
<nubae>
feinbein: which version of italc?
08:28gate_keeper_ has quit IRC
08:29
<feinbein>
1:1.07ubuntu2
08:30
<nubae>
there is a newer version, but I believe it is only in a PPA, hang on let me get the link
08:31
https://launchpad.net/~edubuntu-italc-devel/+archive
08:32
that version should work better
08:33
<feinbein>
i found a script called /usr/bin/ica-laucher'. Is that what became of start_ica?
08:33
<nubae>
I believe so, haven't used it in a while
08:34
<feinbein>
i will check that new version, thanks for helping!
08:35
have to log off a moment
08:35feinbein has quit IRC
08:37feinbein has joined #ltsp
08:46
<feinbein>
are you still there nubae?
08:46toscalix has quit IRC
08:47gate_keeper__ has quit IRC
08:48
<feinbein>
I haven't installed a new version yet. I would like to try and get it to work with the distribution version.
08:49
problem is: if i set up a new host in a new classroom as a teacher it shows as 'No user locked in'
08:49
<nubae>
how are u adding the hosts?
08:50
<feinbein>
as soon as a user logs in at the thinclient i get 'host unreachable'
08:50
<nubae>
it should be done like so localhost:10001 - where 1 is the ip of the terminal
08:50
<feinbein>
i try that
08:52
right, i try with localhost:11001, Has that changed from previous versions?
08:52
<nubae>
no, on thin clients its always been 10000
08:52
<cyberorg>
new version of italc from stgraber should automatically add all clients, use italc-launcher script start italc
08:52
<nubae>
I know the italc docs say differently, but hey
08:52
right, but hes using the old one
08:53
hardy
08:53
<warren>
hmm, italc cannot "see" local apps processes
08:53
I'm guessing
08:53
<nubae>
hmm, someone should test that
08:54
<feinbein>
thanks guys its working now
08:54
next problem: I cant use remote control but just see whats happening
08:54
<nubae>
feinbein: for that upgrade to the new one
08:54
<feinbein>
guess this is because my client runs in a virtual box
08:55
i will test with a real client tomorrow
08:59K_O-Gnom has quit IRC
09:00
<feinbein>
if that doesnt work I'll trty the newer version. Thanks for helping!
09:00
bye for now
09:00feinbein has quit IRC
09:06Q-FUNK has joined #ltsp
09:14feinbein has joined #ltsp
09:16feinbein has quit IRC
09:27Egyptian[Home] has quit IRC
09:27Egyptian[Home] has joined #ltsp
09:28hersonls1 has joined #ltsp
09:29hersonls has quit IRC
09:36Q-FUNK has quit IRC
09:39Loto has quit IRC
09:42|Paradox| has joined #ltsp
09:48
<|Paradox|>
so, ltsp.org is no longer updated, and the sourceforge project is stuck on 4.x, where is the repository and all for LTSp 5?
09:50
oops
09:50
I went to the wrong page
09:50
::blush::
10:07jammcq has joined #ltsp
10:07
<jammcq>
hello all
10:11Ozzzkar has joined #LTSP
10:11
<Ozzzkar>
anyone knows how to get sound on my rdesktop session to an xp pro machine?
10:22
<alkisg>
Ozzzkar, you get no sound at all or the sound is played on the ltsp server?
10:22
<Ozzzkar>
alkisg: never mind, think i found a solution
10:22Ozzzkar has quit IRC
10:22
<alkisg>
ok
10:51DonSilver has joined #ltsp
11:11captain_magnus has quit IRC
11:11nubae has quit IRC
11:11captain_magnus has joined #ltsp
11:22MacIver has quit IRC
11:22dcavall has joined #ltsp
11:23MacIver has joined #ltsp
11:26davidj has joined #ltsp
11:26
<davidj>
jammcq: Hello!
11:26
<jammcq>
hey davidj
11:26
<davidj>
How are you? It's been a long time.
11:26
I got a call from Alex Col. the other day, which reminded me that I needed to come say hello.
11:27
<jammcq>
doing great. how's things down there in N.Carolina?
11:27
what'd Alex want?
11:27
<davidj>
Gas is still hard to find.
11:27alkisg has quit IRC
11:27
<jammcq>
wow, hard there too?
11:27
<davidj>
Alex wanted to know if I knew about Diskless Workstations, and about LTSP.
11:27
<jammcq>
it's easy here, and getting cheap. I saw $2.93 this morning
11:27
<davidj>
It's getting better, but they say it will be another couple of weeks before all of the stations have gas.
11:27
<jammcq>
wow, they're out cold calling now, eh?
11:28
<davidj>
$2.93???
11:28
We're paying $4 and change here for low-test.
11:28
<jammcq>
yeah, I'm guessing we'll be seeing 2.79 by monday or tuesday
11:28alkisg has joined #ltsp
11:28
<davidj>
No, Alex said he called because I'd bought from them before.
11:28
more like a warm call.
11:29
When I asked him how his dad was, he seemed to wake up ;-)
11:29
<jammcq>
heh
11:29
<davidj>
How is Maine this year?
11:30
<jammcq>
november 6th, you coming?
11:31
<davidj>
I don't know. There's a lot going on right now.
11:31
<jammcq>
yeah, but this is Maine' we're talking about
11:32
<davidj>
;-)
11:32
That's true.
11:32
Is Ragnar coming?
11:32
mistik1: Are you coming?
11:33
<jammcq>
he said he'd try, but he hasn't confirmed yet
11:34
<davidj>
I've got two interesting projects going at once, and a third is heating up. If things go as planned, I can make it, but if either hits a bump it might be difficult.
11:40
<jammcq>
http://wiki.ltsp.org/twiki/bin/view/Ltsp/ByTheSea2008
11:40
ther'es the details
11:40
if you can make it, that'd be great
11:46tarzeau has quit IRC
11:48siki has joined #ltsp
12:08alkisg has quit IRC
12:09alkisg has joined #ltsp
12:13
<davidj>
jammcq: I'll try ;-)
12:14dcavall has left #ltsp
12:15gate_keeper_ has joined #ltsp
12:35
<stgraber>
ogra: ping
12:36
ogra: can you upload: https://bugs.launchpad.net/bugs/281415 ?
12:37
ogra: I'll also have a new ltspfs soon including sbalneav's fix for a segfault
12:37alkisg1 has joined #ltsp
12:37alkisg has quit IRC
12:40
<stgraber>
Anyone's has a problem with me tagging a new release of ltspfs ?
12:42davidj has left #ltsp
12:45vagrantc has joined #ltsp
12:57
<stgraber>
ogra: and new ltspfs: http://www.stgraber.org/download/ubuntu/ltsp/
12:59alkisg has joined #ltsp
13:00
<warren>
go ahead, i'll probably use your tagged version
13:01
is this our first tag since we began tagging? =)
13:02
<vagrantc>
??
13:02* vagrantc wonders what folks are talking about tagging
13:05alkisg2 has joined #ltsp
13:05alkisg has quit IRC
13:08* vagrantc got confused by stgraber's commit message for tagging 5.1.28
13:17alkisg1 has quit IRC
13:31
<ogra>
stgraber, ltsp uploaded
13:31
stgraber, but there is no ltspfs
13:31
<stgraber>
ogra: ?
13:32
<ogra>
<stgraber> ogra: and new ltspfs: http://www.stgraber.org/download/ubuntu/ltsp/
13:32
<stgraber>
oh, right, 2s
13:32
<ogra>
i only found ltsp there :)
13:32
which i uploaded
13:34* vagrantc is testing ltsp 5.1.28-1 for debian
13:34
<vagrantc>
keeping up with the joneses
13:34
<stgraber>
ogra: uploaded
13:34
<ogra>
vagrantc, btw, we had to change the 50 to 80 again in ltspfs, udev doesnt really like it being run that early
13:35
<vagrantc>
the cdpinger segfault thing ... is it actually causing failures, or just an ugly cosmetic issue?
13:35
ogra: i just do whatever dh_installudev does.
13:35
ogra: so maybe you really need to patch ubuntu's dh_installudev
13:36
ogra: which typically comes out at 60 on debian
13:36
<ogra>
no, its suposed to use the right thing, but uses the defaults by default :)
13:36
<vagrantc>
?
13:37
<ogra>
stgraber added --priority 80, else it will use the default value for a package with module/HW support
13:37
default meaning 50 :)
13:37
<vagrantc>
for debian, it's actually z60
13:37
default
13:38
i was so hoping we'd have at least one release with one of our packages in sync. *sigh*
13:38
<ogra>
http://paste.ubuntu.com/56365/
13:38
thast the upstream README
13:39
80 rules that run programs (but do not load modules)
13:39
by that definition it has to be 80
13:39
but i know debian hacked up udev anyway
13:40* vagrantc is largely udev clueless
13:40
<ogra>
what does your readme say ?
13:40
(/etc/udev/rules.d/README)
13:41
since we run scripts it shoudl eb 80
13:41
*be
13:41
<vagrantc>
no such file
13:41
<ogra>
oh
13:41
intresting
13:41
stgraber, uploaded as well
13:41* vagrantc looks at /usr/share/doc/udev/README.Debian.gz
13:41
<stgraber>
ogra: thanks
13:42
<ogra>
No such file or directory
13:42
heh
13:42
we only have README
13:42
but then we use upstream unmodified
13:43alkisg2 has quit IRC
13:43
<vagrantc>
ogra: default is z60, which i think runs later than everything else.
13:43
<ogra>
ah
13:43
<vagrantc>
note the "z"
13:43
<ogra>
well, it defaulted to 50 for us which resulted in a bunch of variables not being available
13:44
<vagrantc>
yeah. oh well.
13:44
<ogra>
i guess i have to SRU that in hardy as well
13:44
<vagrantc>
that the only diff?
13:44
if so, maybe we could do a build-time conditional
13:46
<ogra>
i think so, yes
13:47
stgraber, ?? ^^^
13:47highvoltage has quit IRC
13:47highvoltage has joined #ltsp
13:48
<stgraber>
we probably can, I don't think we need a SRU for Hardy as we didn't use the ID_TYPE thing back then
13:49
<ogra>
yeah, i will take care for that in hardy, dont worry about that one
13:54alkisg has joined #ltsp
13:55
<vagrantc>
well, uploaded ltsp 5.1.28-1 to debian experimental...
13:59
ogra, stgraber: would a build-dep on lsb-release for ltspfs be ok?
13:59
<stgraber>
yeah, I guess we have it by default anyway
13:59
<ogra>
sure
14:01
<vagrantc>
of course, need to implement this in debian/rules ...
14:02
the fun world of makefiles
14:20vagrantc has quit IRC
14:23K_O-Gnom has joined #ltsp
14:23DonSilver has quit IRC
14:46Egyptian[Home] has quit IRC
14:47alkisg has left #ltsp
14:52Egyptian[Home] has joined #ltsp
14:54otavio has quit IRC
14:59alkisg has joined #ltsp
15:12otavio has joined #ltsp
15:20nubae1 has joined #ltsp
15:21nubae1 is now known as nubae
15:26
<nubae>
hey, stgraber, are u an italc developer?
15:29
<stgraber>
nubae: no but I do the packaging in Ubuntu and contributed some patches upstream, why ?
15:35
<nubae>
because I've been asked to develop a school server that includes it
15:36
should be a XS - LTSP hybrid with schooltool, italc and some other goodies... it will be the prototype to be used for the Austrian school system, backed by ministry of education
15:36
unfortunately the XS is Fedora based
15:37
so not sure if there are latest packages for Fedora of italc
15:38
<stgraber>
well, you'll likely need italc 1.0.9 from upstream and probably some of the patches I applied on top of it
15:39
IIRC opensuse is using some of them for their own package
15:39
I don't know who's packaging it in fedora though
15:40
<nubae>
ah just checked, pbone seems to have 1.09-5 rpm packages
16:12vagrantc has joined #ltsp
16:18alkisg has left #ltsp
16:26toscalix has joined #ltsp
16:29toscalix has quit IRC
16:35K_O-Gnom has quit IRC
16:58horochovec has joined #ltsp
16:58
<horochovec>
hello, my pen-drive dont work in debian lenny, ltsp 5
17:00
the 'dmesg' log show me the information of pendrive, but, dont create a icon on desktop to access
17:02
<johnny>
and the desktop is?
17:02
and is the pen drive partitioned
17:02
if the pen drive isn't partitioned, it can't work
17:03
<horochovec>
my dekstop is kde
17:03
<johnny>
that is probably why
17:04
kde doesn't listen in /media
17:04
there's a hack to make that work.. but i don't know anything about it
17:04
do you see your pen drive in /media on the server ?
17:04
<horochovec>
johnny on server yes..
17:04
my problem is on thinclient
17:05
<johnny>
sure.. but it gets mounted on the server
17:05
sorta..
17:05
the problem is kde
17:05
there's some sort of hack
17:05
to make it work
17:05
i don't know anything about it tho
17:05
somebody else here will tho
17:06
<horochovec>
well
17:07
but in the server, the pendrive dialog show
17:07
in the thinclient dont
17:07
<johnny>
i'm telling you the answer i know
17:07
the problem is kde
17:08
there is a hack to fix it, i just don't know where it is, or how it works
17:08
<horochovec>
well, any idea to resolv this on kde?
17:08
<johnny>
no.. i just said i don't
17:08
other people here do
17:08
this channel is not very active on the weekends tho
17:08
it's most active during typical work hours in europe and usa
17:08
<horochovec>
strange, the kde works on server and dont work on thinclient?
17:08
<johnny>
yep
17:09
<horochovec>
johnny where are you from?
17:09
<johnny>
usa
17:09
<horochovec>
cool
17:09
i'm brazilian
17:09
<johnny>
if you do a web search maybe you will find the answer
17:10
search for /media kde local devices
17:34gonzaloaf_laptop has joined #ltsp
17:39rcy has quit IRC
17:40rcy has joined #ltsp
17:51rcy has quit IRC
17:52rcy has joined #ltsp
17:58horochovec has quit IRC
18:08
<stgraber>
ogra: hey, you still around ? I found a very weird bug in nbd-client ...
18:10
ogra: is there any difference between the nbd-client that we have in the initrd and the one in the chroot ? I'm pretty sure it's not possible but one respects the -persist option and the other doesn't
18:11
<vagrantc>
ouch.
18:12
<stgraber>
basically, I boot my thin client, then kill nbd-server on the server side => nbd-client stopped and no disk access
18:12
then I respawn nbd-client manually on the client and do the test again, this time it reconnects to the server
18:12
checking the command line, both are started with the exact same parameters
18:13
and I don't see any reason for that behavior in nbd's code (I'll probably need more time to fully understand it though)
18:14
of course, that bug makes the -persist option a bit useless :) my current workaround will likely be to disconnect the nbd from ltsp-cluster's init script and start nbd-client again but that's a hack and will only be done so our customers won't complain :)
18:19
oh, I guess I found the problem. When nbd is killed server side, nbd-client client side exits and is then respawned. When it dies the first time (the one started from the initrd) the kernel (or whatever else is respawning it) probably tries to restart the one from the initrd but can't as the root changed.
18:26gate_keeper_ has quit IRC
18:28
<vagrantc>
aha.
18:29
<stgraber>
would that make sense ?
18:30
<vagrantc>
seems feasible.
18:30
<stgraber>
if yes, I'll soon have a fix that can be included in our init script so it'll properly respawn nbd-client (unless you have a better idea how to fix that)
18:30
<vagrantc>
if you're using aufs, maybe you could start a new nbd and insert it into the aufs layer
18:31
and remove the old one
18:32
<stgraber>
http://paste.ubuntu.com/56471/
18:33
that's my current fix
18:33
<vagrantc>
stgraber: why ps aux instead of pgrep ?
18:34* stgraber checks pgrep manpage
18:34
<vagrantc>
i think most of the other nbd detection is using pgrep
18:34
<stgraber>
I only used pgrep to get the PID never to get the command line
18:35
<vagrantc>
pgrep -l -f FOO
18:35
maybe i'm mistaken ...
18:35
i did a lot with pgrep for ltspfs related stuff
18:35
<stgraber>
works, I just need to drop the first column so I don't get the pid, other than that it'll work
18:45
vagrantc: http://paste.ubuntu.com/56477/
18:45
I hate that sleep 0.5s but if I do the check right after disconnecting it, nbd-client still returns the "dead" process PID
18:46
<vagrantc>
hope that sleep is long enough...
18:46
might vary from client to client
18:46
<dberkholz>
we're having a bit of an X hack session
18:47
trying to fix xinit across distros so there's less distro-specific weirdness
18:47
<vagrantc>
cool
18:47
<stgraber>
yeah, I tried on a quite powerful lappy, I'll now test on the most crappy client I have just to make sure we don't have a race condition there
18:47
<dberkholz>
we == pdx X people
18:50mistik1 has quit IRC
18:55mistik1 has joined #ltsp
19:01
<johnny>
dberkholz, any chance we'll use Xsession.d or similiar ?
19:01
and drop the need for xdm to be installed for ltsp?
19:02
<dberkholz>
johnny: trying to figure that out. might break that stuff into a separate app
19:02
<johnny>
what do the other distros do?
19:02
where do they package it?
19:02
<dberkholz>
weird crap
19:02
same as gentoo
19:02
but different types of weird
19:02
johnny: xdm is only for the Xsession script right?
19:02
<johnny>
yes
19:03
<stgraber>
vagrantc: I see that ltsp-client-setup isn't taken from the bzr branch (at least in Ubuntu), how are changes done to the init scripts shared between Ubuntu and Debian ?
19:03
<dberkholz>
debian/ubuntu are weird, fedora is weird
19:03
haven't managed to find suse's stuff yet
19:09
<vagrantc>
stgraber: debian uses the ones shipped upstream.
19:09
<johnny>
stgraber, i use the one in initscripts/Gentoo.. which mirrors the one in initscripts/
19:09
<vagrantc>
stgraber: all of my packaging branches are available: https://code.launchpad.net/~vagrantc
19:10
<johnny>
it's just gentooified..
19:10
that reminds me.. i can prolly pull some stuff out of the mask now..
19:11
<stgraber>
ok, I guess I'll need to ask ogra what forced us to have different initscripts in debian/ ... for now I just patched our existing script
19:12
<vagrantc>
stgraber: it's a little weird how i do it ... i have a script debian/initscripts-tweaks with patches and copies the init scripts into place
19:12
<johnny>
dberkholz, i spoke to ulm or opher .. i forget which.. about bringing bzr.eclass into the eclasses/ .. he said he would post ti gentoo-dev, but i didn't see it last time i looked.. think you can put some pressure ?
19:12
<vagrantc>
stgraber: and it's called from debian/rules in the clean phase.
19:14
<stgraber>
right, if we want to get rid of having our own copy of the initscript we'll also have to do something similar for Ubuntu
19:14* vagrantc would be very happy to work towards a smaller diff
19:14
<stgraber>
for example I'm not sure I want to send that nbd-client hack to ltsp's bzr branch as only Debian and Ubuntu have -persist (it's in the initramfs) and I don't think other distros want their nbd to be killed and respawned at boot time
19:15
<vagrantc>
stgraber: well, every distro has different init scripts ... it's mostly just a set of some shared functions
19:16
stgraber: but how each distro calls those functions varies
19:20
<stgraber>
vagrantc: ok, so would adding the nbd -persist workaround to client/initscripts/ltsp-setup be ok ?
19:20
the way I did it for Ubuntu is adding a configure_nbd() function that's called before all the other configure_ in the script
19:20
<vagrantc>
stgraber: i'm not sure ... i don't really use the nbd stuff much yet.
19:21
stgraber: so it would have to be triggered only when NBD root is kicking in
19:22
stgraber: and one of the nbdroot methods i use actually uses nbd-client's native nbdroot support
19:24
<stgraber>
I guess I'll just keep it as an Ubuntu-specifc hack for now, ideally that should be filed as an nbd-client bug, -persist should respawn nbd-client when the connection is dropped, that should be the same nbd-client
19:24
and if another distro has a bug about that -persist thing, we can then see how to best move the hack to ltsp-trunk without causing troubles for distros with different nbd implementation (or no nbd)
19:30
<johnny>
only two distros use initscripts/ltsp-client*
19:30
it sounds like
19:30
err 3 sorry
19:30
debian,ubuntu,opensuse
19:30
warren uses his own stuff
19:31
it doesn't look much like ltsp-client* stuff at all
19:31
mine is almost exactly like it, without the usplash.. and some gentoo specifics with the way our init scripts are written.. and we don't have console-setup either..
19:54shogunx has quit IRC
20:06shogunx has joined #ltsp
20:17
<warren>
I don't have usplash
20:20siki has quit IRC
20:21
<warren>
vagrantc: you forgot distro prefix
20:21
<vagrantc>
warren: apologies.
20:21
<warren>
np
20:24
<johnny>
warren, you don't use ltsp-client-setup
20:24
so that is nill
20:27
<warren>
ok
20:37mikkel has quit IRC
20:48nubae1 has joined #ltsp
20:49nubae has quit IRC
21:03japerry_cat has joined #ltsp
21:08vagrantc_ has joined #ltsp
21:15japerry has quit IRC
21:19vagrantc has quit IRC
21:21joebaker has joined #ltsp
21:35vagrantc_ has quit IRC
21:53gonzaloaf_laptop has quit IRC
22:05ogra has quit IRC
22:20ogra has joined #ltsp
22:28japerry_cat has quit IRC
22:50joebaker has left #ltsp
23:28gentgeen__ has quit IRC
23:28japerry has joined #ltsp
23:38poklib has joined #ltsp
23:38
<poklib>
hello I'm new here
23:39
anybody know how to solve dhcpcd failed
23:39vmlintu has quit IRC
23:43gonzaloaf_laptop has joined #ltsp
23:55gentgeen__ has joined #ltsp
23:55nubae1 has quit IRC