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


Channel log from 31 January 2009   (all times are UTC)

00:21japerry has quit IRC
00:23japerry has joined #ltsp
00:27zirconiumks has quit IRC
00:39cyberorg has joined #ltsp
01:13alkisg has joined #ltsp
01:31alkisg has quit IRC
01:33F-GT has quit IRC
01:35F-GT has joined #ltsp
02:01shrek has joined #ltsp
02:45
<Ryan52>
warren, I have fully working code for the language localization, but it's inefficient and stupid (ldm reads the entire file once for every language that ldminfod gives it), and I have a patch to make it load the info from the file into a hash once and then use that, but it doesn't work and makes ldm crash...and I'm done working on it for today...I'm gonna get it figured out tomorrow, tho.
03:46nubae has joined #ltsp
04:21alekibango has quit IRC
04:24Egyptian[Home] has quit IRC
04:30Egyptian[Home] has joined #ltsp
04:33Egyptian[Home] has quit IRC
04:41hanthana has joined #ltsp
04:44sep has joined #ltsp
05:05alekibango has joined #ltsp
05:57gate_keeper_ has joined #ltsp
05:57sep has quit IRC
06:56shrek has left #ltsp
07:20
<nubae>
hmmm whats the ltsp twitter name again?
07:20|Paradox| has quit IRC
07:44bobby_C has joined #ltsp
07:57hanthana is now known as hanthana|afk
08:02alkisg has joined #ltsp
08:04nubae has left #ltsp
08:04ogra has quit IRC
08:05ogra_ has joined #ltsp
08:05ogra_ is now known as ogra
08:13cyberorg has quit IRC
08:16nubae has joined #ltsp
08:16cyberorg has joined #ltsp
08:26alkisg has quit IRC
08:35alekibango has quit IRC
08:49pmatulis_ has joined #ltsp
08:49pmatulis_ has quit IRC
09:12cyberorg has quit IRC
09:15cyberorg has joined #ltsp
09:28try2free has joined #ltsp
09:42petre has joined #ltsp
09:45hanthana|afk is now known as hanthana
09:51litlebuda has joined #ltsp
09:59try2free has left #ltsp
10:03CAN-o-SPAM has quit IRC
10:14try2free has joined #ltsp
10:16try2free has left #ltsp
10:27nubae has quit IRC
10:32nubae has joined #ltsp
10:37nubae has quit IRC
11:03RobertLaptop has joined #ltsp
11:05nubae has joined #ltsp
11:07RobertLaptop has quit IRC
11:07RobertLaptop has joined #ltsp
11:11nubae has quit IRC
11:24alekibango has joined #ltsp
11:25vagrantc has joined #ltsp
11:35markit has joined #ltsp
11:36
<markit>
hi, is it possible / easy / how to :) have a thin client boot with ltsp, and then imediatly go with rdp to another host?
11:36dirigeant has joined #ltsp
11:36
<johnny>
yes..
11:36
you have to set some variables in lts.conf..
11:36
<markit>
johnny: I've found something in the wiki, but did not understood
11:37
<johnny>
i'm not sure what they are tho
11:37
<markit>
johnny: can you point me somewhere more precise?
11:37
<johnny>
sorry no
11:37
<markit>
johnny: ok, thanks for trying :)
11:37
<johnny>
i don't do anything with windows
11:37
just stick around
11:37
<ogra>
wasnt that in the docs ?
11:37
<markit>
johnny: lucky man
11:37
<ogra>
!docs
11:37
<ltspbot>
ogra: "docs" is For the most current documentation, see http://wiki.ltsp.org/twiki/bin/view/Ltsp/LtspDocumentationUpstream
11:37
<johnny>
also.. try the posts on ltsp-discuss
11:37
or whatever it is called..
11:38
<markit>
ogra: let me check the manual, thanks
11:38
I usually consider google and wiki the only source of knowledge :)
11:38
<ogra>
look for rdesktop, if its in there that should be mentioned
11:38
<vagrantc>
ogra: i think you used the old FSF address in your copyright header updates...
11:38
<ogra>
vagrantc, eek
11:39
i just copy pasted :P
11:39
<markit>
ogra: mmm "rdp" is not present in the manual, strange
11:39
<ogra>
"rdesktop"
11:39
<vagrantc>
ogra: lintian will catch it :)
11:40
<ogra>
vagrantc, well, i didnt build packages, just since some days ago it came up when who did what here i decided to take the blame and add the headers to all files
11:40
<vagrantc>
ogra: thanks for doing that! :)
11:40
<ogra>
its far from ebing done :)
11:40
*being
11:40
<vagrantc>
although i forsee a complicated debian/copyright file coming up...
11:40
<ogra>
i only attacked the toplevel dirs yet
11:41
well, if i need to touch them again i should probably look at the new debian/copyright automation system and make it match best :)
11:42Eghie has joined #ltsp
11:43
<markit>
ogra: thanks, much better :)
11:44
mm just one occurrence
11:54markit has quit IRC
11:54klausade has quit IRC
11:54pscheie has quit IRC
11:54pmatulis has quit IRC
11:54shamino has quit IRC
11:54urandom has quit IRC
11:54markit has joined #ltsp
11:56urandom has joined #ltsp
11:56pmatulis has joined #ltsp
11:56shamino has joined #ltsp
11:58hanthana has quit IRC
12:00pscheie has joined #ltsp
12:06bobby_C has quit IRC
12:14
<markit>
I've found a setup, but don't tell in what file to enter: http://wiki.ltsp.org/twiki/bin/view/Ltsp/Rdesktop
12:16
<ogra>
the wiki doc are all for ltsp 4.x
12:16
<markit>
lts.conf maybe
12:16
<ogra>
*docs
12:17
<markit>
I think I've solved, thanks :)
12:36alekibango has quit IRC
12:56otavio has quit IRC
12:56otavio has joined #ltsp
13:04
<markit>
I'm puzzled.. if I set screen_02 with rdesktop, it boots with rdesktop only (with keyboard not working)
13:04
if I set screen_01 ldm
13:05
then I get an error "screen_01 contains a creen script ldm. Please change to another screen"
13:05
what am I doing wrong?
13:05
(SCREEN_01 uppercase)
13:05
<vagrantc>
markit: use SCREEN_02-SCREEN_10
13:06
markit: if you define no SCREEN_*, it defaults to SCREEN_07=ldm
13:07
<markit>
ah, not screen_01? what if I want to change the main login method? do I have to define screen_01 or screen_07?
13:07
<johnny>
7 for graphics
13:08
01 is reserved on ubuntu
13:08
<vagrantc>
though 7 is only a convention
13:08
<johnny>
for now at least..
13:08
we should move to 01 at some point
13:08
for kms stuff
13:08
<vagrantc>
markit: what is your goal, for the moment?
13:12nonix4 has joined #ltsp
13:13
<nonix4>
Umm is there a way to actually use ltsp-build-client that doesn't end up either building ridiculously oversized images or failing because of random http errors?
13:14
... hardy's one at least either fails using http or tries to put the whole mirror there if using file:///
13:16
<markit>
vagrantc: now works like a charm, thanks a lot
13:17
vagrantc: my goal is that a friend of mine has to provide rdp clients for a customer of him
13:17
and asked me about a gnu/linux based solution
13:17
and I'm experimenting about having the clent to automatically load ltsp and go in rdp login to M$crap server
13:17
later I will try with usb and sound (crossing my fingers)
13:18
now using screen_07 instead of _01 has solved the "deak keyboard" problem
13:18
I've to run, thanks again
13:18markit has quit IRC
13:19alkisg has joined #ltsp
13:20* nonix4 thinks a while... --skipimage could be what is needed?
13:20nonix4 has quit IRC
13:21nonix4 has joined #ltsp
13:23RobertLaptop has quit IRC
13:26keithclark has joined #ltsp
13:32RobertLaptop has joined #ltsp
13:43keithclark has quit IRC
13:47
<ogra>
vagrantc, ugh https://lists.ubuntu.com/archives/edubuntu-users/2009-January/005074.html
13:50
<laga>
ogra: ughß
13:50
?
13:51
<ogra>
the attempt is noble ... thugh the code is ugh....
13:51
*though
13:53
<laga>
heh
13:53
i was wondering about code quality when i saw the .tgz
13:55nubae has joined #ltsp
13:59
<vagrantc>
sure is nice to have all those .original files!
14:00
<ogra>
heh
14:00* vagrantc wonders why they didn't just write a plugin to hook into the regular initramfs-tools...
14:00
<vagrantc>
rather than mangling everything...
14:01
<ogra>
right
14:01
and use update-initramf with -b instead of shuffling the files around
14:01
*update-initramfs
14:02
<vagrantc>
clearly, they just need more experience coding, is all, i guess.
14:02
<ogra>
yep
14:02
but the way it is it can easily trash your chroot ... he should probably have mentioned to make a backup before in his mail
14:03
<vagrantc>
sure.
14:06
<laga>
maybe we can add that code to automatix
14:06
<ogra>
lol
14:06
<laga>
"oh hai, we re-invented the wheel. it'll break your box"
14:06
;)
14:11
<warren>
ogra: please, let us not merge any insanity like that.
14:11
aside from obvious issues like it is impossible for wlan to PXE
14:11
It doesn't belong in ltsp
14:11
<ogra>
warren, lol, definately not
14:11
<warren>
nor should it even be called LTSP
14:11
<ogra>
ove my dead body
14:11
+r
14:12nonix4 has quit IRC
14:12
<ogra>
it would be nice to have something like that as external project that can hook into ltsp ... but first it shold be coded in a sane and non destroying way
14:13
<laga>
maybe we can add network-manager to the initramfs
14:13* laga hides
14:13
<ogra>
now thats a good idea !
14:13
we'll do that when it entered the ubuntu and fedora server releases
14:14* Ryan52 sighs at the Debian packaging for ldm...it breaks my changes.
14:14
<ogra>
fix it :)
14:16nonix4 has joined #ltsp
14:17
<vagrantc>
Ryan52: what's the issue?
14:17
<Ryan52>
vagrantc, I'm just generating the data in the top directory, so "DESTDIR=../debian/ldm/" doesn't work...3 options to fix it. a) DESTDIR=$(shell readlink -f ../debian/ldm/) # or something like that, I suck at make files. b) replace the "make install" line with dh_auto_install, requiring changing the debhelper depends/compat to 7, or c) just replace the whole debian/rules with a dh7 version. which one do you want me to do?
14:18
<vagrantc>
Ryan52: i'm fine with debhelper 7.
14:18
<Ryan52>
okie doke.
14:18
<vagrantc>
Ryan52: since i have no intention of backporting current versions to etch.
14:19
<ogra>
there is no backport of dh7 in debian ?
14:19
<Ryan52>
there is
14:19
<ogra>
ubuntu has one
14:19
ah
14:19
<vagrantc>
in any case, i have no intention of backporting ldm anyways, so it's moot :)
14:20
i mean, i'll backport to lenny, but that has dh7
14:20
<ogra>
ha !
14:20* ogra found the proper debian bug for his arm issues :)
14:21
<Ryan52>
I guess I finally get to figure out if I can push to the packaging repo :)
14:23
<vagrantc>
probably not :)
14:23
<warren>
laga: we're eventually adding NetworkManager to our initrd
14:23
laga: probably in a year or two
14:23
<vagrantc>
whoah.
14:23
<warren>
we already have NetworkManager in gdm
14:23
<ogra>
warren, and your server team allows that ?
14:24
<warren>
ogra: NetworkManager will eventually become a complete replacement
14:24
<ogra>
or do you start creating different initrds ?
14:24
heh, QED
14:24
<warren>
ogra: NetworkManager is already default here
14:25
<ogra>
i mean its great for desktop setups ...
14:25
<warren>
ogra: the people upset are becoming fewer and fewer
14:25
<nubae>
well, I'm still upset
14:25
<ogra>
but i want to see it five network cards with different routes and sub routes
14:25
<warren>
it is interoperable with our old way of configuring network interfaces now
14:25
<ogra>
*see it handle
14:25
<nubae>
and just 2 days ago I had a bunch of people bitching about intrepid and nm 0.7
14:26
<warren>
ogra: yeah, it upsets people where it doesn't handle corner cases. Last year that corner just happened to be 75% of cases. =)
14:26
<ogra>
nubae, well, the prob in ubuntu is that nobody actually makes the complete cut
14:26
<warren>
NM upstream implements its own thing, and your own distro has to implement stubs to interoperate with the old way you configured networks
14:26
<vagrantc>
Ryan52: the packaging repositories should be group-writeable now.
14:26
<ogra>
suckyness in ubuntu is mainly because NM isnt fully there yet and they try to retain all opportunities working
14:26
<nubae>
well, I've started deleting everything but it, and then it seems to be ok
14:27
<Ryan52>
kk
14:27
<nubae>
but u cant have remnants of anything else
14:27
<ogra>
so you have two half breeded solutions that interfere
14:27
<warren>
Sounds like Ubuntu didn't implement the interop tihngs
14:27
<nubae>
but what Ive ended up having to do is, delete networking completely on my laptop and leave just nm
14:27
<warren>
ogra: I personally use both service network and service NetworkManager and it doesn't interfere.
14:27
<nubae>
and on my server delete nm and leave just networking
14:27
<ogra>
it did, but many of them are not even nearly as powerful as the debian model
14:28
<warren>
ifdown and ifup here are integrated with NM
14:28
<ogra>
and getting somwthing right on top of the debian network design isnt easy
14:28
<nubae>
they are supposed to be on buntu too... but they are not
14:28
<vagrantc>
Ryan52: there were a few files i didn't have permission to change, but those are (i hope) all just old tagged releases and such.
14:28
<ogra>
lots and lots of failed attempts can be found in debians history
14:29
you will never see ubuntu server use NM
14:29
likewise for debian
14:29
<Ryan52>
DEBVER=$(shell dpkg-parsechangelog |sed -n -e '/^Version:/s/^Version: //p')
14:29
export ENABLE_NLS=1
14:29
vagrantc, what are those for?
14:29
<ogra>
warren, they are too in ubuntu ... but our /etc/network/interfaces setup is so powerful that ifup/ifdown dont completely cover it
14:30
there are millions of corner cases
14:30
<vagrantc>
Ryan52: "copies" of released versions that pre-date tagging in bzr.
14:30
<nubae>
hey are there plans to get git working with launchpad?
14:30
<ogra>
and you can have a two or three digit number of tools that edit /etc/network/interfaces in different ways ...
14:31
<vagrantc>
Ryan52: oh, sorry, wrong question :)
14:31* Ryan52 was talking about the lines that he pasted...
14:31* ogra points nubae to #launchpad :)
14:31
<vagrantc>
Ryan52: the ENABLE_NLS=1 was needed to get translations working.
14:31
<Ryan52>
for which tool?
14:31
<nubae>
yy... but ogra is a supercow and is sure to give me a quick answer
14:31
:p
14:31
<vagrantc>
Ryan52: not sure if DEBVER is used anymore ... i *think* it was used for the bzr-release target or some such.
14:32
<ogra>
nubae, different team inside canonical :) i'm just a customer as you are :)
14:32
<nubae>
u really haven't heard anything about that?
14:32
<ogra>
i'm doing happy dances that we have signed PPAs now
14:32
<vagrantc>
Ryan52: the first time we got translations working at all, i needed ENABLE_NLS=1 in there.
14:32
<ogra>
i'm not intrested in git
14:32
nor is any of my colleages
14:33
<Ryan52>
ok.
14:33
<ogra>
and our kernel team works on their own server
14:33
<nubae>
ok, I'll go find out... and I'll give the answer here :p
14:33
<ogra>
but i really doubt it
14:33
<nubae>
/join #launchpad
14:33
oops
14:33
stupid empathy
14:33* alkisg wonders why the ubuntu alternate CD does install network manager, even with "install an ltsp server" option selected...:(
14:33
<ogra>
there are not many actual ubuntu devs that use git
14:33
<nubae>
well if any of them work on sugar there are
14:34
<ogra>
alkisg, because ltsp is just an addon to desktop and because NM is fine in that setup
14:34
especially on new installs
14:34bobby_C has joined #ltsp
14:34
<ogra>
NM is only actually problematic if you want backwards compatibility
14:34
but it will fully respect ltsp
14:35
<alkisg>
Bah... I always uninstall it, I had problems with it some time ago, and all I was trying to do was have 2 static ips :)
14:35
<ogra>
right *some time ago*
14:35
like three releases in the past :P
14:35
<vagrantc>
Ryan52: looks like ldm 2:2.0.1-1 is the first version in debian that had translations working...
14:35
<alkisg>
Now I'm using bonding, so again, it's useless.. :)
14:36
<ogra>
right, its crap for bonding
14:36
it works fine in a setup where you dont fiddle with your network setup on the low level
14:36
it wont touch interfaces defined in /etc/network/interfaces
14:37
<Ryan52>
is stgraber's translation stuff done?
14:37
<vagrantc>
Ryan52: no idea.
14:38
i haven't tested ldm-trunk since 2.0.28, actually.
14:43
<laga>
ogra: actually, the ltsp_over_wlan stuff doesnt look too bad.. other than the fact that it's not an initramfs hook and overwrites files
14:43
<stgraber>
Ryan52: not yet no ...
14:44
<Ryan52>
stgraber, is it taggable anyway?
14:44
<stgraber>
Ryan52: I still have to write that .pot/.po/.mo generator script and integrate it with the build process
14:44
Ryan52: probably, you just won't have any translation support for rc.d scripts
14:44* Ryan52 nods
14:46
<ogra>
laga, it does a lot of insane things
14:48
<vagrantc>
Ryan52: for some reason, it didn't get Esperonto, although the previous incarnations of ldminfod did.
14:48
Ryan52: generate.py, that is
14:48
<laga>
ogra: yes, but i expected it to be worse ;)
14:50
<Ryan52>
what's it's locale code?
14:50
<vagrantc>
Ryan52: "eo" is the locale code for esperonto.
14:50
Ryan52: no territory at all
14:51
<Ryan52>
oh.
14:51
hrm.
14:51
<vagrantc>
Ryan52: also curious if we want the @euro locales, as they are not UTF-8, and i think ldm assumes UTF-8 locales
14:52* Ryan52 wonders why this packaging has a binary-indep target that does stuff..
14:53
<vagrantc>
i guess have a small handfull of extra locales won't hurt anything.
14:54
Ryan52: most of the debian/rules came from very early revisions, so there's probably lots of cruft in there.
14:54
Ryan52: much of it goes back to revision 1 of the ltsp project.
14:55
<Ryan52>
o
14:55
<vagrantc>
Ryan52: i'd suggest just starting with the basic debhelper 7 debian/rules file, and adding the obvious specific stuff.
15:00
<Ryan52>
what do you mean by "@euro"?
15:00
I think my code ignores that.
15:01
<vagrantc>
./generate.py | egrep '@euro' | head -n 1
15:01
de_AT@euro Deutsch (Österreich)
15:01
<Ryan52>
P: ldm: no-homepage-field
15:02
should I add ltsp.org as ldm's Homepage?
15:02cliebow has joined #ltsp
15:02* Ryan52 <3 lintian's pedantic mode
15:02
<Ryan52>
:)
15:02
<cliebow>
whats the metapackage name for tools to compile etc?
15:03
<vagrantc>
Ryan52: sounds fine to me, though it would be ideal to find a specific page for ldm.
15:03
<Ryan52>
are there any?
15:03* Ryan52 doesn't think there are..
15:03
<vagrantc>
i don't think there are, but i bet we could make one for each of the major LTSP projects.
15:04
<Ryan52>
meh. that sounds like work :p
15:04
hrm. can I use debcommit with this bzr set up?
15:05
<vagrantc>
Ryan52: i don't *think* so, but i've never tried.
15:06
i think debcommit assumes a debian dir within the Vcs-* location
15:06
whereas the Vcs-* location is actually just the debian dir, in this case.
15:06
and the control file in generated, so i'd be surprised if debcommit could figure it out
15:07
<Ryan52>
how is that relevant?
15:07
woah
15:07* Ryan52 thinks vagrantc needs the machine readable copyright format
15:07
<vagrantc>
how would debcommit figure out where to commit to.
15:07
<Ryan52>
0.o I don't think it does that.
15:07
<vagrantc>
yeah, i haven't done the unpleasant work of rewriting all the copyright stuff. but ogra's recent commits will make that necessary :)
15:08
Ryan52: how does debcommit figure out where to commit to?
15:08
<Ryan52>
it just commits.
15:08
svn ci -m "blah"
15:08
$ grep Vcs `which debcommit`
15:08
$
15:08
<ogra>
doesnt it pull that out of control ?
15:09
james_w in #ubuntu-devel would know if he was around ...
15:09
<vagrantc>
Ryan52: well, i almost assure you, it won't work.
15:09
Ryan52: but there's one way to find out :)
15:09
<ogra>
vagrantc, feel free to leave it to me ... its a nice work for killing time on sundays
15:10
<vagrantc>
ogra: heh. :)
15:10* Ryan52 pushes
15:11
<Ryan52>
okay, so now for the languages that don't have a territory...hrm.
15:11
<vagrantc>
Ryan52: so it's smart enough to figure out that if debian/.bzr exists, ignore .bzr ?
15:12
<Ryan52>
dunno. I didn't use it.
15:12
<vagrantc>
ah. :)
15:14
<Ryan52>
so how do I figure out what's a valid locale code, then?
15:15
I was looking for things with a _, and that didn't give any false positives on Fedora, but I guess it does on Debian.
15:15
<vagrantc>
what's your current logic?
15:15
<Ryan52>
it gets rid of anything after a . or a space.
15:15
and then it looks for a _
15:15
<vagrantc>
Ryan52: it looks like you dropped the - ./configure --prefix=/usr --libexecdir=/usr/lib
15:15
<ogra>
sounds ok to me for a start
15:15
<vagrantc>
Ryan52: was that intentional?
15:16
<Ryan52>
yes, of course.
15:16
dh calls dh_auto_configure
15:16
which does that for us.
15:16
<vagrantc>
nice.
15:18
Ryan52: special-casing "eo" seems silly...
15:19
also appears to be "ia", whatever that is.
15:20Egyptian[Home] has joined #ltsp
15:21
<Ryan52>
so, can I just ignore them? :)
15:21
<vagrantc>
Ryan52: but those are the only two exceptions in found in /etc/locale.gen
15:21
which seems the same as /usr/share/i18n/SUPPORTED
15:22
Ryan52: well, invariably someone would file a bug on them.
15:25alkisg has quit IRC
15:26* Ryan52 fixed it
15:27alkisg has joined #ltsp
15:28
<Ryan52>
how do I do try and except without running any code in the except?
15:28* Ryan52 gets: IndentationError: expected an indented block
15:28
<Ryan52>
if I leave that part empty.
15:28cliebow has quit IRC
15:29
<johnny>
Ryan52, pass ?
15:29
<Ryan52>
thanks
15:29
<johnny>
not that i'm that big of a python person.. i could be wrong
15:29
<Ryan52>
it works!
15:31
vagrantc, have you actually tested my changes?
15:32
or just the generate.py part?
15:33Eghie has quit IRC
15:34
<vagrantc>
Ryan52: "pass"
15:34
Ryan52: just the generate.py part (which should maybe be renamed to generate_locale_list.py or some such?)
15:36
<Ryan52>
ok.
15:37
<vagrantc>
Ryan52: while we have no examples at this time, there are other languages with 3 letter codes :)
15:37
<Ryan52>
ugh.
15:37
<johnny>
the nighmares of language and timezones
15:37
:(
15:37* johnny doesn't envy Ryan52
15:37
<johnny>
Ryan52, try joining the tzdata folks if you want real fun
15:38
<vagrantc>
well, no examples without a corresponding territory.
15:38
<Ryan52>
heh
15:38
vagrantc, if and when that happens, we can figure something out :)
15:38
<vagrantc>
Ryan52: fair enough.
15:39
<re-G>
ltsp over wlan new version: https://lists.ubuntu.com/archives/edubuntu-users/2009-January/005074.html
15:39
<stgraber>
vagrantc: about that ltsp-genmenu thing, just go ahead and merge it, I'll tag and release a new ltsp in Ubuntu once I'm sure it works in my setups.
15:41
<vagrantc>
i'm hesitant to merge it, but as long as all other things simply append to XDG_DATA_DIRS, it should be fine...
15:41
<Ryan52>
okay, I think I'm ready to tag now.
15:41
vagrantc, can you test with ldm itself really quickly, just to make sure that I'm not insane? :)
15:42
it works for perfectly for me, but ya..
15:42
<stgraber>
vagrantc: IIRC desktop-profiles appends to XDG_DATA_DIRS, so as long as it's set before Xsession is called, it should be handled correctly
15:42
<vagrantc>
stgraber: yeah, it is.
15:42
<stgraber>
vagrantc: I've forwarded your mail to the guy doing most of our xdg stuff just to make sure and will try when I've a minute (I hope later today)
15:43
<vagrantc>
Ryan52: i could test later tonight, but not immediately...
15:43
<Ryan52>
oh, ok.
15:44* Ryan52 will just tag then
15:46
<stgraber>
ok, it's already -25C here, sounds like a good idea of going outside before it gets even colder :) See you later.
15:47
<ogra>
dont lick the metal poles
15:47jammcq has joined #ltsp
15:47
<alkisg>
Where is stgraber living? North pole? :O
15:47
<jammcq>
good afternoon folks
15:47
<ogra>
alkisg, not far, yeah
15:48
yo jam
15:48
<jammcq>
howdie oliver
15:48
<ogra>
:)
15:53Egyptian[Home] has quit IRC
15:53Egyptian[Home] has joined #ltsp
15:53nubae has quit IRC
15:58Egyptian[Home] has quit IRC
15:58nonix4 has quit IRC
15:59bobby_C has quit IRC
16:02nonix4 has joined #ltsp
16:03alkisg has quit IRC
16:03vagrantc has quit IRC
16:17vagrantc has joined #ltsp
16:19
<vagrantc>
Ryan52: well, maybe i can give it a whirl in the next 25ish minutes...
16:26nubae has joined #ltsp
16:32
<vagrantc>
Ryan52: or not.
16:32
<Ryan52>
ok
16:37
warren, ldm is ready to be tagged and built for F-9, F-10, and devel...should I "make tag && make build", or do you wanna look it over and/or test first?
16:39
<vagrantc>
7 minutes, package still building...
16:39
<Ryan52>
vagrantc, when you get time, the Debian package just needs "dch -r" and to be uploaded, I think :)
16:39
<stgraber>
vagrantc: I got confirmation from our xdg-guy that desktop-profiles will just append to an existing XDG_DATA_DIRS so your change should work just fine
16:39* johnny still needs tarballs..
16:39
<vagrantc>
Ryan52: yeah, basically testing that
16:42nubae has quit IRC
16:45nubae has joined #ltsp
16:45* vagrantc hears the buzzer
16:46* vagrantc goes to work
16:47topslakr has quit IRC
16:48
<vagrantc>
Ryan52: looks promising so far. will try for a real upload tonight.
16:48
<Ryan52>
ok.
16:49
<vagrantc>
may also have to add debian/changelog entries for 2.0.29+
16:55alekibango has joined #ltsp
17:08nubae has quit IRC
17:10vagrantc has quit IRC
17:18alekibango has quit IRC
17:21alekibango has joined #ltsp
17:25dirigeant has quit IRC
17:32topslakr has joined #ltsp
18:27
<stgraber>
Ryan52: I've done the Makefile for ldmrc, I'm pushing the code now. Do you happen to have some autotools knowledge ?
18:27
<Ryan52>
not really, but maybe some..
18:27
it depends :)
18:27
<stgraber>
my Makefile should be used as part of the build process so the .mo are generated but I have no idea how to do so ...
18:27vagrantc has joined #ltsp
18:28
<Ryan52>
stgraber, push it and I'll play with it.
18:28* Ryan52 thinks he might know how.
18:29
<stgraber>
looking at some other dirs, looks like automagic is looking for Makefile.am in all directories, so I just renamed it for now
18:29
no idea if that works though :)
18:31
didn't work :(
18:33
<ltsppbot>
"vagrantc" pasted "debdiff ldm_2.0.28-1_i386.changes ldm_2.0.31-1~20090131.1_i386.changes" (22 lines) at http://pastebot.ltsp.org/228
18:33
<vagrantc>
Ryan52: ^^ looks like ldmgtkgreet got relocated...
18:33
yet it still seems to work.
18:34
<Ryan52>
weird.
18:36
<stgraber>
Ryan52: renaming it to Makefile.in and adding it to the configure.ac seems to work
18:36
<Ryan52>
vagrantc, push @opts, "--libexecdir=\${prefix}/lib/".sourcepackage();
18:36
hrm..
18:37
*shrug*. we can override it.
18:39
<stgraber>
Ryan52: generate-local-list failed here
18:39
<Ryan52>
what's the error?
18:40
<ltsppbot>
"stgraber" pasted "generate_locale_list.py failed" (13 lines) at http://pastebot.ltsp.org/229
18:41
<Ryan52>
hrm.
18:43
vagrantc, pushed fix.
18:43
stgraber, do you know how to fix it? :)
18:43
cause I don't...:p
18:43
don't I have try/except around it?
18:44
<stgraber>
Ryan52: I'll try to fix it, for now I just commented the locale part in my Makefile as I was testing something else :)
18:44
<Ryan52>
ok.
18:44
if you figure out how to fix that, then please fix the other place where it's doing try and except too :)
18:45
<stgraber>
Ryan52: hmm, can you check why my rc.d po changes and see why autotools doesn't seem to use it ? the Makefile.in is correctly change to Makefile by the autotools but doing a make doesn't call my makefile ...
18:46
<Ryan52>
I'll try :)
18:47|Paradox| has joined #ltsp
18:50* Ryan52 thinks he has it fixed..
18:52
<Ryan52>
stgraber, add po/rc.d to the SUBDIRS list in Makefile.am?
18:54nonix4 has quit IRC
18:54nonix4 has joined #ltsp
18:55
<stgraber>
Ryan52: fixed your python issue
18:55
<Ryan52>
yay!
18:55
thanks.
18:57
<stgraber>
though, it works fine on my system but may break if the system isn't utf-8
18:57
Ryan52: sounds like what I missed :) Trying with it added to Makefile.am now
19:00
<Ryan52>
uhh...0.o. wth?
19:00
<stgraber>
oh, looks like I commited from the wrong directory :)
19:00
commiting the real fix now :)
19:01
<Ryan52>
heh. ok.
19:02
<stgraber>
Next time I do a copy of ldm-trunk to ldm-test I should make sure I don't keep the .bzr :)
19:03
<johnny>
uhmm? shouldn't you just do bzr branch?
19:03* johnny is confused
19:04
<stgraber>
johnny: nope, it's for build testing, I should actually do a bzr export or something similar
19:04markit has joined #ltsp
19:04
<johnny>
ah yeah .. export is good
19:05
export in bzr,git,mtn,hg is simpler than what it takes for svn and cvs
19:05
<stgraber>
Ryan52: worked, I have the .mo generate this time
19:05
<markit>
hi, I want to have local sound in rdesktop session. I run a -r sound:local, but no sound, even if I've the speaker icon :( anything further to do?
19:05
ltsp5 dedian sid
19:05
<stgraber>
I'll just make sure all scrippts use gettext now and rebuild the template + translate to french
19:06
then push that and tag a new release so my customers will stop naging me on Monday :)
19:07
<Ryan52>
stgraber: http://slexy.org/view/s2wgpe4gea
19:08
<stgraber>
oh, looks like we've a problem somewhere :)
19:08
<Ryan52>
ya.
19:09
maybe what you're doing is causing more of the exceptions that I'm catching?
19:09
(exceptions, or whatever you python people call them)
19:09
<stgraber>
hmm, is the function I touched called from inside a try/except ?
19:09
<Ryan52>
yes :p
19:10
<stgraber>
hmm, ok, I'll try with replacing those "pass" by a "raise" :)
19:12petre has quit IRC
19:18
<Ryan52>
hrm. if I don't pipe it into anything, and don't have it reopen stdout to a file, it (might) work normally.
19:19
I can't tell.
19:34
<stgraber>
pushed a fix ... basically adding another except ...
19:34
I also added a line defining two variables one containing the current locale and the other the current encoding
19:34
I thought you may like to use that instead of parsing the environment
19:34alekibango has quit IRC
19:36
<stgraber>
I read a few pages about encoding with python and it seems that in our case we have input files with varying encoding making python's encoding converter to fail ...
19:37
even with using different combinations of encoding and decoding
19:38
<Ryan52>
why did you add "language, output_encoding = locale.getdefaultlocale()"?
19:42
<stgraber>
so you can use "language" instead of os.getenv["locale"] if you want
19:42
(but haven't done that change)
19:43
<Ryan52>
I'm not getting anything out of the environment, I'm just putting stuff into it to change the locale. (which is a horrible hack, but ya..)
19:47
<stgraber>
oh, ok :)
19:48
dropped it
19:49litlebuda has quit IRC
19:59markit has quit IRC
20:01
<stgraber>
translations should be working now, I just pushed the complete fr.po and will tag later tonight or tomorrow
20:25sbalneav has joined #ltsp
20:25
<sbalneav>
Evening all
20:26
<stgraber>
evening sbalneav
20:26
<jammcq>
Scotty !!!!!!!!!!!!!!!!!!!!!!!!!!!!
20:28
<sbalneav>
Hey jammcq!
20:28
<vagrantc>
sbalneav: !!
20:28
jammcq: !!
20:28
<jammcq>
hey vagrantc
20:28* vagrantc waves
20:30adrianorg has joined #ltsp
21:04phantom has joined #ltsp
21:04F-GT has quit IRC
21:12
<warren>
Ryan52: just do it, you're co-maintainer
21:12
Ryan52: i'll test after built
21:12nubae has joined #ltsp
21:13
<Ryan52>
ok.
21:16
BUILDPATH CLEAN: removing /tmp/tmp.bhu015HoZG/ldm-2.0.32/po/rc.d/Makefile.in
21:17
mkdst removes it, and then stuff fails because it's not there.
21:17
gah, it's in our .bzrignore
21:19* Ryan52 pokes around
21:20
<warren>
was Makefile.in .bzrignore before?
21:20
<Ryan52>
yes
21:20
stgraber added po/rc.d/Makefile.in
21:21adrianorg has quit IRC
21:21
<warren>
only generated stuff should be in .bzrignore
21:22
quite possibly Makefile.in should be removed from .bzrignore
21:22
<Ryan52>
we want it removed everywhere else, tho.
21:22* Ryan52 thinks he figured out how to fix it
21:22
<warren>
Ryan52: either that, or mkdst needs to be sensitive to pathnames
21:23
<Ryan52>
copying the way it happens in po
21:23
<warren>
mkdst can't handle pathnames in .bzrignore
21:23
which is a mkdst bug
21:23
<Ryan52>
bzr handles it the same way.
21:24
anyway, I got it.
21:26
<stgraber>
I'll do some upstream change now btw
21:26
having a Makefile in a sub-sub-directory doesn't seem to work well with DESTDIR ...
21:26
<Ryan52>
yes it does.
21:26
well, wait.
21:26
fix your packaging.
21:27
my locales stuff is also broken cause it's in the topdir.
21:27
but it's the debian/rules that's broken.
21:27
2:17 < Ryan52> vagrantc, I'm just generating the data in the top directory, so "DESTDIR=../debian/ldm/" doesn't work...3 options to fix it. a) DESTDIR=$(shell readlink -f ../debian/ldm/) # or something like that
21:27
, I suck at make files. b) replace the "make install" line with dh_auto_install, requiring changing the debhelper depends/compat to 7, or c) just replace the whole debian/rules with a dh7 version. which one do yo
21:27
u want me to do?
21:27
stgraber: pick your choice :)
21:27
(the redone debian/rules with dh7 is done in vagrantc's repo, if you wanna steal it)
21:28
<stgraber>
I guess I'll just go with dh 7 :)
21:28
<Ryan52>
really to fix it you just need to use a full path for DESTDIR, in some way.
21:30* Ryan52 is going to tag another version to rename the Makefile.in
21:30
<Ryan52>
stgraber, anything else before I tag?
21:30
<stgraber>
just checking, wait a sec :)
21:31
ouch, moving from dh4 to dh7 :)
21:31
<vagrantc>
ouch?
21:31
<Ryan52>
heh
21:33
<stgraber>
well, I didn't think we still had debian/compat == 4 for ldm in Ubuntu :)
21:33
<vagrantc>
how many more versions of ldm-trunk do people plan to tag tonight?
21:33
i didn't think debian/compat was needed anymore...
21:33
<stgraber>
vagrantc: so your debian/rules is all dh7 ?
21:33
<Ryan52>
vagrantc, it is.
21:34
<vagrantc>
stgraber: the ldm-experimental-packaging branch, yes.
21:34
stgraber: nothing released yet, but if upstream ever calms down again, maybe i could :)
21:36
<Ryan52>
warren, can the datestamp thing be automatically incrimented somehow?
21:36
(without me writing a script :) )
21:40* Ryan52 wonders how po/Makefile works..
21:40
<vagrantc>
configure.ac:40: required file `po/rc.d/Makefile.in' not found
21:41
ldm 2.0.32 doesn't work so well for me...
21:41
<Ryan52>
yes, I know, I'm fixing it.
21:41
that's because mkdst removes it.
21:41
because it's in .bzrignore
21:41
<vagrantc>
oh hell.
21:41
<Ryan52>
so I tried renaming it to Makefile.in.in like is on po/
21:41
but it didn't work..
21:42
in po/, Makefile.in.in because Makefile and Makefile.in
21:42
but in po/rc.d/ it only becomes Makefile.in
21:42
what kind of black magic is this? 0.o
21:42
<vagrantc>
always looked weird to me.
21:43
<Ryan52>
apparently it's normal.
21:49
<vagrantc>
how important are the fixes since ldm 2.0.31 ?
21:50
<Ryan52>
just alkisg's race condition and my insane language mess. :)
21:50
<stgraber>
Ryan52: that was 2.0.31
21:51
2.0.31 => 2.0.32 is mainly rc.d translations and a small bugfix on the local generator
21:51
*locale
21:51
<Ryan52>
oh, I thought he said in.
21:55
<stgraber>
btw, I just hacked around dh4 for now, I'll work on merging vagrantc's dh7 branch later
21:55
btw, $(shell readlink -m debian/ldm/) that was
21:55
<vagrantc>
i'm wondering if i shouldn't just maintain a branch for debian that merges in each upstream version and keeps the debian dir in place.
21:55* Ryan52 nods
21:59
<vagrantc>
ok, i'll keep building against 2.0.31 with patches until y'all get the po stuff sorted out.
21:59
<Ryan52>
okay, pushed.
21:59
<vagrantc>
Ryan52: why didn't you just add the locale list generator thing to Makefiles to be installed into DESTDIR ?
22:00
<stgraber>
Ryan52: so what's it called then ? Makefile.in.in.in.in ? :)
22:00
<Ryan52>
stgraber, somethin like that :)
22:00
I have configure processing it from Makefile.in.in to Makefile.in and then to Makefile
22:00
doesn't seem to hurt anything
22:00
tho I still wonder how the one in po/ works..
22:01
vagrantc, why not do it the way I did?
22:01
<stgraber>
AFAIK the one in po/ uses even more magic than mine as it's completely handled by autotools/automake/...
22:01
<Ryan52>
ya.
22:01CaScAdE^1arAway has joined #ltsp
22:02
<Ryan52>
vagrantc, stgraber: anything else before I tag?
22:02* Ryan52 wants to make this the last tag of the day ;)
22:03
<stgraber>
I sent .32 to the builder already so nothing for me :)
22:03
<Ryan52>
do you not use mkdst? 0.o
22:04
<stgraber>
I do but something was wrong with my bzrignore it seems :)
22:04
so I actually have .32 with the .in in it ...
22:04
<Ryan52>
ah, ok.
22:04* stgraber wonders why bzr pull doesn't get .bzrignore
22:05
<Ryan52>
?
22:05CaScAdE^FarAway has quit IRC
22:06
<stgraber>
bah, nevermind something broke somewhere :) I seem to have a .bzrignore with the right content but I still got a .tar.gz with the Makefile.in in it ...
22:06
will make sure it doesn't happen next time I generate a .tar.gz :)
22:07
<vagrantc>
stgraber: older version of mkdst?
22:08* Ryan52 goes to tag
22:08Gadi has joined #ltsp
22:08
<Gadi>
vagrantc: hey - thx for the rigor... again :)
22:09* Gadi looked at the bzr branch and it looks good
22:10
<vagrantc>
Gadi: should probably just merge it, then...
22:11
<Gadi>
u bet
22:11
go ahead
22:11
i assume you tested it and it works
22:11
yes?
22:13
<vagrantc>
yeah
22:13
my only fear is if it breaks stuff
22:13
i'm gettiing funny characters for the japanese and chinese language selection in ldm... guessing i just need to install the right font, probably?
22:14
<Gadi>
well, I would say that any desktop that does not check for the existance of the var and simply overwrites it is in the wrong
22:14
<vagrantc>
little boxes with numbers...
22:14
<Gadi>
but, that said, we could always set XDG_HOME_DIRS
22:14
instead of XDG_DATA_DIRS
22:15
I would assume that no distros mess with XDG_HOME_DIRS
22:15
and leave that to the user
22:15
<stgraber>
Gadi: isn't this one even more likely to be overwritten ?
22:16
I checked and the tools using XDG_DATA_DIRS correctly append values to it (well, I only checked gnome :))
22:18* Gadi shrugs
22:18
<Gadi>
let's not let our fears get the better of us
22:18
let's push it upstream and see what happens
22:19
I think it is a preferrable approach to mucking with homedirs
22:22
<stgraber>
Ryan52: I have to do another upload of ldm, so I'll just use .33 when it's out (I forgot to add python-lxml as build-dep with the previous upload)
22:23phantom has quit IRC
22:23
<Ryan52>
oh, did I not push?
22:23
<stgraber>
Ryan52: I pulled a few minute ago and didn't get the .33
22:23* Ryan52 psuhes
22:23
<Ryan52>
*pushes
22:24
done.
22:24
<stgraber>
thanks
22:26
Ryan52: did you tag it ? I see the version commit but not the tag
22:26
oh, it's there
22:26
sorry
22:27
<vagrantc>
the hurrier we go, the faster we wait!
22:27
<Ryan52>
heh
22:31
<vagrantc>
ok, i tested 2.0.31 and it works, although some languages require additional fonts to display properly.
22:32
<Ryan52>
that makes sense.
22:34
oh, crap.
22:34
fwiw, it builds fine on Fedora...:)
22:34
but it don't build for me on Debian.
22:35
$ fakeroot make -f debian/rules clean
22:35
<stgraber>
except that missing build-dep it built fine on Ubuntu too :) the fixed one is on the builder at the moment
22:35
<Ryan52>
blows up on stgraber's makefile.
22:36
<stgraber>
what's wrong with my makefile again ? :)
22:36
<Ryan52>
no distclean target
22:36
<stgraber>
oh, so it fails for ltsp-trunk too then :)
22:36
<Ryan52>
which is needed for debian/rules clean target.
22:36
?
22:36
<stgraber>
it's a copy of ltsp-trunk/po :)
22:37
<Ryan52>
oh.
22:37
*shrug*
22:37
<vagrantc>
Ryan52: ever heard of debclean?
22:37
<Ryan52>
no.
22:37
there are too many scripts in devscripts :)
22:38
<vagrantc>
Ryan52: it's condsiderably shorter to type to run the above mentioned clean target :)
22:38
<Ryan52>
yes.
22:39* Ryan52 thinks vagrantc should just upload 2.0.31 (if it works and everything) and wait for 2.0.34 :)
22:40
<stgraber>
yeah, I'm using 9 builders at the same time :)
22:40
<Ryan52>
9?
22:40
<vagrantc>
Ryan52: what's wrong with 2.0.33 ?
22:40
<Ryan52>
vagrantc, clean target doesn't work.
22:41* Ryan52 thinks that it's a change with dh7
22:41
<Ryan52>
cause the old rules just ran "make clean"
22:41
but dh7 does distclean.
22:41
<vagrantc>
oh, fun.
22:41
<Ryan52>
it worked until stgraber's Makefile :)
22:41
s/stgraber/ltsp-trunk/ :)
22:41
<stgraber>
Ryan52: PPA build (3 archs) and Ubuntu build (6 archs / 8)
22:41
<Ryan52>
oh.
22:43
ubuntu has 8 arches? 0.o
22:44
<stgraber>
Ryan52: when exactly is iso-codes needed ? I added it in ldm and ldminfod but it looks like I also need it as build-dep.
22:44
<Ryan52>
only as a build dep, nothing more.
22:44
so not in depends.
22:44
<stgraber>
oh, so that scary commit message was wrong ?
22:44
<Ryan52>
it was right at the time.
22:44
:)
22:45
<stgraber>
ok :), one more upload on its way then ...
22:45
<Ryan52>
stgraber, don't you test first? :)
22:46
<stgraber>
I did, on my laptop so without a clean build chroot :)
22:46
I'm on my netbook and I don't have pbuilder on it ...
22:46
<Ryan52>
ah.
22:48
<vagrantc>
Gadi: there's still some unfinished business with ltspfs-trunk ...
22:49
<stgraber>
vagrantc: what doesn't work with it ?
22:49
<warren>
Ryan52: it does
22:49
<vagrantc>
stgraber: the local mounting stuff was unreliable, at least for me.
22:49
<warren>
Ryan52: that's why you put the symlink to the spec file in the *-trunk directory
22:49
<Gadi>
vagrantc: I think we need a time when we can focus on it and bang it out
22:50
<Ryan52>
warren, oh. then how am I supposed to build it? :p
22:50
<warren>
Ryan52: mkdst rpm will build whatever is in the current directory with an incremented datestamp
22:50
<Gadi>
there were woo many tests with unreliable hardware
22:50
<Ryan52>
oh, right.
22:50
<Gadi>
*too
22:50
<stgraber>
vagrantc: oh, ok. I'm using it with rdesktop at the office and it seemed to work.
22:50
<warren>
Ryan52: uncomment the other Release line and comment out the regular Release line
22:50* Ryan52 wasn't using mkdst
22:50
<Gadi>
stgraber: do you have any CDROMs there?
22:50
<Ryan52>
warren, ya, I got it.
22:51
<stgraber>
Gadi: nope, that's one of these white geode, so usbkey only.
22:51
the issue was only with cdpinger ?
22:51
<Gadi>
vagrantc had an issue with cdpinger
22:51
that *may* be the only issue
22:51
not sure
22:51
<vagrantc>
i can't remember...
22:52
we fixed a lot of things, but in the end, i didn't feel comfortable releasing it
22:52
stgraber: you made your tarballs without mkdst ?
22:53
<stgraber>
vagrantc: nope, I used mkdst
22:53
<vagrantc>
oh, ok.
22:53
i might just grab 2.0.33 and apply the patch in-line.
22:54
<stgraber>
yeah, this 2.0.33 upload actually built :)
22:55
<warren>
Ryan52: what was the solution to the Makefile.in?
22:55
<vagrantc>
Ryan52: did you review all the changes since 2.0.28 when you updated debian/changelog?
22:55
<Ryan52>
warren, make it turn Makefile.in.in into Makefile.in and then Makefile :p
22:56
vagrantc, oh, no, I only did from the last tag before mine..
22:56
<vagrantc>
Ryan52: ok, i'll review the previous changes, too, then.
23:01
<Ryan52>
warren, okay, ldm is building in all 3.
23:01
<warren>
Ryan52: ok cool
23:02
<Ryan52>
warren, do you wanna explain your testing proceedure for ltsp packages now?
23:02
<warren>
Ryan52: ok, just don't puke how ugly it is
23:02
<Ryan52>
heh. ok.
23:02hanthana has joined #ltsp
23:03
<warren>
hold a sec
23:03
i'm copying your ssh key
23:03
err
23:03
Ryan52: can you upload your ssh pubkey to a URL
23:04
<Gadi>
night, all
23:04Gadi has left #ltsp
23:05
<Ryan52>
ya, one sec.
23:05
warren: http://ryan52.info/~ryan52/authorized_keys
23:10redspike has quit IRC
23:14
<warren>
Ryan52: ok, you here?
23:14
<Ryan52>
yep
23:16
<johnny>
warrenz
23:16
hello
23:17* johnny expects a response
23:18
<johnny>
Ryan52, hello
23:18
<Ryan52>
hi johnny
23:18
<cyberorg>
o/
23:18
<johnny>
see :)
23:19* johnny kicks cyberorg
23:19
<johnny>
jk cyberorg
23:19
<cyberorg>
aww, what for?
23:19
<johnny>
i don't know what the o/ means
23:20* johnny thinks he is too bubbly for warrens taste
23:20
<johnny>
luckily the rest of you can handle me
23:20
<warren>
growl
23:20
<johnny>
hello warren
23:20
<warren>
omnomnomnomnom
23:20
<johnny>
food?
23:21
<cyberorg>
johnny, it looks like a person waving to me o/
23:21
<johnny>
oh
23:21redspike has joined #ltsp
23:21
<johnny>
it's some emoticon i don't know
23:21
<Ryan52>
the internetz are too complicated.
23:21
<johnny>
Ryan52, ?
23:21
<Ryan52>
"o/"
23:21
<johnny>
why do you think that?
23:21
oh
23:22
<cyberorg>
it could also be old pong (remember early days ping pong?)
23:22
<johnny>
cyberorg, Ryan52 is too young :)
23:23* cyberorg is too old
23:23
<johnny>
how old is that?
23:23
<cyberorg>
35
23:23
<johnny>
older than me
23:23
29 or somethin
23:23
<stgraber>
oh, I'
23:24
crappy keyboard :)
23:29
<johnny>
you were saying stgraber ?
23:29redspike has quit IRC
23:37
<cyberorg>
stgraber, how is cluster thing coming aling?
23:37
*along
23:37
<vagrantc>
uploading ldm 2.0.33 to debian experimental...
23:41
<stgraber>
johnny: it was some kind of copy/paste from another chan ... nothing targeted for that channel :)
23:42
cyberorg: very well, the LTSP integration side is perfect, I'm now working on the other components to make them work better (load-balancer, control-center, account-manager mainly)
23:43
<cyberorg>
stgraber, is it in a stage that we can include in our packages?
23:43
<stgraber>
and I'm still not sure if I'll upload the rest of ltsp-cluster to Ubuntu for Jaunty or not, the hard part was what's in the ltsp chroot, for the rest the PPA is currently the best way as there are quite a lot of changes there
23:43
cyberorg: probably, though you
23:44
you'll have to check for debian/ubuntu-isms :)
23:44
I think I got rid of most of them but there are probably some I'm not aware of
23:44
<cyberorg>
is there a doc that i can go through to understand what pieces go where?
23:45
<stgraber>
I have a doc about installing it in an OpenVZ environment (containers), I guess a good part of it is still up to date
23:45
https://wiki.stgraber.org/LTSP-Cluster
23:46
<cyberorg>
cool, will take a look and see if i can figure it out :)
23:48
<stgraber>
looking at that doc, it seems quite good actually :) The part about building the chroot is wrong though but you can get the right options by using --extra-help with upstream's ltsp-build-client (with Ubuntu's plugins)
23:49
control center is good but the attribute list provided in the bzr is out of date, I'm working on a python script to add and update the attributes instead which will make it a lot easier
23:49
loadbalancer is still good too, I have a better configuration file at the office that I'll commit upstream soon (better to load balance with multiple CPUs servers)
23:51
localapps menu has been merged and NX has not been updated with the nxloadbalancer I wrote two weeks ago (but it's not one of the key components)
23:51
<cyberorg>
i'll try and follow the bzr logs :)
23:53
<stgraber>
ok, just updated the NX part and dropped the ltsp localapps menu one
23:55
yeah, the PPA on Launchpad and the bzr is usually a lot more accurate than my doc :)
23:56redspike has joined #ltsp