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


Channel log from 30 January 2012   (all times are UTC)

00:05
<asmok>
sorry, i have to go
00:05asmok has left IRC (asmok!~ltsp@adsl-85-157-204-13.regionline.fi, Quit: Lähdössä)
00:09
<stgraber>
vagrantc: I synced ltsp-docs 1.0 in Ubuntu. Thanks for the release and upload
00:13
<vagrantc>
stgraber: cool.
01:05monteslu has left IRC (monteslu!~monteslu@ip68-109-174-213.ph.ph.cox.net, Quit: Leaving)
01:28vagrantc has left IRC (vagrantc!~vagrant@c-76-105-252-50.hsd1.or.comcast.net, Ping timeout: 255 seconds)
01:30vagrantc has joined IRC (vagrantc!~vagrant@c-76-105-252-50.hsd1.or.comcast.net)
01:41
<stgraber>
vagrantc: I plan on spending part of tomorrow dealing with LTSP in Ubuntu. Starting by trying to reduce the packaging delta on ldm, I might have some patches for you if I find some ways to bring Debian and Ubuntu even closer on ldm's packaging
01:46monteslu has joined IRC (monteslu!~monteslu@ip68-109-174-213.ph.ph.cox.net)
01:47Parker955_Away is now known as Parker955
01:51
<vagrantc>
stgraber: great!
01:51
stgraber: i've been meaning to get a new upload of ltsp into debian ... if we have aufs for the forseeable future, i can switch to NFS+aufs ...
01:51
stgraber: any idea if overlayfs works with NFS ?
01:54
<stgraber>
vagrantc: haven't tried. overlayfs is supposed to be much closer to a proper VFS overlay so it's more likely to work than the previous iterations
01:55
I also want to spend some time tomorrow testing the current trunk and see if it's in an uploadable state (as in, installs, boots, lets me login), if it's, I'll upload that and deal with the issues as they show up
02:00adrianorg has left IRC (adrianorg!~adrianorg@186.215.22.19, Ping timeout: 245 seconds)
02:03
<vagrantc>
stgraber: i managed to get trunk as of a few weeks ago working for debian, at least
02:14
<stgraber>
good, that's a bit reassuring :)
02:14
is that currently in Debian?
02:16
argh, just confirmed my fear, LTSP doesn't really work with resolvconf, now the default in Ubuntu ... looks like I'll need to hack our udhcp init script to deal with resolvconf :)
02:16* stgraber adds that to the TODO for tomorrow
02:25Mobe_ has left IRC (Mobe_!~Mobe@adsl-82-141-65-229.kotinet.com, Ping timeout: 276 seconds)
04:35
<dberkholz>
anyone gonna be at fosdem?
04:48Parker955 is now known as Parker955_Away
04:52freedomrun has joined IRC (freedomrun!~quassel@89.142.248.224)
05:12lifeboy_ has left IRC (lifeboy_!~roland@41.183.40.251, Remote host closed the connection)
05:46alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
06:19
<knipwim>
dberkholz: i plan to be at fosdem on saturday
06:21
<alkisg>
At many places we're trying to discover the ssh pid, the ltsp client hostname, ip address, the client $DISPLAY etc, and we go through hoops to get them.
06:21
Wouldn't it be easier instead of just "echo LTSPROCKS; /bin/sh -", to do something like:
06:22
"echo LTSPROCKS; echo ...a bunch of useful vars > ~/.ltsp-connection-$IP; exec /bin/sh -" ?
06:22
Then we'd just look at the user's home dir and have everything ready... (btw shouldn't we put the "exec" there anyway, to save a bash process?)
06:23
We could then easily implement an "ltsp-connections-info" script, or an "ltsp-exec" script, and LDM_LIMIT_ONE_SESSION could be rewritten better, etc etc
06:55loather-work has left IRC (loather-work!~khudson@wsip-98-175-250-115.sd.sd.cox.net, Quit: This computer has gone to sleep)
07:16Damianos has left IRC (Damianos!~Damianos@adsl-070-145-074-043.sip.cha.bellsouth.net, Remote host closed the connection)
07:41killermike has joined IRC (killermike!~killermik@2.26.103.109)
08:02map7_ has joined IRC (map7_!~map7@ppp118-209-93-114.lns20.mel4.internode.on.net)
08:03
<map7_>
alkisg, I got the fat client working last night. Think I'll use that and just install mythfrontend in the image.
08:04
<alkisg>
map7_: nice, any problems aside from applying the patch?
08:04
I think fat clients is the way to go, and even with a vbox frontend for managing the chroot ;)
08:06
<map7_>
no problems, I just edited the files manually. I might do separate patches so I can apply them in a auto build script
08:12
<alkisg>
!win32-loader
08:12
<ltsp>
alkisg: win32-loader: win32-loader.exe can be used to add a "PXE - Network boot" option to the Windows boot manager. It's developed for Debian and it uses gPXE/iPXE for the network booting. You can download a binary from http://ts.sch.gr/docs/kanonismoi-diadikasies/doc_download/327-ltsp-loader - (if someone can write a wiki page about it it'd be nice :))
08:20freedomrun has left IRC (freedomrun!~quassel@89.142.248.224, Read error: Connection reset by peer)
08:40loather has left IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net, Quit: This computer has gone to sleep)
09:06vagrantc has left IRC (vagrantc!~vagrant@c-76-105-252-50.hsd1.or.comcast.net, Ping timeout: 245 seconds)
09:09alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
09:13alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
09:28loather-work has joined IRC (loather-work!~khudson@wsip-98-175-250-115.sd.sd.cox.net)
09:40loather-work has left IRC (loather-work!~khudson@wsip-98-175-250-115.sd.sd.cox.net, Quit: This computer has gone to sleep)
10:00Steve_The_Pirate has joined IRC (Steve_The_Pirate!~Gary@ip-80-238-8-128.bskyb.com)
10:22freedomrun has joined IRC (freedomrun!~quassel@BSN-142-163-74.dial-up.dsl.siol.net)
10:30adrianorg has joined IRC (adrianorg!~adrianorg@186.215.22.19)
10:32Yet_another_Bill has left IRC (Yet_another_Bill!billy@nat/redhat/x-dkqxdanbnnnlislt, Ping timeout: 240 seconds)
10:32Yet_another_Bill has joined IRC (Yet_another_Bill!billy@nat/redhat/x-hvqxsutevkghnigm)
10:38ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de)
10:49dobber has joined IRC (dobber!~dobber@213.169.45.222)
11:15freedomrun_ has joined IRC (freedomrun_!~quassel@BSN-142-161-119.dial-up.dsl.siol.net)
11:16xsl has joined IRC (xsl!~silence@unaffiliated/xsl)
11:16freedomrun has left IRC (freedomrun!~quassel@BSN-142-163-74.dial-up.dsl.siol.net, Ping timeout: 272 seconds)
11:44killermike_ has joined IRC (killermike_!~killermik@2.26.103.109)
11:45killermike has left IRC (killermike!~killermik@2.26.103.109, Ping timeout: 252 seconds)
11:50killermike_ has left IRC (killermike_!~killermik@2.26.103.109, Ping timeout: 252 seconds)
11:51killermike_ has joined IRC (killermike_!~killermik@2.26.89.68)
11:57killermike_ has left IRC (killermike_!~killermik@2.26.89.68, Ping timeout: 252 seconds)
12:00killermike_ has joined IRC (killermike_!~killermik@2.26.100.222)
12:02[GuS] has joined IRC ([GuS]!~MysT@200.123.112.224)
12:02[GuS] has joined IRC ([GuS]!~MysT@unaffiliated/gus/x-663402)
12:04killermike_ has left IRC (killermike_!~killermik@2.26.100.222, Remote host closed the connection)
12:04killermike_ has joined IRC (killermike_!~killermik@2.26.100.222)
12:12khildin has joined IRC (khildin!~khildin@ip-80-236-228-68.dsl.scarlet.be)
12:18Steve_The_Pirate has left IRC (Steve_The_Pirate!~Gary@ip-80-238-8-128.bskyb.com, Remote host closed the connection)
12:30bengoa has joined IRC (bengoa!~bengoa@2001:1291:229:2:216:cbff:feab:6cc9)
12:39Steve_The_Pirate has joined IRC (Steve_The_Pirate!~Gary@ip-80-238-8-128.bskyb.com)
12:53bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at)
12:58monteslu has left IRC (monteslu!~monteslu@ip68-109-174-213.ph.ph.cox.net, Ping timeout: 252 seconds)
13:10monteslu has joined IRC (monteslu!~monteslu@ip68-109-174-213.ph.ph.cox.net)
13:22ncd209_ has left IRC (ncd209_!~ncd209@jetmaintain12.pndsl.co.uk, Quit: Leaving)
13:29mgariepy has joined IRC (mgariepy!mgariepy@ubuntu/member/mgariepy)
13:49Parker955_Away is now known as Parker955
13:54adrianorg has left IRC (adrianorg!~adrianorg@186.215.22.19, Ping timeout: 245 seconds)
14:08Parker955 is now known as Parker955_Away
14:11Trixboxer has joined IRC (Trixboxer!~Trixboxer@115.124.115.71)
14:44alexqwesa has left IRC (alexqwesa!~alex@alexo-veto.broker.freenet6.net, Quit: Š„Š°Š½Š° X'Š°Š¼ !!!)
14:48brunolambert has joined IRC (brunolambert!bruno@nat/revolutionlinux/x-wydnkygqikuzbwqc)
14:54dead_inside has joined IRC (dead_inside!~dead_insi@76.75.3.174)
15:15alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
15:19mmetzger has left IRC (mmetzger!~mmetzger@99-71-214-196.lightspeed.mdldtx.sbcglobal.net, Remote host closed the connection)
15:40darkpixel_ has joined IRC (darkpixel_!~darkpixel@65.100.44.217)
15:41darkpixel_ has joined IRC (darkpixel_!~darkpixel@curetheitch/staff/darkpixel)
15:42ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood)
15:44ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de)
15:49alexqwesa has joined IRC (alexqwesa!~alex@109.172.15.11)
15:53adrianorg has joined IRC (adrianorg!~adrianorg@186.215.22.19)
16:04adrianorg has left IRC (adrianorg!~adrianorg@186.215.22.19, Read error: Operation timed out)
16:10mmetzger has joined IRC (mmetzger!~mmetzger@99-71-214-196.lightspeed.mdldtx.sbcglobal.net)
16:14staffencasa has joined IRC (staffencasa!~staffenca@128-193-150-233.oregonstate.edu)
16:20loather has joined IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net)
16:31freedomrun_ has left IRC (freedomrun_!~quassel@BSN-142-161-119.dial-up.dsl.siol.net, Remote host closed the connection)
16:35toscalix has joined IRC (toscalix!~toscalix@85.137.159.216.dyn.user.ono.com)
16:37toscalix has left IRC (toscalix!~toscalix@85.137.159.216.dyn.user.ono.com, Remote host closed the connection)
16:38toscalix has joined IRC (toscalix!~toscalix@85.137.159.216.dyn.user.ono.com)
17:08toscalix has left IRC (toscalix!~toscalix@85.137.159.216.dyn.user.ono.com, Excess Flood)
17:09toscalix has joined IRC (toscalix!~toscalix@85.137.159.216.dyn.user.ono.com)
17:13dobber has left IRC (dobber!~dobber@213.169.45.222, Remote host closed the connection)
17:36freedomrun has joined IRC (freedomrun!~quassel@89.142.248.131)
17:50alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
17:56Steve_The_Pirate has left IRC (Steve_The_Pirate!~Gary@ip-80-238-8-128.bskyb.com, Ping timeout: 244 seconds)
17:57xsl has left IRC (xsl!~silence@unaffiliated/xsl, Quit: Connection reset by fear)
18:04ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood)
18:05ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de)
18:11khildin has left IRC (khildin!~khildin@ip-80-236-228-68.dsl.scarlet.be, Quit: I'm gone, bye bye)
18:15alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
18:41alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
18:51artista_frustrad has joined IRC (artista_frustrad!~fernando@200.247.43.2)
18:52vagrantc has joined IRC (vagrantc!~vagrant@c-76-105-252-50.hsd1.or.comcast.net)
18:53
<vagrantc>
stgraber: the dependencies and recommends are clearly from a different interpretation of what ldm-server provides...
18:54
<stgraber>
vagrantc: hehe
18:55
<vagrantc>
stgraber: i.e. it depends on openssh-server because i cannot think of a use-case that doesn't require ssh... it recommends xauth so that LDM_DIRECTX=true works, it recommends ltspfs so that the client-side ltspfs stuff works, and libasound2-plugins so that remote audio works...
18:55
it is a package to install to ensure that a machine can function with the LTSP defaults.
18:56
(and commonly used features)
18:56
which is exactly what recommends are for
18:56
<alkisg>
We don't have recommends on in debian/ubuntu ltsp chroots, do we?
18:56
<vagrantc>
y'all fly by different rules
18:57
but according to debian policies and best practices, i think i have those things set properly.
18:58
<stgraber>
vagrantc: yeah, I can see where these come from, would have preferred some ltsp-server-minimal or similar bringing these AND ldm-server
18:58Steve_The_Pirate has joined IRC (Steve_The_Pirate!~Gary@cpc16-haye17-2-0-cust169.haye.cable.virginmedia.com)
18:58
<vagrantc>
but those features are specific to a working LDM+LTSP environment, not the LTSP server.
18:59
we've got ltsp-server (which provides the minimalist functionality) and ltsp-server-standalone (which provides complete functionality)
19:00
and depends on ldm-server
19:00
the most common use-case (or at least, our default use-case) is an LTSP+LDM environment, and you get that by installing ltsp-server-standalone
19:01
but it's not unheard of to have an LTSP server that only exports the rootfs, and separate LDM servers that include applications and such without any rootfs stuff.
19:03
although i wish we could rename ltsp-server-standalone to ltsp-server and ltsp-server to ltsp-server-core or something like that ... but the names are pretty much impossible to reconcile.
19:04
we should maybe also think of spliting out the server-side localapps stuff into a separate package
19:05
since that should be installable in an LDM server without LTSP installed.
19:06* vagrantc hopes vagrantc hasn't flooded the channel
19:07
<vagrantc>
stgraber: you think it's feasible to go for gtk3 in the near future? the gtk maintainers in debian would be happy to get the conversion out of the way.
19:07adrianorg has joined IRC (adrianorg!~adrianorg@186.215.22.19)
19:08
<vagrantc>
stgraber: we might also want to make the autogen.sh stuff more forcefull, so that autoreconf stuff works "properly"
19:09
stgraber: or at least have a "really, go ahead and re-do everything" flag.
19:09
<stgraber>
vagrantc: sorry, got pulled into an IRC meeting
19:09
vagrantc: I'm 90% done porting to gtk3, the only broken code is the bit that sets the background
19:09
vagrantc: well, that an porting all the themes ;)
19:09
<vagrantc>
stgraber: excellent
19:10
stgraber: oh, regarding gtk2-engines-ubuntulooks vs. gtk2-engines ... the only reason i stuck with ubuntulooks all these years is the size of the package.
19:11
stgraber: installed size of 1m vs. 160k
19:11
but it would probably be better to switch the theme back
19:11
<stgraber>
vagrantc: yeah, I seem to remember you mentioning a dependency on clearlooks would pull quite a few packages in. I guess that's fine for Debian, it's just not doable in Ubuntu as ubuntulooks simply doesn't exist here ;)
19:12
<vagrantc>
stgraber: i'd rather sync than diverge at this point though.
19:12
stgraber: i'd really like to go with gtk2-engines-murrine, but that has a stupid recommends on murrine-themes, which is *huge*
19:13
highvoltage created ldm-themes which uses murrine, and i like it better ... just wish there was a Conflicts-Recommends
19:13
so i could exclude murrine-themes
19:13
but allow it to install if they really wanted it
19:14
<stgraber>
murrine-themes is a suggests of gtk2-engines-murrine in Ubuntu
19:15
<vagrantc>
http://bugs.debian.org/623783
19:15
wontfixed and closed.
19:15
stgraber: you lucky punks!
19:16
<stgraber>
well, we didn't want murrine-themes promoted to main apparently ;)
19:16* highvoltage emerges
19:16* vagrantc waves to highvoltage
19:17
<vagrantc>
stgraber: ah, so you wouldn't go with that either
19:17* highvoltage waves too
19:17
<vagrantc>
stgraber: oh! you mean gtk2-engines-murrine is in main, but the themes aren't
19:17
<stgraber>
vagrantc: right
19:18
<vagrantc>
it doesn't make any sense to me to have an engine recommend a theme
19:18
stgraber: if i can convince the maintainer otherwise, would you be willing to sync on murrine?
19:20
<stgraber>
vagrantc: yep, I don't mind going with murrine, it's in main and probably lighter than our current gtk2-engines (that's probably pretty big)
19:20
<vagrantc>
stgraber: i'll poke em on irc and see if i can't get them to reconsider
19:22* Hyperbyte waves to everybody
19:23
<vagrantc>
highvoltage: time to probably stir the muck with ldm-themes again, with freeze in a few months
19:23
highvoltage: dunno what the wheezy artwork scene is like
19:24
<highvoltage>
vagrantc: I've been following the lists
19:24
<vagrantc>
highvoltage: ah good, i haven't :)
19:24
<highvoltage>
vagrantc: the calls for artwork is only being announced now, so there's nothing yet to package
19:25Parker955_Away is now known as Parker955
19:25
<vagrantc>
highvoltage: that's the sort of thing that's generally freeze exception worthy, given that it's not generally technically difficult
19:25
<highvoltage>
vagrantc: great. I'd hope so :)
19:25
<vagrantc>
highvoltage: and the artwork isn't usually selected till pretty late in the cycle
19:33
<shawnp0wers>
So, does anyone have a script that can fix "stupid" ?
19:33
<vagrantc>
depends on the context
19:34* shawnp0wers has had a long day of "oh noes, when I minimize an application it dissappears because I apparently deleted the window selector
19:37Kens has joined IRC (Kens!c67def69@gateway/web/freenode/ip.198.125.239.105)
19:37
<highvoltage>
shawnp0wers: slay?
19:38
<Kens>
how to set up kiosk mode? Anyone?
19:40bengoa has left IRC (bengoa!~bengoa@2001:1291:229:2:216:cbff:feab:6cc9, Quit: Leaving.)
19:40brunolambert has left IRC (brunolambert!bruno@nat/revolutionlinux/x-wydnkygqikuzbwqc, Quit: brunolambert)
19:43brunolambert has joined IRC (brunolambert!bruno@nat/revolutionlinux/x-crvxulsripjbphck)
19:44
<vagrantc>
alkisg: if epoptes client doesn't handle openssl being missing, we may as well depend or at least recommend it.
19:44
<alkisg>
vagrantc: epoptes-client doesn't need openssl
19:45
<vagrantc>
alkisg: then it should gracefully error out when it tries to use it
19:45
alkisg: rather than spitting out command not found...
19:45
<alkisg>
Sure, I'll do that. The -c switch is just there to help the users, it's a "run-time" dependency
19:45
They may use "cp" instead
19:46
<vagrantc>
alkisg: doesn't seem an unreasonable recommends, to me.
19:46
<alkisg>
(italc doesn't even have an option like that, it suggests transfering the keys using usb sticks :P)
19:46
I was planning to put it to Suggests
19:46
<vagrantc>
suggests is close to pointless
19:46
<alkisg>
But no problem with recommends either
19:46
<vagrantc>
reccommends pulls it in by default usually, but allows them to not have it installed if they really don't want it
19:47
<alkisg>
It's not used by epoptes-client though... just by the user, and only once, and only if he isn't using the ltsp-build-client plugin that copies the keys, etc etc
19:47
<vagrantc>
alkisg: perfect case for recommends.
19:47
<alkisg>
OK
19:47
Recommends it is then :)
19:47
<vagrantc>
or does it pull in a lot of cruft?
19:47
<alkisg>
No idea it's usually preinstalled in ltsp chroots
19:47
I don't know why it wasn't there for him
19:48* vagrantc is surprised a full LTSP chroot wouldn't have it
19:49* alkisg is optimizing the thumbnails logic in epoptes... will be ready for a new release in a couple of days
19:49* vagrantc wonders if debian-edu knows about epoptes
19:50
<alkisg>
A couple of them do, we mentioned it in some edubuntu meeting when they were attending
19:50
<vagrantc>
i know they've been fighting with italc for years
19:57[GuS] has left IRC ([GuS]!~MysT@unaffiliated/gus/x-663402, Remote host closed the connection)
20:00ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood)
20:02ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de)
20:05adrianorg has left IRC (adrianorg!~adrianorg@186.215.22.19, Ping timeout: 272 seconds)
20:08Gremble has joined IRC (Gremble!~Ben@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com)
20:13
<vagrantc>
stgraber: well, the gtk2-engines-murrine maintainer doesn't really seem willing to consider the issue. ugh.
20:13
<stgraber>
argh...
20:14
what's his reason for having a gtk engine pull themes at install?
20:15
<vagrantc>
stgraber: i haven't seen one yet, other than "we disagree"
20:20toscalix has left IRC (toscalix!~toscalix@85.137.159.216.dyn.user.ono.com, Read error: Connection reset by peer)
20:24
<highvoltage>
alkisg: when can we do an edubuntu blog post on epoptes?
20:24
<alkisg>
highvoltage: I think it'd be best when the new version is published in precise... so maybe in a week?
20:25
<vagrantc>
stgraber: and 12:15 < Corsac> vagrantc: this is a *recommends*, not a depends
20:26
<highvoltage>
alkisg: ok, I'll poke you again in a week :)
20:26
<alkisg>
:)
20:26
<vagrantc>
stgraber: which doesn't seem to justify it any better to me
20:26
<stgraber>
vagrantc: yeah and with Recommends being installed by default it's still wasting space...
20:26
<vagrantc>
stgraber: for, as far as i can tell, no reason.
20:27
stgraber: given that the only things that depend/recommend gtk2-engines-murrine are ... other themes.
20:27
<stgraber>
vagrantc: you could create a package that Conflicts with murrine-themes and is itself a Recommends of ldm? That'd give you a Conflict-Recommends :)
20:27
<vagrantc>
stgraber: that would be heinously ugly, but it would work.
20:27
<highvoltage>
stgraber: didn't we at some point did that in Edubuntu :)
20:28
<stgraber>
highvoltage: kind-of, edubuntu-live does something like that ;) It's a package present at DVD build time but not present post-install, so we can blacklist stuff without preventing the user from installing them later one
20:28
*on
20:28
<vagrantc>
stgraber: given all that, i think we should just stick with clearlooks.
20:28
stgraber: which on debian is shipped in gtk2-engines
20:28* vagrantc wishes clearlooks was still packaged independently.
20:30
<stgraber>
right, same as Ubuntu, then, so we'd be in sync for that bit. I pushed the ltsp-cluster-info fix upstream, so that's another delta we get rid of. So the only thing left is debian/watch which I don't care about if we're in sync anyway
20:30
so with switching to gtk2-engines and dropping the current patch + using the latest upstream, Ubuntu could simply sync ldm from Debian
20:31
<vagrantc>
stgraber: well, the watch file was simply (maybe ubuntu shipped a newer upstream version) and (maybe debian shipped a new upstream version)
20:31
stgraber: that sounds great!
20:32
stgraber: so the security risk with the ssh-hostchecker scripts is that one server could impersonate another
20:32
<stgraber>
I already got Colin to unblacklist ldm in the package importer, so now it's like any other package in the archive, people are asked to merge as long as we have an Ubuntu delta and as soon as we don't, it gets directly imported from Debian
20:33
<vagrantc>
stgraber: how do you upload delta-less?
20:34
<stgraber>
Launchpad lets us "sync" from Debian, that basically copy the source package as-is from Debian and builds it, so we loose our Ubuntu changelog but from that point, any change in Debian is automatically imported (until the auto-import freeze)
20:34
<vagrantc>
got it
20:35
<stgraber>
after that, we can do manual syncs using "syncpackage -d sid ldm" (or experimental if we know it's what we want)
20:35
<vagrantc>
stgraber: you think it'll be synced for precise?
20:35
stgraber: or post-precise?
20:35adrianorg has joined IRC (adrianorg!~adrianorg@186.215.22.19)
20:36
<stgraber>
if we get a new upstream ldm by FeatureFreeze (mid-February IIRC) and you change to use clearlooks, yes, no reason we wouldn't sync
20:36
feature freeze is on the 16th of February
20:39
<vagrantc>
think it's almost time for ltsp 5.3.0 ?
20:39
<stgraber>
well, I'd prefer to make sure it works before we do that ;)
20:39
<vagrantc>
heh, well sure
20:40
<stgraber>
other than that, yes I think we may want to bump ltsp's version in the near future (still keeping LTSP 6 for when we have all of sbalneav new magic)
20:40
<vagrantc>
with the initramfs-scripts.d code, it pretty much assumes writeable / ... which i'd like to put off a little longer
20:40
<stgraber>
for ldm, I want to wait for it to be on gtk3 before doing any major version bump
20:41
<vagrantc>
stgraber: and alkis's boot the chroot to maintain the chroot stuff
20:41
although with the initramfs-scripts.d code, i think we're close with that.
20:53
highvoltage: i'm tempted to push a gtk2-engine-murrine-without-murrine-themes package doing what stgraber suggested.
20:54
it seems stupid, but it'd work...
20:55
it comes down to ubuntulooks being a few hundred k, clearlooks amounting to 1m, murrine (with murrine themes) 5m, murrine (without murrine themes) a few hundred k.
20:55
<highvoltage>
vagrantc: how about forking it to gtk2-engine-ltsp? then we can change it further?
20:55map7_ has left IRC (map7_!~map7@ppp118-209-93-114.lns20.mel4.internode.on.net, Ping timeout: 252 seconds)
20:55
<highvoltage>
(not sure if that's a good idea, but just throwing it out there anyway)
20:56
<vagrantc>
highvoltage: kind of want to avoid maintaining more than we have to
20:56
<highvoltage>
ok
20:56killermike_ has left IRC (killermike_!~killermik@2.26.100.222, Remote host closed the connection)
20:56
<vagrantc>
gtk2-engines/clearlooks seems to be a reasonable middle ground
20:59
stgraber: so should i package up trunk as ldm 2.2.5 today and get that rolling?
21:00
<stgraber>
vagrantc: sounds good. I haven't tested the current trunk with the new ssh stuff, but if that looks good to you, go ahead
21:04asmok has joined IRC (asmok!~ltsp@adsl-85-157-204-13.regionline.fi)
21:04
<vagrantc>
stgraber: i think the big difference with ubuntu and debian in the dependencies is how we treat ltsp-server vs. ldm-server ... i think debian makes it possible to use the LDM server vs. LTSP server functionality where appropriate, rather than assuming all LTSP servers provide all the ldm-server functionality.
21:11Gremble has left IRC (Gremble!~Ben@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com, Quit: I Leave)
21:14
<alkisg>
In ldm: "echo LTSPROCKS; /bin/sh -"
21:14
==> we could at least save a bash process by using "exec /bin/sh -", stgraber if you're checking the ldm code, could you try to put it there so that we have it for precise?
21:16
<muppis>
alkisg, booting with win32 loader and gpxe is terribly slow and sometimes hangs in kernel loading.
21:16
<stgraber>
alkisg: Pushed up to revision 1407.
21:16
<alkisg>
stgraber: thanks :)
21:17
<stgraber>
alkisg: you know you could have done it yourself right? :)
21:17
<alkisg>
stgraber: not the testing, not easily
21:17
I don't have the build-deps handy for precise
21:17
<vagrantc>
alkisg: you've tested that, right? :)
21:18
<stgraber>
alkisg: well, I consider most of the current changes in current ldm-trunk to be untested anyway
21:18
<alkisg>
Nope, not at all,it's just an idea
21:18
<stgraber>
alkisg: at least this one is going to be pretty visible and obvious if it fails ;)
21:18
<alkisg>
It should save us one bash process per user
21:18
Hehe right
21:18
muppis: I have a new one with fresh ipxe from git, here: http://people.ubuntu.com/~alkisg/boot/win32-loader.exe
21:18
<vagrantc>
actually, dash, no?
21:18
which is a much smaller savings
21:19
<stgraber>
yeah, dash in this case
21:19
<alkisg>
vagrantc: no, it's the ssh shell we're saving
21:19
And we're keeping dash
21:19
<vagrantc>
ah!
21:19
<stgraber>
ah yeah, we get rid of a bash instead of having bash + dash
21:19
<muppis>
alkisg, I'll try to member to try that tomorrow.
21:20
<alkisg>
I don't know if anyone saw a question I did some hours ago...
21:20
Exactly after echoing LTSPROCKS, we could write some env vars in a file in the user's home dir
21:20
E.g. in ~/.ltsp-connection.$PPID
21:20
For example, the ssh pid, the user's $DISPLAY, client hostname etc
21:21
So we could easily implement LDM_LIMIT_ONE_SESSION etc without going through hoops to get those vars
21:21
<vagrantc>
hmmmm.
21:21
<alkisg>
ltsp-exec, ltsp-connections-info...
21:24
muppis: you're using local network and tftp, not downloading the kernel over http, right?
21:24
It should have the same speed as using gpxe from any other media, i.e. it should download the kernel in a couple of seconds
21:25
<stgraber>
alkisg: yeah, I suppose we could do something like that. I'd rather have one file containing one PID per line where the PID is the one of sshd. We could probably have that file generated at each login using pgrep.
21:26
alkisg: then the rc.d hook can simply check if it's more than one line long and exit if it's
21:26
<alkisg>
stgraber: I'd like to have more info for the client there, not just the pid
21:26
E.g. "you are already logged in at $HOSTNAME, $VT..."
21:27
<muppis>
alkisg, yes. Using nic pxe everything works as should, but with gpxe it mostly hangs or is unusable slow.
21:27Trixboxer has left IRC (Trixboxer!~Trixboxer@115.124.115.71, Quit: "Achievement is not the end, its the beginning of new journey !!!")
21:28
<alkisg>
muppis: the win32-loader shouldn't affect the gpxe speed. So for any bad driver performance, you should mention that in #ipxe, after testing with the latest version
21:28
The ipxe speed isn't at all related to the boot loader
21:28
<muppis>
Ok.
21:31
Now need catch some sleep.
21:45Parker955 is now known as Parker955_Away
21:45alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
21:46asmok has left IRC (asmok!~ltsp@adsl-85-157-204-13.regionline.fi, Quit: Lähdössä)
21:56asmok has joined IRC (asmok!~ltsp@adsl-85-157-204-13.regionline.fi)
22:12asmok has left IRC (asmok!~ltsp@adsl-85-157-204-13.regionline.fi, Quit: Lähdössä)
22:16
<vagrantc>
stgraber: darn. ldm fails to build :(
22:16
<stgraber>
vagrantc: weird, can you paste the build log?
22:17
<vagrantc>
stgraber: something to do with AM_GLIB_GNU_GETTEXT not found in library
22:17
stgraber: configure.ac:38:
22:17
<stgraber>
hmm
22:18
<vagrantc>
i first tried without mkdst, and then i tried with, and both fail
22:19
<stgraber>
trying a test build here
22:19
<vagrantc>
maybe somethinng new in sid
22:19
<stgraber>
it builds fine here, though I haven't tried in a clean way
22:20
I did: bzr export /tmp/blah && cd /tmp/blah && apt-get build-dep ldm && sh autogen.sh && ./configure && make
22:30muppis has left IRC (muppis!muppis@viuhka.fi, Ping timeout: 240 seconds)
22:34brunolambert has left IRC (brunolambert!bruno@nat/revolutionlinux/x-crvxulsripjbphck, Quit: brunolambert)
22:52
<vagrantc>
hrm.
22:52
no luck with wheezy either...
23:05ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood)
23:06Steve_The_Pirate has left IRC (Steve_The_Pirate!~Gary@cpc16-haye17-2-0-cust169.haye.cable.virginmedia.com, Quit: Leaving)
23:06
<vagrantc>
so i tried a rebuild of 2.2.4 and that worked fine.
23:08ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de)
23:13
<vagrantc>
stgraber: what versions of auto* are you running?
23:15
<stgraber>
vagrantc: autoconf 2.68, automake 1.11.1
23:16freedomrun has left IRC (freedomrun!~quassel@89.142.248.131, Remote host closed the connection)
23:16
<vagrantc>
automake 1.11.2 here ...
23:17
that's not likely to be significantly different
23:17Steve_The_Pirate has joined IRC (Steve_The_Pirate!~Gary@cpc16-haye17-2-0-cust169.haye.cable.virginmedia.com)
23:18
<vagrantc>
here i was thinking to tackle the low-hanging fruit
23:19
config.status: error: cannot find input file: 'po/Makefile.in.in'
23:20
maybe my autoreconf stuff isn't calling the po stuff that mkdst did?
23:23
which was removed well before 2.2.4 ... so that's not really it
23:25Parker955_Away is now known as Parker955
23:25bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Ping timeout: 244 seconds)
23:35Steve_The_Pirate has left IRC (Steve_The_Pirate!~Gary@cpc16-haye17-2-0-cust169.haye.cable.virginmedia.com, Ping timeout: 276 seconds)
23:38
<vagrantc>
stgraber: build log: http://people.debian.org/~vagrant/ldm/
23:38
doh.
23:38
not yet.
23:43
ok, good to go, if you've got a moment to look...
23:51
stgraber: maybe it's the new ltsp-cluster stuff you added?
23:51
<stgraber>
vagrantc: could be but my test run was on current trunk
23:52dead_inside has left IRC (dead_inside!~dead_insi@76.75.3.174, Quit: Leaving...)
23:53* vagrantc tries without
23:54
<stgraber>
vagrantc: did you try calling autogen.sh instead of calling autoreconf? they seem to have a pretty different result here
23:54
(as in, autoreconf fails, autogen.sh doesn't)
23:54
<vagrantc>
oh, right...
23:55
i did that with something else, and don't think i'm getting it right here...
23:56
<stgraber>
ok, I think I got autoreconf to work but it takes some convincing
23:56
hmm, actually, not it doesn't work
23:56
config.status: error: cannot find input file: `po/Makefile.in.in'
23:56
I definitely don't get this one when running autogen.sh
23:59
<vagrantc>
i'm having dh_autoreconf call autogen.sh ... should work.
23:59
that did it!
23:59
doh! i had that problem with another build.