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


Channel log from 24 December 2015   (all times are UTC)

00:40Phantomas has left IRC (Phantomas!~ftsamis@ubuntu/member/phantomas)
00:46TatankaT has left IRC (TatankaT!~tim@heracles.oma.be, Ping timeout: 246 seconds)
01:00Ark74 has left IRC (Ark74!~Luis@177.239.209.188, Quit: Saliendo)
04:55gbaman has left IRC (gbaman!~gbaman@host81-139-185-126.in-addr.btopenworld.com, Remote host closed the connection)
04:56gbaman has joined IRC (gbaman!~gbaman@host81-139-185-126.in-addr.btopenworld.com)
05:04Fenuks has joined IRC (Fenuks!~Fenuks@gate.ifmieo.nspu.ru)
06:11vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
07:34alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Ping timeout: 256 seconds)
07:35uXus has left IRC (uXus!~uXus@217.77.222.72, Remote host closed the connection)
07:42ricotz has joined IRC (ricotz!~ricotz@p5B2AA92D.dip0.t-ipconnect.de)
07:42ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz)
07:47uXus has joined IRC (uXus!~uXus@217.77.222.72)
08:43alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
09:22Phantomas has joined IRC (Phantomas!~ftsamis@ubuntu/member/phantomas)
09:29Phantomas has left IRC (Phantomas!~ftsamis@ubuntu/member/phantomas)
09:29Phantomas has joined IRC (Phantomas!~ftsamis@ubuntu/member/phantomas)
11:24alkisg is now known as work_alkisg
11:38TatankaT has joined IRC (TatankaT!~tim@heracles.oma.be)
11:44gvy has joined IRC (gvy!~mike@altlinux/developer/mike)
11:56Fenuks has left IRC (Fenuks!~Fenuks@gate.ifmieo.nspu.ru, Ping timeout: 246 seconds)
13:07robb_nl has joined IRC (robb_nl!~robb_nl@ip-80-236-242-182.dsl.scarlet.be)
13:26TatankaT has left IRC (TatankaT!~tim@heracles.oma.be, Ping timeout: 250 seconds)
13:28TatankaT has joined IRC (TatankaT!~tim@heracles.oma.be)
13:28Gremble has joined IRC (Gremble!~Ben@host-92-27-135-217.static.as13285.net)
13:28Gremble is now known as Guest49460
14:34gbaman has left IRC (gbaman!~gbaman@host81-139-185-126.in-addr.btopenworld.com, Remote host closed the connection)
14:34gbaman has joined IRC (gbaman!~gbaman@host81-139-185-126.in-addr.btopenworld.com)
14:37gvy has left IRC (gvy!~mike@altlinux/developer/mike, Ping timeout: 272 seconds)
14:51gvy has joined IRC (gvy!~mike@altlinux/developer/mike)
15:02bitchecker_ has left IRC (bitchecker_!~bitchecke@31.131.20.132, Remote host closed the connection)
15:03bitchecker has joined IRC (bitchecker!~bitchecke@31.131.20.132)
15:19bennabiy has left IRC (bennabiy!~bennabiy@unaffiliated/bennabiy, Remote host closed the connection)
15:55gvy has left IRC (gvy!~mike@altlinux/developer/mike, Ping timeout: 276 seconds)
16:05Guest49460 has left IRC (Guest49460!~Ben@host-92-27-135-217.static.as13285.net, Remote host closed the connection)
16:15gvy has joined IRC (gvy!~mike@altlinux/developer/mike)
16:25muppis has left IRC (muppis!muppis@palo.kirves.fi, Ping timeout: 250 seconds)
16:31ogra_` has left IRC (ogra_`!~ogra_@p5098ed03.dip0.t-ipconnect.de, Ping timeout: 255 seconds)
16:31ogra_ has joined IRC (ogra_!~ogra_@80.152.237.3)
17:07gbaman_ has joined IRC (gbaman_!~gbaman@host81-139-185-126.in-addr.btopenworld.com)
17:08gbaman has left IRC (gbaman!~gbaman@host81-139-185-126.in-addr.btopenworld.com, Read error: Connection reset by peer)
17:26bennabiy has joined IRC (bennabiy!~bennabiy@unaffiliated/bennabiy)
18:04vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
18:04* bennabiy waves.
18:04
<bennabiy>
Howdy all
18:04* vagrantc waves
18:05
<bennabiy>
I love when I accidentally wipe out a filesystem because of improperly mounted roots
18:05* bennabiy sighs
18:07
<bennabiy>
vagrantc: which do you prefer for a hypervisor, kvm or vmware ESXi (for bare metal VM server)
18:09
<vagrantc>
bennabiy: i've only ever used free/libre software variants, so kvm.
18:09
typically with libvirt these days, though i've thought about exploring other management interfaces
18:09
<bennabiy>
vagrantc: thought so.
18:09
<vagrantc>
i used to use xen before i had access to hardware that supports /dev/kvm
18:10
<bennabiy>
I am looking for something which plays well with graphical interfaces as well as virtual servers
18:10
<vagrantc>
long, long ago, i used virtualbox, but that required out-of-tree kernel modules which left a bad taste in my mouth.
18:10
<bennabiy>
I know KVM can handle a GUI, but it is not strong when it comes to desktop environments
18:10* vagrantc nods
18:11
<bennabiy>
yes, and now kvm and virtualbox compete on the kernel level for ability to access the hardware
18:11
I used to be able to run kvm and virtualbox machines at the same time, and all was well
18:12
<vagrantc>
ironically, it's because virtualbox moved to a more standardized interface, probably :)
18:12
<bennabiy>
heh
18:13
Did you all have a BTS this year?
18:16
<vagrantc>
i think some folks got together ... couldn't make it myself
18:20
<bennabiy>
Did you get affected by the storms over there?
18:21
floods and all
18:21* bennabiy just got back from Idaho
18:21epoptes_user4 has left IRC (epoptes_user4!57e4c2d2@gateway/web/freenode/ip.87.228.194.210, Ping timeout: 252 seconds)
18:22
<vagrantc>
bennabiy: not to far from here was some pretty ugly flooding, but fortunately i'm a bit further uphill
18:22
<bennabiy>
That is good
18:23
We are starting a little community about an hour or two north of you
18:23
Raymond, WA
18:23
<vagrantc>
should make it easier to meet in person someday, then. :)
18:24
<bennabiy>
Yes, although I am not sure when I will be back out that way
18:24
within the next month or so I am going back to NC
18:24
But I should have a little more time on hand to work on projects
18:25
<vagrantc>
is LTSP amoungst them? :)
18:25
<bennabiy>
Considering it is one of the main things we use for our computer labs, I would hope so
18:26* vagrantc needs to put in some solid LTSP time, but has other pots on the fire right now
18:26* bennabiy nods
18:26
<bennabiy>
I can relate
18:26
Why was it, when we were younger we seemed to have more time in a day?
18:27
Do things just take longer these days?
18:27
<vagrantc>
i just don't tend to do 12-14 hour hack sessions anymore...
18:28
down to more like 2-8 ...
18:28
at some point a made a commitment to get more sleep :)
18:29
<bennabiy>
I understand that one
18:30
<vagrantc>
what has been stealing my time is putting together a diverse arm based build network ... really need to write up a post about that
18:30
<bennabiy>
What did you think about alkisg's suggestion about the bounty system for LTSP?
18:31
really! That would be an interesting read
18:31
I tend to keep ARM at arms length
18:31
<vagrantc>
it's really not mature enough yet for your typical plug-it-in and use it user yet... but getting tantalizingly close!
18:32
<bennabiy>
We use Pii2s for our phone systems
18:32
I am still trying to find the best thin clients for my labs
18:33
<vagrantc>
bennabiy: i like the idea of the bounty system for the most part ... LTSP needs something to give development a jump-start!
18:33
any time you bring funding into an otherwise volunteer project, it can get tricky, though
18:33
<bennabiy>
My issue is that most of the people in our communities are totally computer illiterate and I need to be able to have something reliable (hence my time to devote towards making sure it "Just Works"
18:33
yes
18:33
I wondered about that
18:36
<vagrantc>
hrm.
18:36
<bennabiy>
I liked the thought of being able to compensate for work done, like to get it to work in a situation for a company / school / group , including adding features, but it almost feels like it would lose an aspect of its freeness
18:36
<vagrantc>
seems like ltsp.org and wiki.ltsp.org are down
18:36
<bennabiy>
who hosts them?
18:37gvy has left IRC (gvy!~mike@altlinux/developer/mike, Quit: Leaving)
18:39* vagrantc doesn't remember
18:40
<bennabiy>
be back in a bit
18:41
lunch time
18:41
<vagrantc>
work_alkisg: you remember who hosts ltsp.org ?
18:41
Hyperbyte: ^^ ??
18:42work_alkisg is now known as alkisg
18:42
<alkisg>
Hi guys
18:42
jammcq had a friend in some web hosting business afaik
18:42
I don't think I ever heard his name or his company name
18:42
<vagrantc>
sbalneav: did you end up releasing a new libpam-sshauth package?
18:43
<alkisg>
I'll restart apache at ltsp.org
18:43
<vagrantc>
sbalneav: and how did the lightdm-webkit greeter go?
18:43
<alkisg>
$ uptime
18:43
13:43:14 up 139 days, 9:35, 1 user, load average: 0.05, 0.01, 0.00
18:43
...and maybe reboot the server as well :)
18:44
<bennabiy>
alkisg: might not hurt
18:44
<vagrantc>
clearly not getting security updates
18:44
well, not rebooting to apply them...
18:45
<alkisg>
Lucid has been eol for some time afaik anyway
18:46
<vagrantc>
eeyk.
18:50
<alkisg>
(08:36:23 μμ) bennabiy: ...but it almost feels like it would lose an aspect of its freeness
18:50
==> I don't think there's any serious open source project that doesn't have paid developers
18:50
<vagrantc>
ltsp had several people paid to work on it for quite some years
18:51
<alkisg>
And it'd be very nice to again have paid developers. Open source isn't about starved devs giving away their time for free.
18:51
<vagrantc>
and some businesses that used it as a core of their business actively contributing ...
18:52* alkisg crosses fingers and attempts an ltsp.org reboot, after the apt-get update...
18:52
<vagrantc>
that said, there is a bit of a difference with the project actively soliciting direct donations ... but i think it's something worth trying as long as we pay attention
18:55ltsp has joined IRC (ltsp!bot@ltsp.org)
18:55
<alkisg>
The amount was 1/10 of what paid developers would get for that programming time
18:55
<vagrantc>
https://mako.cc/writing/funding_volunteers/funding_volunteers.html
18:57
<alkisg>
"In short, when it comes to voluntary work and paid labor, you can have one or the other but not both."
18:57
I completely don't agree with that
18:58
<vagrantc>
that's one line in an entire paper
18:58
<alkisg>
I've spend 20+ years of programming projects on a voluntary basis, and then getting some compensation for them, to be able to care for my family etc
18:58* quinox once got bitcoins for a contribution... didn't bother to claim them
18:59
<alkisg>
E.g. I give up some extra jobs, and invest the time on an open source project; the users were usually very willing to compensate for the income that I lost in order to invest on that open source project
18:59
At that's been true for me for 20 years, I don't know how generic that can be
19:00
I think the key is why the developers are doing the work
19:00
<vagrantc>
and i've spent nearly as long doing largely volunteer work on free software projects
19:00
<alkisg>
If they're doing it for the money in the first place, then some failures will happen
19:01
<vagrantc>
though for large stretches of that time it was related to the work i was paid (poorly) for
19:01
<alkisg>
If they're doing it because the want to contribute something to others, then they will succeed as long as they have enough income to live like that
19:04
<vagrantc>
alkisg: i don't disagree with you, but i also think it's foolish to ignore that it may have some negative consequences.
19:04
<alkisg>
Like what? Driving volunteers out of ltsp? :)
19:05
<vagrantc>
it's possible ... as few as they have been lately, less would be worse.
19:06
<alkisg>
It's also possible that volunteers might come if they see bounties though. And no, not the bad ones that only care about the money.
19:06
But to be honest, I don't expect anything else to happen, only 1 person answered in the mailing list
19:07
<quinox>
I suppose it can be a nice show of gratitude
19:07
<alkisg>
Gratitude = donations, paid work = bounties
19:07
Or paid support
19:08
<quinox>
gratitude for the random developer dropping by to fix something
19:08
<vagrantc>
alkisg: yeah... arguably some controversy over it might be healthier that the so-far non-response we've seen
19:08
<alkisg>
quinox: if you see the ltsp-trunk commits, we've been fixing things for ages; I don't think anyone has actually donated something yet
19:09
vagrantc: one other way to attract developers would be to leave something broken; e.g. if people can't login with systemd, they'll other drop ltsp or someone will stand up and fix it :)
19:09
*either
19:09* vagrantc suspects the former
19:10
<alkisg>
Maybe it's time for something new then
19:10
<vagrantc>
how long was tftp booting broken in Ubuntu's LTS release?
19:11
<alkisg>
I think LTSP in Ubuntu has been broken since 12.10, at least without the greek schools ppa or without a lot of workarounds
19:11
<vagrantc>
with bugs reported pretty much since the day of release
19:11
or even before
19:11
<quinox>
It worked for me with vanilla Ubuntu up to 14.10
19:12
<alkisg>
What desktop environment?
19:12
<quinox>
fat clients with awesome, KDE and Gnome
19:12
<vagrantc>
there was a long stretch where you had to manually copy pxelinux.0 and other such files
19:13
<alkisg>
And without XDG_CURRENT_DESKTOP I doubt that Gnome and KDE would load properly their menus
19:13
Or without XDG_SEAT etc
19:13
Awesome might not need the XDG* variables, I haven't tried it
19:15
<quinox>
awesome isn't a desktop env anyway, it doesn't do anything except window managing
19:15* alkisg notes that ltsp.org is very fast now after the reboot :)
19:16
<vagrantc>
!alkisg
19:16
<ltsp>
alkisg: The LTSP oracle. Our beacon of hope in the world of LTSP. With the guidance of this divine emperor, we shall prevail.
19:16
<quinox>
the pxelinux.0 thing started with 14.10 --> https://bugs.launchpad.net/ltsp/+bug/1387773
19:16
<vagrantc>
and the bot works.
19:16
<alkisg>
!vagrantc
19:16
<ltsp>
Error: "vagrantc" is not a valid command.
19:16
<alkisg>
Heh, how did you get away without a motto?!
19:16
!v
19:16
<ltsp>
v: vagrantc!!!!!!!!!!!!!!!!!!1!!!!!
19:17
<bennabiy>
hah!
19:17
<alkisg>
mmmm it needs rewriting :)
19:17
<vagrantc>
while enthusiastic, it doesn't say much. :)
19:17
<quinox>
tbh, if a project isn't on github I don't even bother sending in patches most of the time
19:18* vagrantc would definitely like to switch to git, even if isn't thrilled about github
19:18
<quinox>
I like github's model, personal fork to hack around and if the changes end up being decent file a PR with 1 press of a button
19:18
<alkisg>
github doesn't support translations though, nor build recipes
19:18
<quinox>
the auto integration tools are also very good
19:18
travis etc.
19:19
<alkisg>
What is travis?
19:19
<quinox>
https://travis-ci.org/
19:19
free for open source projects
19:19
<alkisg>
Unit testing?
19:19
<quinox>
for example, yes
19:19
code quality as well
19:20
<vagrantc>
alkisg: i guess we could continue to use launchpad and then git-remote-bzr to push/pull updates near releases
19:20
<alkisg>
LTSP mostly shouldn't be about code, but about making sure other projects don't break netbooting
19:20
<vagrantc>
but likely to always be a release behind then
19:20
alkisg: true enough!
19:21
<alkisg>
vagrantc: we can host the code in github, I don't mind at all, and we can use launchpad for translations, build recipes etc
19:21
<quinox>
are they automated tests for it?
19:21* bennabiy nods
19:21
<vagrantc>
unit-testing booting a thin client into dozens of different desktop environments is an interesting challenge
19:21
<bennabiy>
github might make it a little more friendly for people to contribute
19:21
<alkisg>
quinox: how useful would it be for you if a test told you "the client didn't boot"?
19:22
<quinox>
it would help new developers - I'm always much more confident when I file a PR when I see unittests passing
19:22
yeah LTSP is a bit tricky
19:22
it's better than no feedback :)
19:22
<alkisg>
E.g. the clients can't shut down with systemd. It takes a week to locate and send a patch to systemd for that. I don't think unit testing would help ltsp there. It would help systemd, sure, but not ltsp.
19:22
<vagrantc>
there probably are *some* unit-tests we could write for some of the code, but that's not likely to be the biggest source of bugs.
19:24
<quinox>
so skip the unittest and go straigh to system tests... somehow
19:24
<alkisg>
The problem is not testing, it's fixing
19:24
We know of a lot of bugs that don't get fixed
19:24
<bennabiy>
LTSP = lots'o tiny strungtogether programs
19:24
<quinox>
:)
19:25
<alkisg>
E.g. we want to get rid of LDM, but we haven't had the time/manpower to do it for years now
19:25
<bennabiy>
ok, so some not so tiny
19:25
<vagrantc>
at least 3+ years ...
19:26* alkisg also doesn't have much motivation to do it since LightDM breaks the keyboard layout, and switching to lightdm would mean to have to spend a month in getting the bugs fixed + upstreaming them
19:26
<vagrantc>
any other display managers any better?
19:27
we could probably become (sbalneav already is) upstream for the webkit frontend for lightdm and fix it there, maybe?
19:27
<alkisg>
gdm 2 was working fine, gdm 3 had the same issue
19:28
<vagrantc>
alkisg: and LDM actually works? :)
19:28
<alkisg>
Yes because it doesn't try to be smart; it respects that xorg knows about keyboard layouts
19:29
<vagrantc>
right.
19:29
<alkisg>
With the wayland movement, gnome is trying to implement the keyboard layout handling in a new way, and it doesn't do a good job at it
19:29
I don't know if the frontend or the backend is responsible for the layout. There's also accountservice there involved, there were a couple of related bugs there as well
19:31
<quinox>
(random rant: with ubuntu 15.10 pulseaudio decided to turn on flat-volumes ... what a horrible choice that was)
19:32
it scales the audio output of all your programs depending on other programs, drove me nuts in the first 3 minutes
19:33
anyway, I'm not saying LTSP has to move the Github, but as a developer I like the github experience so much that bitbucket/mailinglists/launchpad feels like a drag and un-fun to participate in
19:38* vagrantc suspects a switch to github would be a good move
19:40
<quinox>
should LTSP get a bounty pool ping me and I'll chip in a couple hundred $ for seeding
19:41
<vagrantc>
there's http://wiki.ltsp.org/wiki/Bounties
19:41
but there's no real actual proposals yet
19:41
well, the proposed bugfix is real :)
19:41
<quinox>
I have no experience with when from either perspective, but I know Weblate (a pretty awesome project) and Neovim use https://salt.bountysource.com/
19:42
<vagrantc>
we have no system set up to actually collect or distribute funds
19:43
<quinox>
wow, neovim has 400 monthly supporters
19:43
<vagrantc>
heh. i supporter for every two teams
19:44
<alkisg>
I've seen that tightvnc uses bountysource
19:44
It got about $200 in total for all those years :)
19:44
*tigervnc
19:45
<quinox>
Weblate uses it like this: https://github.com/nijel/weblate/issues/918 a link is automatically added to fund a ticket
19:45
<alkisg>
https://bugs.launchpad.net/ltsp/ and click on the bounty tag
19:45
I.e. https://bugs.launchpad.net/ltsp/+bugs?field.tag=bounty
19:46
I only did that as a proof of concept of course, since only one person replied in the ML, so I don't think bounties will take off
19:47
Maybe it's time for ltsp to be replaced by some other project
19:47
<quinox>
it's a tag with no info right?
19:48
<alkisg>
All the bounty information (except the monetary bits) should be in the bug report
19:48
I.e. what it is about, how to reproduce, who is it assigned to, if it needs to be SRUed etc
19:51
<bennabiy>
alkisg: suggestions?
19:51
<alkisg>
bennabiy: about what, bounties?
19:52
<bennabiy>
alkisg: <alkisg> Maybe it's time for ltsp to be replaced by some other project
19:53
<alkisg>
I was thinking for a new "configd" project, with a lot of new ideas that make even init-ltsp.d obsolete
19:53
*of
19:54
If I have enough time for it some day, I might put it to kickstarter
19:55
<bennabiy>
LTSP meets about 85% of my users needs
19:55
<alkisg>
I was thinking that it would be generic enough to allow for netbooted installations, normal installations (standalone), or even cached installations
19:56
<bennabiy>
that would be nice
19:56
documentation!
19:56
<alkisg>
LTSP breaks too often and it takes a lot of time to get it fixed
19:56
<bennabiy>
bounties on documentation!
19:56
<vagrantc>
bounties on documentation is interesting
19:57* bennabiy imagines a day when the LTSP wiki is full of relevant up-to-date information
19:58
<vagrantc>
it would create an incentive for people to actually ferret out the features from the developers, or for the developers to write it themselves in the first place
19:58
<bennabiy>
yes
19:58* vagrantc finds wikis are really hard to curate
19:59* bennabiy makes note to get all the information out of the developers before the bounties kick in...
19:59
<vagrantc>
they may start out with good docs, but random people with varying degrees of understanding will make wild edits of dubious quality in no time
19:59
<bennabiy>
yes
19:59
<vagrantc>
it really requires constant villigence
20:00
<bennabiy>
and LTSP has not had the best track record of staying up to date...
20:00
<vagrantc>
it's great for bootstrapping docs, but not for long-term maintenance
20:00
<alkisg>
In my last 2-3 pages in the ubuntu wiki, I ended my page with "please don't edit it, I'll be maintaing the current page myself"
20:00
<bennabiy>
alkisg: I noticed that
20:00
<alkisg>
I think it's much better if there's an appointed maintainer for each page
20:00
Something like debian, with appointed maintainers for each project
20:01
<bennabiy>
I have to be honest, before I got to know you a bit, I thought it seemed a bit proud or such, but I do respect it and appreciate it
20:01
alkisg: I agree
20:02
<alkisg>
I did change the pages a number of times when people reported issues or misunderstandings, i.e. I do appreciate feedback
20:02
But my previous pages, I stopped editing them when 10s of people added inaccurate information
20:02
I had to rewrite them at that point, and I couldn't take the time to do that every few months
20:02
<vagrantc>
chmod -R 777 /
20:03
<quinox>
A++ works for me
20:03
<vagrantc>
oops, that was supposed to go in my next wiki edit
20:03
<bennabiy>
It would be good if there could be a section of a wiki where potential entries could go for review, and then if approved, could be posted to the appropriate place
20:04
<vagrantc>
quinox: numbers are easier for computers to understand that letters and punctuation.
20:04
<quinox>
we've got this one developer that won't be receiving sudo powers for at least a few more years.. he googles around and runs the first command he finds without even looking at it
20:04
<vagrantc>
gatekeepers have their own issues, though ...
20:04
<quinox>
I like a=rwx myself
20:04
<vagrantc>
yeah, but that takes more processing power.
20:05
<bennabiy>
vagrantc: like partiality?
20:05
<vagrantc>
bennabiy: partiality, insufficient bandwidth to deal with legitimate updates
20:05
<bennabiy>
good point
20:06
<vagrantc>
if there aren't enough gatekeepers, one going missing for a while can stall things out ...
20:06
but you can hardly stall the docs more than they are now...
20:06
<bennabiy>
again, good point
20:06
<vagrantc>
(and spam defense can be a massive demotivator
20:07
)
20:07
<bennabiy>
I don't even remember if I have access to edit the docs
20:07
<vagrantc>
though, if the docs were on github in a git repository, pull requests would be much more manageable
20:07
<bennabiy>
yes
20:08
what would be the method of integrating into PPAs and such to deal with the lag between releases?
20:08
pull from github into bzr/launchpad?
20:08
<vagrantc>
i seem to recall alkisg suggesting that you can do PPAs with external repositories
20:08
<bennabiy>
oh really?
20:08
<vagrantc>
but yeah, worst case would be git -> bzr syncing
20:09
<bennabiy>
I know you can import external packages, but...
20:09
hmm, fork launchpad and make it sit over github instead?
20:11
I do think if we can do something to get the documentation up to date, it will go a long way towards LTSP progress
20:11
I wonder how many people pull up the old LTSP manual, try it out, run into difficulties and then quit without ever coming here
20:11
<alkisg>
bennabiy: wiki.ltsp.org is world-writeable
20:11
And noone writes to it
20:11
Why would they write if we used git instead of a wiki?
20:12
<bennabiy>
does sbalneav have the files for the manual still? I might be able to touch it up if I had access
20:12
<alkisg>
!ltsp-docs
20:12
<ltsp>
Error: "ltsp-docs" is not a valid command.
20:14
<alkisg>
...vagrantc, who is this v6sa that contributes more than you in ltsp? https://launchpad.net/ltsp/ :D
20:14
<bennabiy>
hrm, I actually remembered my login for the wiki
20:14
translations?
20:14* alkisg searches for ltsp-docs...
20:15
<alkisg>
https://code.launchpad.net/ltsp
20:16
!learn ltsp-docs as https://code.launchpad.net/~ltsp-docwriters/ltsp/ltsp-docs-trunk
20:16
<ltsp>
The operation succeeded.
20:17
<alkisg>
PPAs can pull from external sources, yup, even git
20:20
<Hyperbyte>
alkisg, merry christmas!
20:20
<alkisg>
Merry Christmas Hyperbyte!
20:21
!Hyperbyte
20:21
<ltsp>
Hyperbyte: The LTSP manifest clearly states that LTSP developers can only code under the ecstatic music of our almighty D.J. LTSPROCKS, but HyperByte rocks better! (his web design skills are OK too, check out ltsp.org :))
20:25
<bennabiy>
!bennabiy
20:25
<ltsp>
bennabiy: "wait a minute... who put that there?!"
20:25
<bennabiy>
hrm
20:31
<quinox>
Does the wiki contain something like https://ghost.qtea.nl/tmp/ltsp_boot_flow.png ?
20:32
I've put this together over the years, but if I can simply link to the wiki with the same info that would be nice
20:32
(a deviation from the standard is that we don't use the LTSP server for DHCP)
20:34
<bennabiy>
Does DisklessWorkstations still deal in clients? Their page seems a bit outdated
20:34
<Hyperbyte>
alkisg, wtf....
20:34
lol
20:40bennabiy has left IRC (bennabiy!~bennabiy@unaffiliated/bennabiy, Remote host closed the connection)
20:47bennabiy has joined IRC (bennabiy!~bennabiy@unaffiliated/bennabiy)
20:52
<alkisg>
quinox: there are a few things not accurate in that .png
20:53
The dnsmasq + tftpd-hpa is a bit rare combination, usually it's either plain dnsmasq, or isc-dhcp + tftpd-hpa
20:53
The NBD port now is 10809, name-based instead of port-based configuration for exports
20:53
Those are configured in /etc/nbd-server, not in inetd
20:54
The client login, start x etc are not configured in initramfs-tools, initramfs ends at about stage 6
20:56
<quinox>
thanks, I'll update my info
20:56
the dnsmasq runs on our linux gateway - was there before we started with LTSP
20:57* alkisg waves, 'night guys...
20:57alkisg is now known as work_alkisg
20:57* bennabiy waves
20:57
<bennabiy>
Night!
21:27
<quinox>
my point was that I'd like this kind of flow documented on the wiki
21:27* quinox goes to bed as well
22:10
<quinox>
mmm since LTSP can auto-login clients based on MAC it should actually be no problem to boot a client straight into KDE/Gnome/Awesome and report success back
22:10
if you can set up a temp network and semi-fake boot the client
22:14
QEMU can boot from tftp apparently
22:15
you can point it at a tftp folder and boot, no daemon required
22:17
and there's Xephyr / Xvfb
22:18* quinox might play around with it tomorr
22:18
<quinox>
w
22:19
(booting from a folder isn't so nice since you need the ssh server anyway)
22:21ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving)
22:58
<vagrantc>
using qemu's tftp emulation doesn't really test the tftp configuration scripts ltsp provides very well
23:05robb_nl has left IRC (robb_nl!~robb_nl@ip-80-236-242-182.dsl.scarlet.be, Ping timeout: 245 seconds)