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


Channel log from 16 May 2017   (all times are UTC)

02:36lucascastro has joined IRC (lucascastro!~lucas@186.227.186.18)
05:14gehidore has left IRC (gehidore!~username@unaffiliated/man, Quit: WeeChat 1.7)
05:15gehidore has joined IRC (gehidore!~username@unaffiliated/man)
05:20gehidore has left IRC (gehidore!~username@unaffiliated/man, Client Quit)
05:20gehidore has joined IRC (gehidore!~username@unaffiliated/man)
05:20lucascastro has left IRC (lucascastro!~lucas@186.227.186.18, Remote host closed the connection)
05:22lucascastro has joined IRC (lucascastro!~lucas@186.227.186.18)
05:38gehidore has left IRC (gehidore!~username@unaffiliated/man, Quit: WeeChat 1.8)
05:40gehidore has joined IRC (gehidore!~username@unaffiliated/man)
05:48lucascastro has left IRC (lucascastro!~lucas@186.227.186.18, Remote host closed the connection)
06:00johnsmith has joined IRC (johnsmith!b4974832@gateway/web/freenode/ip.180.151.72.50)
06:20Statler has joined IRC (Statler!~Georg@p4FC1FE8A.dip0.t-ipconnect.de)
06:20mikkel has joined IRC (mikkel!~mikkel@mail.dlvs.dk)
06:27ricotz has joined IRC (ricotz!~ricotz@p5B2A962A.dip0.t-ipconnect.de)
06:27ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz)
07:12trendzetter2 has left IRC (trendzetter2!~trendzett@mail.sbat.be, Ping timeout: 245 seconds)
07:41
<bcg>
If you're switching to git soon then I'll just forget this bzr fighting. It still doesn't work for me, does nothing.
07:42
<alkisg>
The main problem is that launchpad doesn't yet support syncing translations with git branches
07:43
We may do it with an automatically synced bzr branch though... it has support for git=>bzr cloning
08:19Statler has left IRC (Statler!~Georg@p4FC1FE8A.dip0.t-ipconnect.de, Remote host closed the connection)
08:22wim1 has joined IRC (wim1!~Thunderbi@WEGC33.UNI-GRAZ.AT)
08:40adrianor1 has joined IRC (adrianor1!~adrianorg@179.177.209.49.dynamic.adsl.gvt.net.br)
08:44adrianorg has left IRC (adrianorg!~adrianorg@187.58.153.23, Ping timeout: 272 seconds)
08:46Statler has joined IRC (Statler!~Georg@mail.lohn24.de)
08:49ogra_ has left IRC (ogra_!~ogra_@ubuntu/member/ogra, Ping timeout: 240 seconds)
08:49ogra_ has joined IRC (ogra_!~ogra_@p5098ed03.dip0.t-ipconnect.de)
09:15gvy has joined IRC (gvy!~mike@altlinux/developer/mike)
10:02Trendzetter has joined IRC (Trendzetter!~trendzett@mail.sbat.be)
10:03
<Trendzetter>
hi everyone! I am using LTSP to take exams
10:03
right now I am looking into epoptes
10:03
I have set up a debian desktop for epoptes server
10:04markus_e92 has left IRC (markus_e92!~markus_e9@91-115-155-93.adsl.highway.telekom.at, Ping timeout: 260 seconds)
10:05
<Trendzetter>
when I run epoptes it I get an error 13
10:05
<alkisg>
Trendzetter: use copy/paste to show the exact errors
10:07markus_e92 has joined IRC (markus_e92!~markus_e9@62-46-26-118.adsl.highway.telekom.at)
10:09
<Trendzetter>
https://pastebin.com/raw/ZBsVwb5V
10:11
<alkisg>
Trendzetter: that means that you are not part of the epoptes group, or that you just got into that group but haven't logged out yet
10:11
See epoptes.org/installation for instructions
10:11
<Trendzetter>
I added to the epoptes group but indeed I didn't log out afterwards..
10:12
<alkisg>
You can use `newgrp epoptes`
10:12
To avoid logout
10:12
newgrp epoptes <enter>
10:12
epoptes <enter>
10:13
<Trendzetter>
nice! now booting some clients and see if I get the connections established! thanks
10:14
<alkisg>
np
11:57kjackal_ has joined IRC (kjackal_!~quassel@2a02:587:310c:4d00:b425:1f06:53fc:28c5)
12:40SYS64738_ has joined IRC (SYS64738_!~Amuro_Ray@159.213.93.166)
12:43SYS64738 has left IRC (SYS64738!~Amuro_Ray@159.213.93.166, Ping timeout: 240 seconds)
12:47SYS64738_ has left IRC (SYS64738_!~Amuro_Ray@159.213.93.166, Quit: leaving)
12:47SYS64738 has joined IRC (SYS64738!~jhonny5@159.213.93.166)
12:49wim1 has left IRC (wim1!~Thunderbi@WEGC33.UNI-GRAZ.AT, Quit: wim1)
12:52sbalneav has left IRC (sbalneav!~sbalneav@wnpgmb0311w-ds01-165-236.dynamic.mtsallstream.net, Ping timeout: 272 seconds)
12:57
<Trendzetter>
I am not sure if I understand the ltsp-pnp epoptes architecture correctly. On the ltsp-pnp in set the hostname "server" to resolv it to 192.168.103.97 (my epoptes desktop server). When I boot a client terminal and I check to what "server" resolvs it is the ip of the ltsp-pnp server (192.168.130.248).
12:58
If I am using a different epoptes server than ltsp-pnp server than I should use a different name for the epoptes server than "server"
12:58
?
12:59
I will give it a try if your buzy with other stuff
12:59
I'll let you know how it progresses
13:18
I am still stuck. No clients are being detected
13:23
I tried configuring the client sessions by logging on to the ltsp-pnp in the client profile and running "sudo epoptes-client -c 192.168.103.97". The certificate was fetched without issues.
13:23
then i rebooted the client
13:25
but the client session is not detected. using netstat I could not see a connection to the epoptes server(192.168.103.97)
13:30
<alkisg>
Trendzetter: run this on the clients: ping server
13:30
As you see, it resolves to the ltsp server
13:31
If you want another server, configure /etc/default/epoptes-client accordingly, so that ping works
13:31
<Trendzetter>
I tried "ping server". It pings the ltps-pnp server instead of the epoptes desktop
13:31
<alkisg>
Check epoptes.org/installation for instructions
13:31
<Trendzetter>
so I configured the client using fixed ip 192
13:31
ip 192.168.103.97
13:31
<alkisg>
The epoptes server is an ltsp fat client?
13:32
Or is it a standalone computer that is not part of ltsp?
13:32
<Trendzetter>
I am using your tutorial pnp, thats not fat client
13:33
<alkisg>
!ltsp-pnp
13:33
<ltsp>
ltsp-pnp: ltsp-pnp is an alternative (upstream) method to maintain LTSP installations for thin and fat clients that doesn't involve chroots: https://help.ubuntu.com/community/UbuntuLTSP/ltsp-pnp
13:33
<alkisg>
ltsp-pnp is fat clients, yes
13:33
<Trendzetter>
the epoptes server is a debian virtual machine on esx
13:33
<alkisg>
OK
13:33
So, you're supposed to install epoptes-server there, and epoptes client in ltsp
13:33
Not epoptes in ltsp at all
13:35
So, in ltsp, apt purge epoptes
13:35
<Trendzetter>
ok so I make sure that no epoptes server is installed on the ltsp server and check the contents of /etc/default/epoptes-client
13:35
<alkisg>
Open /etc/default/epoptes-client and set SERVER=debian-vm (or whatever the hostname is)
13:35
Or put the ip directly there
13:35
SERVER=192.168.103.97
13:36
Then run `sudo epoptes-client -c` to get the server certificate
13:36
Then ltsp-update-image -c / to create the updated ltsp image
13:36
Reboot clients, ready
13:36
<Trendzetter>
ok. checking that out now
13:39wim1 has joined IRC (wim1!~Thunderbi@WEGC33.UNI-GRAZ.AT)
13:52johnsmith has left IRC (johnsmith!b4974832@gateway/web/freenode/ip.180.151.72.50, Ping timeout: 260 seconds)
13:57
<Trendzetter>
just wanted to confirm that the clients are now indeed visible from the epoptes desktop server
13:59ben_roose has joined IRC (ben_roose!~roose@roose.cs.wichita.edu)
14:01
<alkisg>
Cool
14:11mikkel has left IRC (mikkel!~mikkel@mail.dlvs.dk, Quit: Leaving)
14:13vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
14:34GodFather has joined IRC (GodFather!~rcc@2600:1008:b050:3013:9825:bd2b:639f:c9ed)
14:35GodFather has left IRC (GodFather!~rcc@2600:1008:b050:3013:9825:bd2b:639f:c9ed, Client Quit)
14:48kjackal_ has left IRC (kjackal_!~quassel@2a02:587:310c:4d00:b425:1f06:53fc:28c5, Quit: No Ping reply in 180 seconds.)
14:49kjackal_ has joined IRC (kjackal_!~quassel@2a02:587:310c:4d00:a423:99ca:c80f:28cd)
14:58lucascastro has joined IRC (lucascastro!~lucas@186.227.186.18)
14:59wim1 has left IRC (wim1!~Thunderbi@WEGC33.UNI-GRAZ.AT, Quit: wim1)
15:00wim1 has joined IRC (wim1!~Thunderbi@WEGC203033.UNI-GRAZ.AT)
15:03wim1 has left IRC (wim1!~Thunderbi@WEGC203033.UNI-GRAZ.AT, Client Quit)
15:26vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
15:58forum has joined IRC (forum!~Thunderbi@193-154-16-227.adsl.highway.telekom.at)
16:05forum has left IRC (forum!~Thunderbi@193-154-16-227.adsl.highway.telekom.at, Ping timeout: 268 seconds)
16:32gvy has left IRC (gvy!~mike@altlinux/developer/mike, Quit: Leaving)
16:39lucascastro has left IRC (lucascastro!~lucas@186.227.186.18, Remote host closed the connection)
17:49kjackal_ has left IRC (kjackal_!~quassel@2a02:587:310c:4d00:a423:99ca:c80f:28cd, Quit: No Ping reply in 180 seconds.)
17:50vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
17:50kjackal_ has joined IRC (kjackal_!~quassel@2a02:587:310c:4d00:b425:1f06:53fc:28c5)
17:56JerryT has joined IRC (JerryT!~jerry@wsip-70-165-106-163.om.om.cox.net)
18:01
<vagrantc>
alkisg: attempting to push my git branch now
18:01
<alkisg>
vagrantc: cool
18:01
Yeah, I was proposing to use git on launchpad; but if you prefer github, we can do that too...
18:02
We'll have to do an extra step for the translations; I hope we'll manage
18:02
<vagrantc>
https://code.launchpad.net/~ltsp-upstream/+git/ltsp
18:02
<alkisg>
Maybe they'll need to be manually synced now, but that also saves us from the frequent lang commits issue
18:02
<vagrantc>
i prefer launchpad, honestly
18:03
<alkisg>
No objections from me
18:03
<vagrantc>
there's enough ambiguity with GPL code on github
18:03
<alkisg>
I may even revert epoptes to launchpad, to be there with ltsp and sch-scripts...
18:03* vagrantc would be fine with that
18:04
<vagrantc>
phantomas got excited and moved everything to git ... and then... ah life. :)
18:04
<alkisg>
So, once you put ltsp+ldm+ltspfs to git/launchpad, send a mail to the ltsp-developers list, so that everyone stops using the old trees
18:04
<vagrantc>
er, github
18:04
<alkisg>
Yeah he hopes he'll be able to work on epoptes again in the future, but let's see what comes up...
18:05
<vagrantc>
so the translations still need a bzr branch that we have to manually merge?
18:05
<alkisg>
https://code.launchpad.net/~ts.sch.gr/sch-scripts/trunk => The old, bazaar-based code repository of sch-scripts. Now it's moved to git: https://code.launchpad.net/sch-scripts
18:05
I made a note to the old bzr tree, we can do similar notes for ltsp
18:05
<vagrantc>
there's also some flag we need to switch to make git the default and bzr the "alternative" VCS on launchpad
18:06
<alkisg>
Yes otherwise build recipes fail :)
18:06
We don't want to mark bzr as the alternative, we want to mark that only git is used
18:06
<vagrantc>
there are a few outstanding issues with build recipes we never really sorted out ... :/
18:06
<alkisg>
The bzr branches will still be there, they won't be removed
18:07
I don't think launchpad supports pulling translations from git branches,
18:07
<vagrantc>
but they'll use a different URL, if i recall, so they will effectively be hidden
18:07
<alkisg>
so there are two ways to manage them,
18:07
<vagrantc>
alkisg: do you know if there's a launchpad feature request for that? if not, we should open it
18:07
<alkisg>
I think they're slowly working on that
18:08
...one is to have an automatic bzr mirror of our git tree, and tell launchpad to use that for translations, but since it's a mirrror (=read only), to push the results to another, third tree,
18:08
<vagrantc>
a little nudge on relevent bugs sometimes can get people to think about it again and make progress
18:08
alkisg: that sounds pretty good
18:09
<alkisg>
and the second way is to manually push the .pot to a second bzr branch, and have launchpad maintain translations there, using only the .pot/po files
18:09
<vagrantc>
i like the read-only branch idea
18:09
because then we can just merge the changes, rather than having to manually copy them into the branch
18:10
we could even do merge --squash, so we get one commit for all the translations
18:10
<alkisg>
Do we want to merge the translations back to the initial tree, or do we want to keep them in the separate tree?
18:10
<vagrantc>
i like branches that we can simply merge, myself
18:11
<alkisg>
We can have a do-release script that pulls from 2 branches and creates a single .tar.gz, if we want that
18:11* vagrantc would rather just merge them back into the master branch
18:11
<alkisg>
OK. That would be a manual process
18:11
Maybe we can do that before releases
18:11
<vagrantc>
but then git tags are meaningful
18:12
e.g. you could in theory recreate the tarball from git
18:12
and i can sign the git tags
18:13
hmmm... not sure i can push the tags for only upstream
18:13
i guess i could also push the debian branches to launchpad
18:14
<alkisg>
Yey!
18:14
<vagrantc>
not in the way you're excited about :P
18:15
<alkisg>
:D
18:15
It will still make synching easier
18:15
That alioth site is very frequently broken in several ways
18:16
E.g. mismatched certificate, wrong redirects, whatever
18:16
<vagrantc>
ouch
18:18
alkisg: can you see the tags if you: git clone https://git.launchpad.net/~ltsp-upstream/+git/ltsp ?
18:18
it's not clear to me if they are pushed or not
18:19
ah, they're visible from: https://git.launchpad.net/~ltsp-upstream/+git/ltsp/
18:19
but https://code.launchpad.net/~ltsp-upstream/+git/ltsp only shows a brief summary
18:20
<alkisg>
vagrantc: are we going to use tags anywhere else than for releases?
18:20
<vagrantc>
alkisg: i don't intend to
18:20kjackal_ has left IRC (kjackal_!~quassel@2a02:587:310c:4d00:b425:1f06:53fc:28c5, Ping timeout: 240 seconds)
18:20
<alkisg>
OK, I can understand that much then. Otherwise I'd need a tutorial :D
18:20
<vagrantc>
although there are a few oddball tags in there
18:21
<alkisg>
What's after ltsp 5.5.9? 5.5.10 or 5.6.1?
18:21
<vagrantc>
somehwere along the way this alkis person committed a "current" tag :P
18:21
<alkisg>
I committed a tag? Never!
18:21
<vagrantc>
alkisg: depends on how much we feel like changing it
18:21
<alkisg>
It will signify the transition to git :D
18:22
<vagrantc>
personally, i've been thinking we should call it ltsp 5.17 starting in january 2017
18:22
and 5.18 starting in january 2018
18:23
etc.
18:23
<alkisg>
I would prefer ltsp 17.04.1 :D
18:23
<vagrantc>
i wouldn't object
18:23
<alkisg>
I already name sch-scripts that way
18:24
<vagrantc>
i like the date being embedded in the release
18:24
i guess the zero-padding helps for sort order?
18:24
<alkisg>
I think so, it might also make users more easily understand that it's a date
18:25
<vagrantc>
with what we have now, people generally say "version 5-something
18:25
"
18:25
which... is... practically meaningless at this point
18:25
<alkisg>
Yeah I don't even remember in which 5-something was ltsp-pnp implemented
18:26
<vagrantc>
exactly
18:26
we really should have called it 6 when we moved to init-ltsp.d and implemented ltsp-pnp stuff
18:26
<alkisg>
True
18:26
I asked for that but I was turned down :D
18:26
<vagrantc>
though, to the end-user, there wasn't much difference
18:27
versioning based on feature differentness is actually kind of hard, which is part of why i like date based versioning
18:27
it's really simple to implement
18:27
for some reason i balk *slightly* at zero-padded versions, though
18:27
<alkisg>
Suppose stretch has ltsp 17.04
18:28
<vagrantc>
no, stretch froze in 16.11
18:28
<alkisg>
OK
18:28
<vagrantc>
i added some features since, but not much
18:28* vagrantc would like debian to version based on the date of freeze
18:28
<alkisg>
Some time later you backport it to jessie, but you cherry pick some commits. What do you call that, 16.11.1?
18:29
And, if you put those commits to stretch too? Or more commits? 16.11.2?
18:29
<vagrantc>
alkisg: backport would be 16.11~0bpo$RELEASEVERSION+$N
18:29
<alkisg>
What is $RELEASEVERSION?
18:29
<vagrantc>
debian jessie is 8, stretch is 9
18:29
<alkisg>
OK
18:30
<vagrantc>
there's a standard protocol for backport versioning in debian, security update versioning, point-release updates, etc.
18:31
<alkisg>
And if you want the equivalent of an "SRU" for stretch? I.e. cherry pick some more bug fixes, after stretch is out?
18:31
How is that called?
18:31
<vagrantc>
+deb9u$N
18:31
<alkisg>
Cool
18:31
<vagrantc>
if i'm remembering exactly correctly
18:31
<alkisg>
I like that
18:32
<vagrantc>
might be subtle variations, but that's the idea
18:32
maybe point release/SRU and security updates use the same versioning
18:33
<alkisg>
In PPAs, I might end up having two versions, like... 16.11.1~201703160820~xenial1
18:33
...but I don't think that's too bad either
18:34
<vagrantc>
let's think more about versioning once stretch releases :)
18:35
<alkisg>
Well I decided I won't bother with SRUs, so I don't need ltsp releases for ubuntu, so I don't mind waiting for the debian cycle to take its course
18:35
For ppas, the date is fine
18:35
<vagrantc>
so, ltsp upstream is converted over to git, as far as i can tell ... will think a bit more about the debian branch(es) later
18:35
<alkisg>
What about ldm, ltspfs, ltsp-docs?
18:35
<vagrantc>
yeah, was going to move on to ldm and ltspfs next
18:35
<alkisg>
Nice
18:36
<vagrantc>
ltsp-docs seems so out of date... it might be easier to start over
18:36
<alkisg>
Hmm, we'll probably want to disable translations until we get the new translations support ready
18:36
Otherwise they'll still be committed to the old branch
18:37
...and I hope we won't lose the new translations that will come until we get the translations support ready... hmm...
18:40
<vagrantc>
can we leave a branch as is, but mark it as read-only?
18:40
for launchpad bzr branches or whatever
18:40
alkisg: and were you saying we can mirror the git branch as a bzr branch for translations in an automated way?
18:42
ok, ldm pushed
18:46
and ltspfs pushed
18:47
last commit to ltspfs was in late 2014
18:47
really minor packaging changes in late 2015
18:47
guess it's been working
18:47
or it's so broken people are too embarrased to report bugs
18:48* vagrantc suspects it works well enough
18:48
<vagrantc>
sometimes the sign of a good project is the lack of needing to make lots of updates
18:49
i should really prioritize the libpam-* stuff so as to stop having to ask people to stop abusing themselves with LDM_XSESSION
18:52
alkisg: should we switch the default VCS to git now... that might "break" the translations since the URLs will change
18:57markus_e92 has left IRC (markus_e92!~markus_e9@62-46-26-118.adsl.highway.telekom.at, Ping timeout: 260 seconds)
18:59
<vagrantc>
there were two mails sent to ltsp-developer in 2016, and one of them was spam
19:01
<alkisg>
Hehe
19:01
One for 2017 then will be enough
19:01markus_e92 has joined IRC (markus_e92!~markus_e9@62-46-26-118.adsl.highway.telekom.at)
19:01
<alkisg>
We can mirror git to bzr, but I don't know yet how to properly set up the translations
19:02
I'll look it up during the sch-scripts internationalization
19:02
I don't think we have many new translations coming nowadays, so I hope we won't lose much
19:04
<vagrantc>
wpw. o
19:04
wow, i'm really confused about what this switch to git button does
19:05
ok, so https://code.launchpad.net/~ltsp-upstream/ still shows bzr branches
19:06
https://code.launchpad.net/ltsp/ shows no branches anymore
19:08
apparently i need to import them from ~ltsp-upstream user to the ltsp project ... ?!
19:25
needed to mark the individual git branches as part of the ltsp project
19:25
done for now
19:37
announced to ltsp-developer
20:30Statler has left IRC (Statler!~Georg@mail.lohn24.de, Remote host closed the connection)
21:08kjackal_ has joined IRC (kjackal_!~quassel@ppp-2-87-52-67.home.otenet.gr)
21:13kjackal_ has left IRC (kjackal_!~quassel@ppp-2-87-52-67.home.otenet.gr, Ping timeout: 246 seconds)
21:27ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving)
21:44GodFather has joined IRC (GodFather!~rcc@2600:1008:b028:e943:9dd7:37e9:c80c:d831)
22:41ben_roose has left IRC (ben_roose!~roose@roose.cs.wichita.edu, Quit: Leaving)
23:25yanu_ has left IRC (yanu_!~yanu@178-116-57-70.access.telenet.be, Ping timeout: 240 seconds)
23:36vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
23:38JerryT has left IRC (JerryT!~jerry@wsip-70-165-106-163.om.om.cox.net, Ping timeout: 268 seconds)
23:47GodFather has left IRC (GodFather!~rcc@2600:1008:b028:e943:9dd7:37e9:c80c:d831, Ping timeout: 272 seconds)