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


Channel log from 6 May 2015   (all times are UTC)

00:47telex has left IRC (telex!teletype@freeshell.de, Remote host closed the connection)
00:48telex has joined IRC (telex!teletype@freeshell.de)
01:52circ-user-1K6fy has joined IRC (circ-user-1K6fy!~circuser-@2601:4:900:72a:685d:9ba6:bd5c:2f8b)
04:03TatankaT_ has left IRC (TatankaT_!~tim@193.190.253.114, Ping timeout: 240 seconds)
04:04TatankaT has joined IRC (TatankaT!~tim@193.190.253.114)
05:17work_alkisg is now known as alkisg
05:33ricotz has joined IRC (ricotz!~rico@ubuntu/member/ricotz)
05:49book` has left IRC (book`!~book`@105.ip-167-114-152.net, Ping timeout: 248 seconds)
06:05book` has joined IRC (book`!~book`@105.ip-167-114-152.net)
06:16mikkel has joined IRC (mikkel!~mikkel@mail.dlvs.dk)
06:18vlt has joined IRC (vlt!~nobody@lvps87-230-93-209.dedicated.hosteurope.de)
06:27AlexPortable has joined IRC (AlexPortable!uid7568@gateway/web/irccloud.com/x-vrbgakbqtgezjtwz)
06:39* highvoltage wonders if vagrant had a hand in https://www.linux.com/news/featured-blogs/200-libby-clark/827669-video-84-year-old-volunteer-rebuilds-sends-linux-laptops-to-africa
07:08
<muppis>
Which dhcpd is used in 14.04 to boot client?
07:10
<alkisg>
muppis: isc-dhcp as usual?
07:10
But you can install dnsmasq too...
07:10
Or are you asking the exact version/
07:10
?
07:11
http://packages.ubuntu.com/search?keywords=isc-dhcp-server
07:13
<muppis>
Just asking between isc-dhcp / dnsmasq, because none seems to be installed by default.
07:14
Or I'm just not looking from right place...
07:15
<alkisg>
Did you install ltsp-server, or ltsp-server-standalone?
07:18
<muppis>
-standalone, but I was just looking from wrong place. Founded already. :D
07:24
Hmm.. Created a LDAP user.
07:26
Äh. Didn't meant to write that. :D
07:26
Couldn't log in Lubuntu desktop from client. Failsafe Xterm works.
07:27
Openbox as well.
07:28
<alkisg>
muppis: there's a bug report about lubuntu.desktop session in ltsp
07:29
<muppis>
Oh, now I remember. You mentioned that before. :D
07:33
Thank you, again. :)
07:33
<alkisg>
np
08:04
<muppis>
!lubuntu
08:04
<ltsp`>
Error: "lubuntu" is not a valid command.
08:09
<muppis>
!learn lubuntu as To fix log in problem with Lubuntu-desktop run 'sudo dpkg-reconfigure dash' and answer No to question to use dash as /bin/sh.
08:09
<ltsp`>
The operation succeeded.
08:09
<muppis>
!learn lubuntu as To fix log in problem with Lubuntu-desktop run 'sudo dpkg-reconfigure dash' at server and answer No to question to use dash as /bin/sh.
08:09
<ltsp`>
The operation succeeded.
08:10
<muppis>
!learn lubuntu as To fix login problem with Lubuntu-desktop run 'sudo dpkg-reconfigure dash' at server and answer No to question to use dash as /bin/sh.
08:10
<ltsp`>
The operation succeeded.
08:10
<muppis>
:D
08:12
<alkisg>
Better point to the bug report instead, as that's a really bad fix
08:12
It might break a lot of things and make the system slower
08:21
<muppis>
!help
08:21
<ltsp`>
(help [<plugin>] [<command>]) -- This command gives a useful description of what <command> does. <plugin> is only necessary if the command is in more than one plugin.
08:22
<muppis>
!learn lubuntu as To fix login problem with Lubuntu-desktop check this bug report: https://bugs.launchpad.net/ubuntu/+source/ltsp/+bug/1330252
08:22
<ltsp`>
The operation succeeded.
08:23
<alkisg>
!lubuntu
08:23
<ltsp`>
lubuntu: (#1) To fix log in problem with Lubuntu-desktop run 'sudo dpkg-reconfigure dash' and answer No to question to use dash as /bin/sh., or (#2) To fix log in problem with Lubuntu-desktop run 'sudo dpkg-reconfigure dash' at server and answer No to question to use dash as /bin/sh., or (#3) To fix login problem with Lubuntu-desktop run 'sudo dpkg-reconfigure dash' at server and answer No to question to use dash as (1 more message)
08:23
<alkisg>
!forget lubuntu 1
08:23
<ltsp`>
The operation succeeded.
08:23
<alkisg>
!forget lubuntu 1
08:23
<ltsp`>
The operation succeeded.
08:23
<alkisg>
!forget lubuntu 1
08:23
<ltsp`>
The operation succeeded.
08:23
<alkisg>
!lubuntu
08:23
<ltsp`>
lubuntu: To fix login problem with Lubuntu-desktop check this bug report: https://bugs.launchpad.net/ubuntu/+source/ltsp/+bug/1330252
08:25
<alkisg>
muppis: did you try my workaround from comment #3? Did it work for you?
08:26
<muppis>
Yes, it worked.
08:41
Now I need figure out how to transfer punch of LDAP users from old system.
08:42don384 has joined IRC (don384!4f8607ab@gateway/web/freenode/ip.79.134.7.171)
08:43
<don384>
hello
08:44
<alkisg>
hi
08:44
<don384>
help me, centos 6.6 install
08:44
ltsp install ok
08:46
<alkisg>
I don't think anyone maintains LTSP for CentOS, you should probably try another distribution like Debian, Ubuntu, OpenSUSE etc
08:47
<don384>
k12linux installs and works ok
08:47
<alkisg>
And what is the problem?
08:47
<don384>
new ltsp problems
08:48
<alkisg>
What problems?
08:48
<don384>
ltsp-update-image
08:49
doesn't work
08:49
<alkisg>
Yes LTSP doesn't work on CentOS
08:49
Because there is no maintainer there
08:49
<don384>
it's so sad
08:50
<alkisg>
Yup... unfortunately it's not an LTSP issue that noone here uses that distribution...
08:50
s/uses/maintains/
08:51
<don384>
can i do ot myself somehow? is it hard?
08:51
<alkisg>
Can you write bash scripts?
08:51
<don384>
yes i can
08:52
<alkisg>
It involves solving all the problems you find, and then merging the solutions to the ltsp code base
08:53
<don384>
ok, thank you? will try
08:58vsuojanen has left IRC (vsuojanen!~valtteri@83-136-248-31.uk-lon1.host.upcloud.com, *.net *.split)
08:58jbermudes_ has left IRC (jbermudes_!~quassel@76.74.178.212, *.net *.split)
09:00vsuojanen has joined IRC (vsuojanen!~valtteri@83-136-248-31.uk-lon1.host.upcloud.com)
09:00jbermudes_ has joined IRC (jbermudes_!~quassel@76.74.178.212)
09:06don384 has left IRC (don384!4f8607ab@gateway/web/freenode/ip.79.134.7.171, Ping timeout: 246 seconds)
09:17
<highvoltage>
:)
09:59adrianorg has left IRC (adrianorg!~adrianorg@177.204.158.154.dynamic.adsl.gvt.net.br, Ping timeout: 240 seconds)
10:01adrianorg has joined IRC (adrianorg!~adrianorg@177.156.224.242)
10:28highvoltage has left IRC (highvoltage!~highvolta@ubuntu/member/highvoltage, Ping timeout: 264 seconds)
10:31alkisg is now known as work_alkisg
10:52highvoltage has joined IRC (highvoltage!~highvolta@ubuntu/member/highvoltage)
11:00telex has left IRC (telex!teletype@freeshell.de, Remote host closed the connection)
11:02telex has joined IRC (telex!teletype@freeshell.de)
11:05professor has joined IRC (professor!c3391360@gateway/web/freenode/ip.195.57.19.96)
11:05professor is now known as Guest16001
11:06Guest16001 has left IRC (Guest16001!c3391360@gateway/web/freenode/ip.195.57.19.96, Client Quit)
11:21Faith has joined IRC (Faith!~paty@unaffiliated/faith)
11:54khildin has joined IRC (khildin!~khildin@ip-213-49-85-178.dsl.scarlet.be)
12:59khildin has left IRC (khildin!~khildin@ip-213-49-85-178.dsl.scarlet.be, Ping timeout: 276 seconds)
13:07mgariepy_ has left IRC (mgariepy_!~mgariepy@styx-204.ccs.usherbrooke.ca, Quit: Leaving)
13:07mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy)
14:14mikkel has left IRC (mikkel!~mikkel@mail.dlvs.dk, Quit: Leaving)
15:11work_alkisg is now known as alkisg
15:12nocturn has joined IRC (nocturn!~nocturn@unaffiliated/nocturn)
15:12
<nocturn>
Hi, Is LTSP still working on Ubuntu 14.04
15:14
<alkisg>
It should be
15:19
<nocturn>
Thanks alkisg
16:34vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
17:00
<vagrantc>
alkisg: how does the localapp switch to using unicode work with older versions that still use strings?
17:00
alkisg: is it backwards-compatible?
17:02
<alkisg>
vagrantc: the change only affects non-ascii characters
17:02
If 8u is used, non-ascii work, otherwise they don't
17:02
<vagrantc>
yes, i understand that
17:02
<alkisg>
The reader side doesn't use 8s or 8u, it's the same on both cases
17:02
<vagrantc>
aha
17:03
<alkisg>
Open two tabs:
17:03
terminal tab1: xprop -notype -root -spy LTSP_COMMAND | while read junk_x junk_y LTSP_COMMAND ; do LTSP_COMMAND=${LTSP_COMMAND%\"}; LTSP_COMMAND=${LTSP_COMMAND#\"}; echo "To: $LTSP_COMMAND"; done
17:03
terminal tab2: xprop -root -f LTSP_COMMAND 8s -set LTSP_COMMAND "Hi - Γεια"
17:03
That breaks the "Γεια" string, but of course the ascii part, "Hi", works
17:04
If you switch 8s to 8u, both work
17:04
(a LANG=xx.UTF-8 locale is needed of course)
17:06
scripts that use xprop 8u: ./ltsp-remoteappsd ./doc/examples/ltsp-restart.desktop ./doc/examples/ltsp-shutdown.desktop ./ltsp-localappsd ./ltsp-remoteapps ./ltsp-localapps ./ltsp-genmenu
17:07
<vagrantc>
ok, good.
17:07
unicode attempted to maintain comptibility with ASCII, so that moving forward would break as little as possibl
17:08
so good.
17:08
<alkisg>
vagrantc: in a couple of places we're using LANG=C
17:08
<vagrantc>
true enough
17:08
<alkisg>
I think it would be better if we used LANG=C.UTF-8
17:08
What do you think?
17:08* vagrantc isn't sure when C.UTF-8 was introduced in Debian
17:09
<vagrantc>
i think it was fairly recent
17:10
of course, having been using debian for ~15 years ... recent is a relative term ...
17:10
first mention in the debian locales changelog is 2011 ...
17:11
so it should be fine to switch, at least from a debian perspective
17:12
<alkisg>
I see /usr/share/locale/C.UTF-8 in a wheezy chroot...
17:12
<vagrantc>
looks like it was introduced in wheezy, it's not available by default in squeeze
17:14
so i'm fine with switching :)
17:14
<alkisg>
Nice :)
17:14* vagrantc stares at ALTLinux
17:15
<alkisg>
vagrantc: stgraber mentioned he's mostly OK with syncing debian ltsp with ubuntu's
17:15
Do you want us to arrange some time to work on that?
17:15
<vagrantc>
that'd be nice
17:16
<alkisg>
Many people are already using your packaging (from the Greek schools PPA), so we shouldn't need much work...
17:16
<vagrantc>
i'm sure whatever divergence needs to stay could be made conditional at build-time
17:16
now would be a good time to start, since it's early in the debian release cycle
17:16
<alkisg>
It's early in Ubuntu 15.10 too, and it's not an LTS release either
17:16
stgraber: ^?
17:17
highvoltage: ^ I think it will affect the edubuntu ltsp installer too?
17:17
Do you need something there?
17:17
<vagrantc>
i won't have much time this week, but perhaps the following week
17:17
<alkisg>
Note that we can probably add ltsp-pnp or even casper support in the installer as well
17:18
<vagrantc>
yes, i wanted to make ltsp-pnp an option for the installer
17:18
dunno what casper support entails
17:18
<alkisg>
Ubuntu live CDs have a squashfs file system
17:19
<vagrantc>
debian-live was at one point a fork of casper?
17:19
<alkisg>
One can expose that one over the network, for live fat client demonstration
17:19
<vagrantc>
ah, so basically livecd ready ltsp-pnp?
17:19
<alkisg>
Yup
17:19
<vagrantc>
nice
17:19
<alkisg>
ltsp-pnp plays anywhere :P
17:19
<vagrantc>
could probably do the same with debian-live
17:20telex has left IRC (telex!teletype@freeshell.de, Remote host closed the connection)
17:20
<alkisg>
Yes, I think so, but it would need ltsp bits inside the initramfs
17:20
So a custom debian-live cd
17:20
...while the edubuntu dvd is already ...customized
17:20
<vagrantc>
simply a matter of installing ltsp-client into the build image
17:20
pretty much...
17:20
<alkisg>
And a "start live server" script, yup
17:21
vagrantc: I can also implement a real "guest user" if we want it
17:21
For fat clients only, user guest/password guest everywhere
17:21
<vagrantc>
sure
17:21
<alkisg>
No sshfs, just nbd-mounted /home, or tmpfs
17:22
So that we don't have to add a bunch of users to the live server, for the clients to login with
17:22
<vagrantc>
if it's a guest session, why have a persistant home at all?
17:22telex has joined IRC (telex!teletype@freeshell.de)
17:22
<alkisg>
nbd-client mounts a swap /home, then mkfs ext4 over it, and use it as a temporary home
17:23
<vagrantc>
swap home?
17:23
<alkisg>
The swap export of nbd
17:23
Just to have 512 MB as /home
17:23
<vagrantc>
why not just leave it in tmpfs? to save ram?
17:23
<alkisg>
It's an easy way to have 512 MB of /home, when one doesn't have enough ram
17:23
Yes
17:24
And it's easy to increase the temporary /home too, with NBDSWAP=xxx size
17:24
SWAPSIZE or whatever we call it
17:24
Of course if the nbd mount fails, we can fall back to tmpfs
17:24
<vagrantc>
alkisg: we'd probably want to create another volume for all that
17:24
alkisg: in /etc/nbd-server/config.d/
17:24
<alkisg>
A second nbd-client call
17:25
Ah
17:25
Why?
17:25
<vagrantc>
if the client uses swap, it will be exporting the same file
17:25
<alkisg>
No
17:25
The client does: nbd-client -N swap /dev/nbd1
17:25
That's for swap
17:25
Then, nbd-client -N swap /dev/nbd2
17:25
That's for home
17:25
mkfs.ext4 /dev/nbd2 etc
17:26
<vagrantc>
yes, and the nbd-server will return the same file in both cases
17:26
<alkisg>
The server then makes 2 different files
17:26
Hmmm
17:26
You're right
17:26
<vagrantc>
that would be nice, but it's not how it's set up currently
17:26
<alkisg>
We should remove the nbd file immediately after creation
17:27
That way we wouldn't have such issues
17:27
...but I don't think nbd-server supports that easily...
17:27
OK, another .conf then
17:27
You're right
17:27
<vagrantc>
right
17:28
ideally, it'd create the file and then delete it, but hold the inode open until the nbd-server process was done
17:28
<alkisg>
Yes
17:28
<vagrantc>
which i think is how we had it back when we called from inetd, but that had other issues
17:29
<alkisg>
But anyway with the separate .conf there are other advantages
17:29
E.g. one can have kiosk images with stock /home
17:29
That gets reset on logon
17:29
<vagrantc>
true enough
17:29
<alkisg>
(the nbd-server cow trick)
17:30
<vagrantc>
oh yeah, that's supported now
17:30
has been for some while, i guess
17:31* vagrantc will have to switch to defaulting to NBD for stretch until the overlay + NFS issues are sorted out
17:31
<vagrantc>
(stretch == jessie+1)
17:32
actually... it'd be nice to be able to try multiple different rootfs methods at boot
17:33
<alkisg>
vagrantc: we already support easy switching between nfs, nbd, aoe, right?
17:33* vagrantc might have to use a custom boot script again as the initramfs-tools scripts are getting more and more problematic
17:33
<vagrantc>
alkisg: yes, but you have to define them at the boot prompt
17:33
alkisg: what if it detected them at boot?
17:33
<alkisg>
automatically?
17:34
<vagrantc>
yes
17:34
:)
17:34
<alkisg>
See if the server supports nbd, otherwise use nfs, etc etc?
17:34
<vagrantc>
at least as an option, or a fallback
17:34
<alkisg>
We'd have to use our own script then
17:34
<vagrantc>
right
17:34
<alkisg>
Hmmm or rather... :D
17:34
How hacky do you think mount --bind /proc/cmdline is, for a few seconds? :D
17:34
Haha
17:35
<vagrantc>
??
17:35
<alkisg>
Well
17:35
Suppose we have a script at init-premount
17:35
That initializes networking etc
17:35
<vagrantc>
oh, i see where this is going
17:35
<alkisg>
Then, it port scans the server and checks for nbd/nfs whatever
17:35
But, in order to avoid rewriting nbd/nfs support, it fools those scripts with /proc/cmdline
17:35
Then, at init-bottom, it unmounts /proc/cmdline
17:36
Messy, but it should be rather portable
17:36
<vagrantc>
also, i need to get the loop mount support integrated into initramfs-tools
17:36
i think ubuntu's still carrying a patch for that, and it's silly that debian doesn't support it
17:37
<alkisg>
I think the best thing there would be sequential mount=xxx directives
17:37
mount=nfs, then mount=loopback, then mount=nbd...
17:37
And one could do anything then... the nfs/loopback/nbd would be simple initramfs plugins
17:38khildin has joined IRC (khildin!~khildin@ip-213-49-85-178.dsl.scarlet.be)
17:40
<vagrantc>
well, predictibly, bind-mounting /proc/cmdline is a little sketchy :)
17:41
at least, it doesn't seem obviously a good thing :)
17:41
<alkisg>
Essentially the "mount" cmdline parameters I'm talking about are like /etc/fstab lines, just not space-separated
17:41
If those mount parameters were supported, we wouldn't need such hacks
17:41
We would just have an ltsp mount plugin and an ltsp mount cmdline parameter
17:42
The problem is in the initramfs design, it doesn't provide appropriate hooks
17:42
If there are no plans to use dracut etc, maybe it'd be better to start a long conversation about it with Maks...
17:43
<vagrantc>
the initramfs-tools folsk are hesitant to incorporate anything that doesn't emulate linux built-in bootprompt syntax ...
17:44
<alkisg>
Sure, but when there are actual problems (like lack of hooks), they should provide some solutions, other than rewriting nfs/nbd/aoe/whatever scripts
17:46
Btw I'm playing with UEFI these days... I've succeeded doing a lot of things, even booting 32bit kernels under 64bit uefi, but I'm missing a few others, like proxydhcp support...
17:49khildin has left IRC (khildin!~khildin@ip-213-49-85-178.dsl.scarlet.be, Ping timeout: 246 seconds)
17:51
<alkisg>
https://www.mail-archive.com/dnsmasq-discuss@lists.thekelleys.org.uk/msg09234.html
17:52
If we lose proxydhcp support in uefi, I'll start thinking of using dhcp over ipv6 for the initial pxe request, and then use ipv4 inside the initramfs :D
17:53
<vagrantc>
wow!
17:53* alkisg needs to go, brb
17:53
<alkisg>
bbl
17:53* vagrantc waves
17:53alkisg is now known as work_alkisg
17:55
<vsuojanen>
I was looking Ubuntu wiki page UbuntuLTSP/ltsp-pnp but it was not listed in the main page https://help.ubuntu.com/community/UbuntuLTSP
17:57
it is there but I was wondering why left out from the main list
18:32danau11 has joined IRC (danau11!~durban@12.197.179.122)
18:32danau11 has left IRC (danau11!~durban@12.197.179.122)
18:47vsuojanen has left IRC (vsuojanen!~valtteri@83-136-248-31.uk-lon1.host.upcloud.com, *.net *.split)
18:47jbermudes_ has left IRC (jbermudes_!~quassel@76.74.178.212, *.net *.split)
18:47cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, *.net *.split)
18:47uXus has left IRC (uXus!~uXus@217.77.222.72, *.net *.split)
18:47book` has left IRC (book`!~book`@105.ip-167-114-152.net, *.net *.split)
18:49book` has joined IRC (book`!~book`@105.ip-167-114-152.net)
18:49cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg)
18:49uXus has joined IRC (uXus!~uXus@217.77.222.72)
18:49vsuojanen has joined IRC (vsuojanen!~valtteri@83-136-248-31.uk-lon1.host.upcloud.com)
18:49jbermudes_ has joined IRC (jbermudes_!~quassel@76.74.178.212)
19:18cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, *.net *.split)
19:18uXus has left IRC (uXus!~uXus@217.77.222.72, *.net *.split)
19:19cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg)
19:19uXus has joined IRC (uXus!~uXus@217.77.222.72)
20:21Faith has left IRC (Faith!~paty@unaffiliated/faith, Quit: Saindo)
20:45Grembler has joined IRC (Grembler!~Ben@cpc29-aztw22-2-0-cust128.18-1.cable.virginm.net)
21:02Grembler has left IRC (Grembler!~Ben@cpc29-aztw22-2-0-cust128.18-1.cable.virginm.net, Quit: I Leave)
21:34vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
21:57AlexPortable has left IRC (AlexPortable!uid7568@gateway/web/irccloud.com/x-vrbgakbqtgezjtwz, Quit: Connection closed for inactivity)
22:09ricotz has left IRC (ricotz!~rico@ubuntu/member/ricotz, Quit: Ex-Chat)
22:32FGXR6 has left IRC (FGXR6!~phantom@ppp121-44-60-22.lns20.syd4.internode.on.net, Ping timeout: 250 seconds)
22:33FGXR6 has joined IRC (FGXR6!~phantom@ppp121-44-60-22.lns20.syd4.internode.on.net)
23:05telex has left IRC (telex!teletype@freeshell.de, Remote host closed the connection)
23:06telex has joined IRC (telex!teletype@freeshell.de)