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


Channel log from 14 January 2013   (all times are UTC)

00:15gentgeen__ has left IRC (gentgeen__!~kevin@c-98-236-71-64.hsd1.pa.comcast.net, Ping timeout: 272 seconds)
01:09shogunx has left IRC (shogunx!~shogunx@2001:4978:106:1:885d:63d1:b453:ba19, Ping timeout: 245 seconds)
01:19Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Ping timeout: 276 seconds)
01:22shogunx has joined IRC (shogunx!~shogunx@2001:4978:106:1:b431:66eb:4a59:56af)
01:51vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Quit: leaving)
02:08adrianorg_ has left IRC (adrianorg_!~adrianorg@177.132.218.134, Ping timeout: 246 seconds)
02:51andygraybeal_ has left IRC (andygraybeal_!~andy@h44.218.213.151.dynamic.ip.windstream.net, Ping timeout: 255 seconds)
03:08PhoenixSTF has left IRC (PhoenixSTF!~rudi@78.29.134.164, Remote host closed the connection)
04:32jammcq has left IRC (jammcq!~jam@c-98-209-67-190.hsd1.mi.comcast.net, Quit: leaving)
05:30telex has left IRC (telex!~telex@freeshell.de, Ping timeout: 264 seconds)
05:32telex has joined IRC (telex!~telex@freeshell.de)
05:36telex has left IRC (telex!~telex@freeshell.de, Read error: Connection reset by peer)
05:38telex has joined IRC (telex!~telex@freeshell.de)
05:43muppis_ is now known as muppis
06:18komunista has joined IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk)
06:28work_alkisg is now known as alkisg
06:34komunista has left IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk, Quit: Leaving.)
06:56vmlintu has joined IRC (vmlintu!~vmlintu@cs180030.pp.htv.fi)
07:11vilas has joined IRC (vilas!31f881b2@gateway/web/freenode/ip.49.248.129.178)
07:12
<vilas>
I have got the error - LTSP - login fails - no response from server
07:13
I have move arch i386 to amd64, also update the sshkey and kernels and rebuild image
07:14
in arch i386, users can login but in amd64 it is showing "no response frome server"
08:10
<Hyperbyte>
vilas, you have two chroots, one for i386 and one for amd64, right?
08:10
<vilas>
yes
08:10
in /opt/ltsp
08:10dobber has joined IRC (dobber!~dobber@213.169.45.222)
08:11
<Hyperbyte>
You're using fat clients, I assume?
08:11
<vilas>
yes
08:12
<Hyperbyte>
Okay
08:12
Well I'm thinking it -has- to be the sshkeys
08:12
!screen_02
08:12
<ltsp>
screen_02: To get a root shell on an Ubuntu thin client: https://help.ubuntu.com/community/UbuntuLTSP/ClientTroubleshooting#Using_a_shell_SCREEN
08:12
<Hyperbyte>
Do that, and then on screen 02, the root prompt, on both clients, type:
08:12
ssh <user>@server
08:12
Leave the @server bit intact
08:13
For example, 'ssh vilas@server'
08:13
<vilas>
yes, I can login into server from thinclient
08:13
<Hyperbyte>
If it is an ssh key problem, the i386 chroot should ask if you're sure you want to continue connecting to host 'server'
08:13
<vilas>
via shell
08:13
<Hyperbyte>
And the amd64 chroot should log right in.
08:14
Note that you should try this before you log in via LDM.
08:15
<vilas>
it is not ask sshkey
08:15
just i have reboot the thinclient and agin check it
08:16
<Hyperbyte>
And you can log in from both chroot, without problems, and without confirmation for the ssh key?
08:16
*chroots
08:17
<vilas>
yes
08:18
i have login from thin client via tty1, i can login from user without asking sshkey
08:18
but I can't login from KDE ubuntu 12.04
08:19
from GUI
08:19
<Hyperbyte>
You've lost me. Which chroot is that?
08:19
<vilas>
i can also do chroot
08:19
chroot is /opt/ltsp/amd64
08:20
<Hyperbyte>
So the amd64 is also KDE?
08:20
<vilas>
yes
08:20
<Hyperbyte>
And the i386 chroot is Gnome?
08:20
<vilas>
it's also kde
08:20
<Hyperbyte>
Okay
08:20
09:18 <vilas> but I can't login from KDE ubuntu 12.04
08:20
So what does this mean?
08:22
<vilas>
Ubuntu is base server 64 bit as a GNOME
08:22
and Kubuntu 12.04 fat thin client as a KDE
08:22
64 bit
08:23
<Hyperbyte>
I'm sorry, I need to go... maybe someone else can help you further.
08:23
<vilas>
my thinclient booted in KDE
08:23
successfully
08:23
but users cannot login in
08:24
okey
08:40cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, Ping timeout: 248 seconds)
08:54bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at)
09:49Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
09:55bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Quit: Goin' down hard)
09:56bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at)
10:00bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Client Quit)
10:00Gremble has joined IRC (Gremble!~Ben@cpc29-aztw23-2-0-cust144.18-1.cable.virginmedia.com)
10:00bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at)
10:02bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Client Quit)
10:02bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at)
10:23mkuzu has joined IRC (mkuzu!~quassel@178.250.90.62)
10:44
<vilas>
not able to login through ldm using amd64 thin client but can get tty1 ..can someone help
10:46Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Ping timeout: 248 seconds)
10:50meamy has joined IRC (meamy!~hannes@pd95cdee4.dip0.t-ipconnect.de)
10:51
<alkisg>
!screen_02 | echo vilas:
10:51
<ltsp>
vilas: screen_02: To get a root shell on an Ubuntu thin client: https://help.ubuntu.com/community/UbuntuLTSP/ClientTroubleshooting#Using_a_shell_SCREEN
10:51
<alkisg>
vilas: from there, type: ssh vilas@server
10:51
Substitute vilas with an existing username, but leave "server" exactly as it's written
10:52
<vilas>
give me minute checking ....
10:53
can i replace the ip with "server"
10:54
<alkisg>
vilas: do that from screen_02 on a client
10:54
And, don't put an ip there
10:54
Leave the "server" word
10:54
<vilas>
ok
10:55
screen_02 means tty1 or tty2 ?
10:55
i m getting tty1
10:55
<alkisg>
Did you read the link above?
10:56
You're supposed to read the wiki page above, first
10:56
<vilas>
ok
10:56
okey
10:56
i have create a lts.conf file
10:56
just a moment
10:59Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
11:04
<vilas>
ctrl+alt+f2 getting direct root login
11:04
and I can login as user in server via ssh "$ ssh vilas@server"
11:05
but I am not getting log in via KDE GUI mode
11:07
<meamy>
hi i have a problem. I'm stetting up an LTSP-Cluster environment, I was able to get the root, control and the appserver running (the dhcp server runs on the root server and the load balancer on the control server) and also to create one amd64 client chroot thats boots. But now I want to bluild an nother amd64 client (test1) with a different chroot. the Clientcomputer gets the correct pxelinux.0 file but it boot in to a busyboxshell and on the
11:07
server side i get an syslog error message: nbd_server[936]: negotiation failed; nbd_server[936]: Exiting.
11:07
When I switch back to the amd64 client (in the dhcpd.conf) every think is working again. The different Server and Chroots are all 12.04 Ubuntu
11:08
any idear how i can solve this
11:08
?
11:09
<vilas>
@alkisg: gone throught wiki but not able to login through ldm
11:09
<alkisg>
(01:04:52 μμ) vilas: and I can login as user in server via ssh "$ ssh vilas@server" ==> did you get a warning about "trusting server xxx" ?
11:09
<vilas>
no
11:10
<alkisg>
vilas: on the server, check /var/log/auth.log
11:10
See if you get entries for the user that tries to log in
11:10
(tries to log in on the client)
11:14
<knipwim>
Phantomas: reading yesterday's logs
11:14
yes, I execcute epoptes as root
11:14
<alkisg>
knipwim: that's not yet supported
11:14
Try as a user in the epoptes group instead
11:14
<knipwim>
the ip command is not available for regular users
11:14
only for root
11:15
<vilas>
<alkisg> : Jan 14 16:44:25 EV-TC-SRV01 sshd[22388]: Accepted password for vikas from 172.16.0.20 port 49461 ssh2 Jan 14 16:44:25 EV-TC-SRV01 sshd[22388]: pam_unix(sshd:session): session opened for user vikas by (uid=0) Jan 14 16:44:26 EV-TC-SRV01 sshd[22528]: subsystem request for sftp by user vikas Jan 14 16:44:26 EV-TC-SRV01 sshd[22528]: Received disconnect from 172.16.0.20: 11: disconnected by user Jan 14 16:44:26 EV-TC-SRV01 ssh
11:15
<knipwim>
that's way started it as root in the first place
11:15
<alkisg>
knipwim: you mean wrt the defined PATH?
11:15
knipwim: because I can run `ip` fine here as non-root...
11:15
knipwim: can you try to symlink ip to e.g. /usr/bin/ip ?
11:15
Just as a test...
11:15
<knipwim>
it's in /sbin/ip
11:16
when i get the whole path it works
11:16
<alkisg>
knipwim: yes, and /sbin is in my user path here
11:16
<vilas>
@alkisg: output of /var/log/auth.log
11:16
<knipwim>
so a symlink to /usr/bin/ip might work
11:16
<alkisg>
knipwim: $ echo $PATH
11:16
/home/alkisg/bin:/usr/lib/lightdm/lightdm:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
11:16
knipwim: or, add /sbin to the user path temporarily
11:16
vilas: sounds like the user does log in
11:17
<vilas>
@alkisg: yes
11:17
<alkisg>
vilas: and then maybe xorg crashes, and it makes you think it doesn't log in
11:17
<knipwim>
alkisg: for testing that might work
11:17
<alkisg>
vilas: check /var/log/Xorg.7.log (or .old), and the user's .xsession-errors
11:17
The first one from screen_02 on the client
11:17
The second one, on the server
11:17
/home/vilas/.xsession-errors
11:18
<vilas>
@alkisg: but not any users can login throgh ldm
11:18
<knipwim>
alkisg: i'll test when i get home
11:18
thx for now
11:18
<alkisg>
np
11:19
vilas: yes, it sounds like a problem in either graphics drivers, or in KDE, you need to troubleshoot those, I gave you some hints on how to start above.
11:21yalu has left IRC (yalu!~yalu@91.180.93.68, Read error: Operation timed out)
11:21Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Remote host closed the connection)
11:22Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
11:22knipwim has left IRC (knipwim!~wim@ip4da83870.direct-adsl.nl, Read error: Operation timed out)
11:22yalu has joined IRC (yalu!~yalu@91.180.93.68)
11:22knipwim has joined IRC (knipwim!~wim@ip4da83870.direct-adsl.nl)
11:22
<vilas>
@alkisg: Xsession: X session started for vikas at Mon Jan 14 16:51:36 IST 2013
11:23
@alkisg: open: Permission denied
11:24
@alkisg: in /home/vikas/.xsession-erros
11:26
<meamy>
mm nobody saw this error before "nbd_server[936]: negotiation failed; nbd_server[936]: Exiting"? What im trying to do is to get mutiple chroots working in one Cluster environment. so that i can have for example 10 clientcomputer with the default chroot, and different spezialcomputer with they own chroot. THe way i want to do this is over the dhcp server but for strange reason i just get one chroot booting and for the other this strange error in
11:26
<alkisg>
vilas: right, work on those errors, it's not ltsp related anymore...
11:26
<meamy>
the server syslog
11:30mkuzu has left IRC (mkuzu!~quassel@178.250.90.62, Remote host closed the connection)
11:31
<vilas>
@alkisg: for your information when i am using 32bit arch i am able login through ldm but when switch to 64 bit arch no success..anyway thankyou..for your support...
11:31xet7 has joined IRC (xet7!~xet7@a91-156-233-189.elisa-laajakaista.fi)
11:34
<Hyperbyte>
meamy, interesting bits are probably on the client, not the server.
11:34
Do this:
11:34
!quiet-splash
11:34
<ltsp>
quiet-splash: to disable the splash screen in Ubuntu, in order to see any boot error messages, run sudo gedit /var/lib/tftpboot/ltsp/i386/pxelinux.cfg/default and remove quiet splash plymouth:force-splash vt.handoff=7
11:34
<Hyperbyte>
For the affected chroot. You'll see the machine spit out lots of debug information, the last bits before it stops will tell why it has stopped. :)
11:41
<meamy>
ok i did that: block nbd0: Receive control failed (result -32); block nbd0: shutting down socket; block nbd0: queue cleared
11:43alkisg is now known as work_alkisg
11:44
<meamy>
mm and then it fails to mount the root file system
12:02adrianorg has joined IRC (adrianorg!~adrianorg@187.113.219.235)
12:09[GuS] has joined IRC ([GuS]!~MysT@213-117-16-190.fibertel.com.ar)
12:09[GuS] has joined IRC ([GuS]!~MysT@unaffiliated/gus/x-663402)
12:13
<Hyperbyte>
meamy, you could try mounting the filesystem manually, from another client to see what's going wrong
12:14
But my bet is that you just have something wrong in either your client config in /var/lib/tftpboot/ or your nbd server config in /etc/nbd-server
12:17
<meamy>
yep the weird thing is that they are expect from the name similar chroots. As soon as I get the dmesg file out of the client (the busybox has no netcat or telnet installed arg) i post it
12:21
<vilas>
@alkisg: @Hyperbyte -- SOLVED
12:22
@alkisg: @Hyperbyte -- SOLUTIONS:
12:22
@alkisg: @Hyperbyte -- sudo chmod 777 /var/tmp
12:23
@alkisg: @Hyperbyte -- sudo chmod o+t /var/tmp
12:23
@alkisg: @Hyperbyte -- sudo chmod 777 /tmp
12:23
@alkisg: @Hyperbyte -- sudo chmod o+t /tmp
12:24
@alkisg: @Hyperbyte -- Thanks once again
12:29
<Hyperbyte>
:)
12:34JuJuBee has joined IRC (JuJuBee!~mike_knic@24-148-112-102.ip.mhcable.com)
12:34
<meamy>
@Hyperbyte in /etc/nbd-server/conf.d/ are two config files they are similar expect from the name (ltsp_amd64.conf ltsp_test1.conf). how do i mount the root file system from the working client out of the initramfs shell. tried to mount a nfs dirktion to safe the dmesg file but that was not working (it freezed without any error)
12:35PatrickDickey has joined IRC (PatrickDickey!~quassel@2001:470:1f11:830:b064:da1b:647:e431)
12:35
<JuJuBee>
Hyperbyte: Thanks for the help
12:36
<Hyperbyte>
meamy, the two config files should not be similar. They should have a different exportname in them.
12:36
JuJuBee, sure, hi. :)
12:38
<JuJuBee>
Hyperbyte: so I already run an ubuntu server (file server, dhcp, iptables, dansguardian, squid, http(s) etc) for my class. My school just got new computers and offered them to me and we will re-distribute mine. The rest of the school uses Windoze on the schools domain. I have my own dedicated internet connection (cable modem).
12:38
<Hyperbyte>
I'm using LTSP at work here (for ~10 thin clients or so)... we wanted to elimiate noise & heat at the workplaces... so all clients run their software on the server.
12:39
I also vlunteer at a local radio station, there we don't have such a server beast, and all clients are basically things we can get left and right whenever we have an opportunity... so those are fat clients. All clients have same software, it's deployed over the network.
12:39
:)
12:39
JuJuBee, you could do two things.
12:39
1. run regular LTSP
12:39
2. run LTSP "pnp" edition
12:39
<meamy>
Hyperbyte, sorry my mistake they also have different exportnames (according to they names so they are correct). yeah that is basicly what im trieing to do a network mixed with fat clients an thin clients
12:40
<Hyperbyte>
The first one, creates a seperate OS for the client, inside the server OS, where you can maintain everything and when done, build an image for all clients to use.
12:41
The downside to this is that you need to be familiar with console, because you cannot maintain the chroot graphically.
12:41
<meamy>
so you just have one image running in the whole installation?
12:41
<JuJuBee>
Hyperbyte: I am ok with cli
12:42
<Hyperbyte>
Second one, ltsp-pnp, builds the network boot image for the clients from the host (server) OS. This is useful because you can maintain everything graphically. The downside is that, whatever you put on the server, usually finds it's way to the clients unless you specifically tell LTSP not to.
12:43
So first one might be the best choice. You can just apt-get install ltsp-server and build a client. Or, if you want to check it out first, see how it works, run an Edubuntu live CD. It can set up a live thin client server for you.
12:43
<JuJuBee>
Hyperbyte: does that also open up a seurity hole (students have access to things like the mysql db etc now?)
12:43
<Hyperbyte>
JuJuBee, not sure. In your case, I'd go with the first option.
12:43
<JuJuBee>
Ah, I have never used edubuntu
12:43
<Hyperbyte>
You don't have to. I run LTSP on regular Ubuntu.
12:43
<JuJuBee>
Sounds like easier to manage anyway.
12:44
<Hyperbyte>
JuJuBee, I think so, but people who are not familiar with console disagree. :)
12:45
<JuJuBee>
So I can set up one of my newer computers as my new server then (better CPU). Then "sudo apt-get install ltsp-server-standalone" and "sudo ltsp-build-client" and configure the client...?
12:45
I am not very familiar with chroot, but I have done it once or twice. I can read up on it if need be.
12:45
<Hyperbyte>
JuJuBee, -standalone includes dhcp, nameserver, etc. It's for networks that have no server yet whatsoever.
12:46
<JuJuBee>
So if I already have dhcp etc.. what would I install?
12:46
<Hyperbyte>
With clients like yours, you're going to use fat clients. So the hardware strain LTSP places on the server would be minimal. Think 2% load minimal.
12:47
I would set up a test server with Ubuntu, apt-get install ltsp-server, ltsp-build-client and then just see how it works for you
12:47
Eventually, I'd just put LTSP on the existing dhcp/etc server.
12:47
<JuJuBee>
What is diff between fat and thin?
12:47
<Hyperbyte>
Thin and fat clients both network boot and present the user a login prompt to log in to the server with their credentials on the server.
12:48
After that, fat clients download the network image and run all software in the image locally on their own hardware.
12:48
<JuJuBee>
I see.
12:48
<Hyperbyte>
Thin clients however, log in to a graphical desktop session on the server and run all their software on the server's hardware.
12:48
<JuJuBee>
so still no OS installed locally
12:48
with both
12:48
<Hyperbyte>
That's the beauty of LTSP. :)
12:49
<JuJuBee>
OK, I will set up a test server and client and play. To set up server, I just install ubuntu as normal then install ltsp?
12:49
<Hyperbyte>
If I want to modify something on the clients, let's say I want to place a certain shortcut in the Gnome menu... I put it in /usr/share/applications/ in the client OS on the server, run ltsp-update-image.... and then all clients have it after a reboot.
12:49
JuJuBee, yep. :)
12:50
<JuJuBee>
I was reading https://help.ubuntu.com/community/UbuntuLTSP/LTSPQuickInstall but am confused about the section on installing on an already running desktop system... they mentioned a spare nic ?
12:51
<Hyperbyte>
JuJuBee, those instructions are for this user case:
12:51
http://www.ltsp.org/images/ltsp_diagram.gif
12:52
That's the most common usercase. Networks that have no server at all, migrate to a Linux server that does dhcp, dns, ltsp and routing (hence the need for two nics)
12:52
You want LTSP without the -standalone, so just regular ltsp-server
12:52
And you don't need two NICs.
12:52
<JuJuBee>
I have a server with 2 nics as the diagram suggests.
12:52vilas has left IRC (vilas!31f881b2@gateway/web/freenode/ip.49.248.129.178, Quit: Page closed)
12:52
<JuJuBee>
One public facing and one LAN facing
12:53
<Hyperbyte>
Right, if I were you, I'd eventually install LTSP there.
12:53
And create "the diagram" :-)
12:53
<JuJuBee>
OK.
12:53
<Hyperbyte>
However, as a test, I'd just take a seperate server.
12:53
<JuJuBee>
I plan that
12:54
<Hyperbyte>
Since you already have dhcpd and router configured on your Ubuntu server, your eventual setup there would be identical to the one you're doing as a test anyways.
12:54
<JuJuBee>
I may even make my test server my new real server (it has better hardware)
12:54
<Hyperbyte>
:)
12:54
In that case you could also install ltsp-server-standalone
12:54
That configures dhcp and other things needed for LTSP too
12:54
But then you'll have to isolate it from the rest of the network while you test. :)
12:54
<JuJuBee>
If it works well, then I may even give the newer computers to the faculty and keep my older ones which are still dual core with 4G RAM
12:55
<Hyperbyte>
Dual core with 4GB RAM is more than enough to run Ubuntu. :)
12:55
<JuJuBee>
I can create a vlan on my switch to take care of that
12:55
<Hyperbyte>
Well, go for it. :)
12:55
<JuJuBee>
Thanks
12:56PatrickDickey has left IRC (PatrickDickey!~quassel@2001:470:1f11:830:b064:da1b:647:e431, Ping timeout: 245 seconds)
12:56
<Hyperbyte>
If you have further questions, don't hesitate to ask here of course.
12:56
<JuJuBee>
I Appreciate the help.
12:56
<Hyperbyte>
Generally speaking this channel -is- a bit slow-moving, since we're all gainfully employed thankfully, so.... busy busy busy. :)
12:57
But we look from time to time. You know how it goes. :)
12:57
And you're welcome, of course. :)
13:06MastaaK has joined IRC (MastaaK!c3dd3a29@gateway/web/freenode/ip.195.221.58.41)
13:08[GuS] has left IRC ([GuS]!~MysT@unaffiliated/gus/x-663402, Ping timeout: 252 seconds)
13:11gentgeen__ has joined IRC (gentgeen__!~kevin@98.236.71.64)
13:27gvy has joined IRC (gvy!~mike@altlinux/developer/mike)
13:34ltspuser_61 has joined IRC (ltspuser_61!42b0170c@gateway/web/freenode/ip.66.176.23.12)
13:34
<ltspuser_61>
hi everyone
13:35
has anyone here been able to use a squid proxy with ltsp and limit some terminals to internet?
13:36ltspuser_61 has left IRC (ltspuser_61!42b0170c@gateway/web/freenode/ip.66.176.23.12, Client Quit)
13:40mgariepy has joined IRC (mgariepy!mgariepy@ubuntu/member/mgariepy)
13:50alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
13:53
<Hyperbyte>
mhmmm
13:58PhoenixSTF has joined IRC (PhoenixSTF!~rudi@78.29.134.164)
14:11
<MastaaK>
Hello everyone, it's me again... I will make a LTSP server as soon as I can and I would know if it's possible to make a link with samba and ltsp
14:12
I mean when a user log on an ltsp client his documents will be the one on the samba server, even the desktop if it's possible
14:13
I would be glade if someone know or just paste a link!
14:13
<alkisg>
Is that possible with Linux in general?
14:13
If yes, it's also possible with LTSP :)
14:13
<MastaaK>
alkisg: I don't know, that's what I search
14:13
<alkisg>
Why ask in #ltsp then?
14:13
Your distro channel would be more appropriate...
14:14
Or maybe some samba irc channel
14:14
<MastaaK>
I know, but maybe someone here know too...
14:14
<Hyperbyte>
MastaaK, yes, it's possible.
14:14
<alkisg>
Well noone prohibits you from asking that question in the supermarket either... I'm just saying, you have more chances there
14:14
:)
14:15
<MastaaK>
I've already post un #ubuntu channel ;)
14:15
<alkisg>
Try #samba, or the samba channel
14:15
They're the ones that would know about samba
14:16
<Hyperbyte>
alkisg, have you tried #samba?
14:16
<alkisg>
Nope
14:16
<Hyperbyte>
It's way too specific.
14:16
And too quiet. #ubuntu is a good choice.
14:16
<alkisg>
If they don't respond in IRC, they might have mailing lists
14:16
<MastaaK>
Hyperbyte: Can you tell me a link?
14:16
<alkisg>
Or forums
14:17
E.g. there might be problems with the file permissions (samba doesn't support all linux permissions),
14:17
or sockets, or file locks,
14:18
I'd expect samba users to know more about samba than ltsp users...
14:18Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
14:22
<Hyperbyte>
Nah, I think he just wants to map the Windows' profile (My documents, Desktop) dirs to the Linux ones
14:23
<JuJuBee>
Hyperbyte: if I am setting up a new server, do I install ltsp-server or -standalone? I am assuming I don't need to install dhcp3 then?
14:23
<Hyperbyte>
ltsp-server = LTSP
14:23
ltsp-server-standalone = LTSP with dhcp preconfigured
14:23
Right, alkisg?
14:25
<alkisg>
Hyperbyte: yes, ltsp-server-standalone also depends on a dhcp server and some other stuff
14:25
apt-cache show ltsp-server-standalone | egrep '^Depends|^Recommends'
14:25
vs apt-cache show ltsp-server | egrep '^Depends|^Recommends'
14:31
<Hyperbyte>
alkisg is one of the LTSP developers. :-)
14:31
Or, as we like to say:
14:31
!alkisg
14:31
<ltsp>
alkisg: The LTSP oracle. Our beacon of hope in the world of LTSP. With the guidance of this divine emperor, we shall prevail.
14:32
<andygraybeal>
eh, what is everyone doing about this oracle java thing? does anyone bother to install oracle java, or does everyone have the icetea plugin instead?
14:35
<Hyperbyte>
Hi Andy! Depends on the java applet. If it's properly written, it works with icedtea. If not, you need the Oracle java thing. This is hardly LTSP related though. :-)
14:36
<andygraybeal>
hmm, Hyperbyte, well that's not what i was asking about
14:36
i'm wondering if you are all removing oracle java plugin from your ltsp machines.
14:36dead_inside has joined IRC (dead_inside!~dead_insi@76.75.3.174)
14:37
<andygraybeal>
and i figure it is ltsp related, cause i run ltsp and i want to know what you guys are doing too.
14:37
i would be sure that schools would be using java
14:38
and now that there is a big security exploit with oracle java.. i wondered what everyone one else was up to.
14:38
<Hyperbyte>
There is?
14:38
<andygraybeal>
yea, just google 'java' in the news section
14:39
<ogra_>
only in oracles java ... do any distros ship that at all nowadays ?
14:39
<andygraybeal>
no one ships it
14:39
<ogra_>
yeah
14:40
<JuJuBee>
Hyperbyte: if I use -standalone, then I can basically use my current config to modify the config it installs? What dhcp server does it install? it's own or something like dhcp3?
14:40
<alkisg>
Hyperbyte: get to work! https://translations.launchpad.net/ltsp/ltsp-trunk/+lang/nl :P
14:41
JuJuBee: isc-dhcp, but if you already have a dhcp3 server, then you don't need -standalone
14:41
<Hyperbyte>
JuJuBee, run alkisg's commands from above. It tells you all the things both packages install.
14:42
<JuJuBee>
I did not realize that was directied to me, sorry
14:43
Will do
15:06
<meamy>
@hyperbyte just want to let you know, solved my problem after several changes in the nbd-server config restarts, Manuel debuging it work but dont ask me what the problem was i have no idea, it just started to work for no good reason :D
15:08
bye the way is there a extra mailing-list for LTSP-Cluster?
15:10komunista has joined IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk)
15:22
<JuJuBee>
Hyperbyte: when I configure dhcp, in the conditional statement about substring option vendor... the filename "/ltsp/i386/..." I want to create 64 bit images, should I change i386 to x64?
15:23
The directions I am reading I believe are about creating a 32 bit client image
15:25komunista has left IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk, Quit: Leaving.)
15:34adrianorg has left IRC (adrianorg!~adrianorg@187.113.219.235, Read error: Connection reset by peer)
15:34
<||cw>
JuJuBee: I believe the i386 pxe menu can load either, just need to make sure the pxe config files sends the correct kernel
15:36komunista has joined IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk)
15:36
<JuJuBee>
so I should just leave it i386 in the dhcp config?
15:37
<||cw>
that's my understanding. I don't have any x64 clients though
15:40gothaggis has joined IRC (gothaggis!~gothaggis@corp.heavyhammer.com)
15:41
<gothaggis>
hello - printing was working fine for months for our thin clients - now they are not able to print - i checked the CUPS error log and it seems to show 'Returning IPP client-error-not-possible for Print-Job' whenever a client tries to print. Any ideas?
15:50komunista has left IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk, Quit: Leaving.)
15:53
<gothaggis>
not sure how I fixed it exactly - tried restarting cups, didn't work - tried canceling all jobs - didn't work, tried printing a test page and it said there were too many jobs...so i added MaxJobs 0 to the .conf file, restarted cups and now everything seems to print again.
15:57komunista has joined IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk)
16:09
<meamy>
htop
16:09
ubs sorry
16:09
<alkisg>
htop: command not found
16:15
<effenberg>
zsh: correct 'htop' to 'top' [nyae]?
16:15komunista has left IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk, Quit: Leaving.)
16:18sbalneav has left IRC (sbalneav!~sbalneav@mail.legalaid.mb.ca, Ping timeout: 245 seconds)
16:19sbalneav has joined IRC (sbalneav!~sbalneav@mail.legalaid.mb.ca)
16:22komunista has joined IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk)
16:30vmlintu has left IRC (vmlintu!~vmlintu@cs180030.pp.htv.fi, Ping timeout: 264 seconds)
16:39dhani has joined IRC (dhani!b4f8af41@gateway/web/freenode/ip.180.248.175.65)
16:40
<JuJuBee>
I was following a how-to and in the dhcp config it references filename "/ltsp/i386/pxelinux.0" but I have no /ltsp directory on the server. I installed ltsp-server-standalone
16:41
<alkisg>
JuJuBee: that's relative to your tftp dir, /var/lib/tftpboot
16:43Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
16:49mikkel has joined IRC (mikkel!~mikkel@80-71-132-15.u.parknet.dk)
16:51vagrantc has joined IRC (vagrantc!~vagrant@c-98-232-129-196.hsd1.or.comcast.net)
16:51vagrantc has joined IRC (vagrantc!~vagrant@freegeek/vagrantc)
16:51dobber has left IRC (dobber!~dobber@213.169.45.222, Remote host closed the connection)
17:11PhoenixSTF has left IRC (PhoenixSTF!~rudi@78.29.134.164, Quit: Leaving)
17:13
<knipwim>
alkisg: i made a symlink for ip in /usr/bin, but now i'm getting a Connection with epoptes failed:
17:14
An error occurred while connecting: 2: No such file or directory.
17:14JesseC has joined IRC (JesseC!~JesseCWor@wsip-98-175-20-126.br.br.cox.net)
17:15
<JuJuBee>
alkisg: ok, I will look for that...
17:16
<vagrantc>
alkisg, Phantomas: so are the ltsp/ldm branches being translated in launchpad now?
17:16
<Phantomas>
vagrantc: Yes! :)
17:16
<vagrantc>
cool!
17:17
<Phantomas>
Yep, it's easier that way!
17:17* vagrantc wonders why on earth there's a whole separate branch for ALTLinux
17:17
<vagrantc>
not branch ... er, translation system
17:19
<JuJuBee>
alkisg: doesn't appear to be a tftp server installed... I did use apt-get install lltsp-server-standalone but don't see tftp server
17:20
nm, spelling counts :)
17:20
<vagrantc>
hrm. it incorrectly sets the revision dates...
17:20
<alkisg>
knipwim: is the epoptes daemon running? root 2493 0.0 0.4 21744 16380 ? S 14:34 0:07 /usr/bin/python /usr/bin/twistd --pidfile /var/run/epoptes.pid --logfile /var/log/epoptes.log epoptes
17:21
vagrantc: let's check what launchpad pushes for a few days, in case we need to revert things
17:21JesseC has left IRC (JesseC!~JesseCWor@wsip-98-175-20-126.br.br.cox.net, )
17:21
<alkisg>
Maybe even bzr push --overwrite things, if needed... :D
17:21
vagrantc, knipwim: so maybe we shouldn't push code changes before checking the translations
17:22
<vagrantc>
it seems to delete fuzzy strings?
17:22
<alkisg>
Yeah
17:23
vagrantc: maybe they "deprecated" them in favor of their "suggestions" system
17:23
<vagrantc>
does it automatically update the .pot files?
17:23
<alkisg>
Nope, and I don't think we want it to
17:23
<Phantomas>
https://bugs.launchpad.net/launchpad/+bug/312476
17:24
<alkisg>
vagrantc: did you read the mail I sent yesterday?
17:24
<vagrantc>
alkisg: no :)
17:24
<alkisg>
vagrantc: basically we need a "translator-credits" string somewhere in our sources, and to regenerate the .pot files
17:25
So that all .pot files do have such a string
17:27
<knipwim>
alkisg: check, that was it
17:27gvy has left IRC (gvy!~mike@altlinux/developer/mike, Quit: Leaving)
17:27
<knipwim>
concerning the epoptes thing
17:27
<alkisg>
nice
17:31meamy has left IRC (meamy!~hannes@pd95cdee4.dip0.t-ipconnect.de, Quit: meamy)
17:33adrianorg has joined IRC (adrianorg!~adrianorg@187.113.249.84)
17:34meamy has joined IRC (meamy!~hannes@pd95cdee4.dip0.t-ipconnect.de)
17:36
<knipwim>
alkisg: is it normal just to start the server with the whole /usr/bin/python thing ?
17:36
<alkisg>
knipwim: we have an initscript for that...
17:37
$ grep start-stop-daemon /etc/init.d/epoptes
17:37
start-stop-daemon --start --startas /usr/bin/twistd --quiet --oknodo --pidfile "$PIDFILE" --exec "$DAEMON" -- \
17:37
...
17:37
<knipwim>
ah check, you also have the initscript in the package
17:37
not in the sources
17:37
<alkisg>
knipwim: there's a debian/ folder in the sources
17:38
Some things like the initscripts, the /etc/default/epoptes* configuration files are there
17:38
knipwim: ideally, other distros would send us /distro folders to merge upstream
17:39
And e.g. if some is using sysvinit, they could have a symlink to the debian/ initscripts, and the same with configuration files, etc
17:39* alkisg would welcome 1 upstream committer for each distro...
17:40
<vagrantc>
i should really give epoptes some testing again...
17:40
hopefully it still works as well as it did last time i tried, but it's been a while
17:43autoditac has joined IRC (autoditac!~rouven@pd95cdee4.dip0.t-ipconnect.de)
17:48
<autoditac>
hello all! whilst being in a changing & testing cycle, building images for the ltsp fat clients is a pita cause it takes an incredible amount of time. isn't it possible, since these are nbd images, to loop mount the nbd on the terminalserver read-write, chroot into it and do the updates/ installs directly inside the nbd image instead in the main chroot and then rebuilding the image with ltsp-update-image?
17:49
<vagrantc>
autoditac: no, squashfs is read-only filesystem
17:49
autoditac: depending on what distro/version you're running, you could use NFS while you're developing, and then build with NBD when you've finished.
17:49
<alkisg>
Or you could use a btrfs loopback nbd image
17:50
<vagrantc>
autoditac: it's a little slower on the booted environment, but doesn't require rebuilding the image
17:50
or other filesystem types
17:50
<alkisg>
I think I've written a wiki paragraph for that...
17:50vector has left IRC (vector!~vector@host-1-185-230-24.midco.net, Remote host closed the connection)
17:50
<autoditac>
vagrantc, do i understand correctly, that nfs would be ok as well but is a little bit slower than nbd?
17:51
<vagrantc>
autoditac: yup
17:51
autoditac: or using an editable filesystem, like alkisg suggested
17:53vector has joined IRC (vector!~vector@host-1-185-230-24.midco.net)
17:53* vagrantc seems to recall ldm-trunk having to manually specify which languages are supported
17:56gvy has joined IRC (gvy!~mike@altlinux/developer/mike)
17:57vector has left IRC (vector!~vector@host-1-185-230-24.midco.net, Remote host closed the connection)
17:59vector has joined IRC (vector!~vector@host-1-185-230-24.midco.net)
18:04adrianorg_ has joined IRC (adrianorg_!~adrianorg@187.113.218.69)
18:06
<JuJuBee>
I am getting closer. I am now getting PXE-E32: TFTP open timeout error. I am getting a dhcp address fine. When I ls /var/lib/tftpboot/ltsp/amd64 I see the pxelinux.0 but not the nbi.img that the else in dhcp config refers to.
18:07
<alkisg>
JuJuBee: did you run ltsp-build-client --arch amd64 etc etc?
18:07adrianorg has left IRC (adrianorg!~adrianorg@187.113.249.84, Ping timeout: 260 seconds)
18:07
<alkisg>
ls /opt/ltsp/amd64/boot => you should have vmlinuz, nbi.img and the rest there
18:08
<JuJuBee>
alkisg: I ran ltsp-build-client
18:09
<alkisg>
JuJuBee: put the result of ls -lhaR /var/lib/tftpboot to pastebin
18:09
<JuJuBee>
gonna be hard, my test network not online
18:09
<alkisg>
And the result of ls -lhaR /opt/ltsp/amd64/boot/
18:09
Ah
18:10
<JuJuBee>
I have /opt/ltsp/amd64 -> a directory like OS, and /opt/ltsp/images/amd64.img
18:10
<alkisg>
And what do you have in /opt/ltsp/amd64/boot?
18:10
Is nbi.img, vmlinuz etc there?
18:11
<JuJuBee>
No it is not.
18:11
<alkisg>
Then your ltsp-build-client probably failed
18:11
sudo chroot /opt/ltsp/amd64 dpkg -l 'linux*' | grep ^ii
18:12
Is a kernel installed?
18:14
<JuJuBee>
Seems to be linux-image-3.2.0-35 generic
18:14
and linux-image-generic
18:14
<alkisg>
And yet no vmlinuz there? That's weird
18:15Gremble has left IRC (Gremble!~Ben@cpc29-aztw23-2-0-cust144.18-1.cable.virginmedia.com, Quit: I Leave)
18:15
<alkisg>
You should have /opt/ltsp/amd64/boot/vmlinuz-3.2.0-35-generic then
18:15
<JuJuBee>
I do
18:15
<alkisg>
(08:10:34 μμ) alkisg: Is nbi.img, vmlinuz etc there?
18:15
(08:11:17 μμ) JuJuBee: No it is not.
18:15
<JuJuBee>
Did not see the vmlinuz
18:16
sorry
18:16
<alkisg>
Try: sudo chroot /opt/ltsp/amd64 /usr/share/ltsp/update-kernels
18:16
<JuJuBee>
ok
18:18
Done, "skipping yaboot, etherboot, netabootwarp & sparc....
18:19
<alkisg>
Skipping etherboot?
18:20
Looks like you don't have mkelfimage installed
18:20
Anyways, you don't need that
18:20
<vagrantc>
not that that really matters much... :)
18:20
<alkisg>
So, you do have vmlinuz, but not nbi.img, so all look fine in your tftp dir
18:21
(nbi.img is broken anyway, but it was weird that you didn't have mkelfimage)
18:21
!tftp
18:21
<ltsp>
tftp: Here's a page to help you troubleshoot TFTP problems in Ubuntu: https://help.ubuntu.com/community/UbuntuLTSP/Troubleshooting/TFTP
18:23
<alkisg>
ltsp-client-core depends on mkelfimage though
18:24
at least on ubuntu... JuJuBee, which distro are you using?
18:27
<JuJuBee>
kubuntu 12.04
18:28risca has joined IRC (risca!~risca@h240n6-n-a31.ias.bredband.telia.com)
18:28
<JuJuBee>
What how-to would you suggest? I don't mind starting over.
18:29
<alkisg>
ls /opt/ltsp/amd64/usr/sbin/mkelfImage
18:29
Is that file there?
18:30
I haven't seen a how-to for KDE, that overcomes kde/ltsp bugs
18:30sbalneav has left IRC (sbalneav!~sbalneav@mail.legalaid.mb.ca, Remote host closed the connection)
18:30
<alkisg>
Some people do use that, but I know that they're manually configuring some things to make the installation smoother
18:31autoditac has left IRC (autoditac!~rouven@pd95cdee4.dip0.t-ipconnect.de, Read error: Operation timed out)
18:31
<JuJuBee>
not that file is not there, but mkdiskimage is
18:31
and other files
18:31autoditac has joined IRC (autoditac!~rouven@pd95cdee4.dip0.t-ipconnect.de)
18:31
<JuJuBee>
so maybe I will restart again tomorrow, I have a meeting soon and an appointment after.
18:32
<alkisg>
Yeah something's missing in your installation
18:32
<JuJuBee>
I will use ubuntu
18:32
rather than kubuntu
18:32
<alkisg>
That should be much easier :)
18:32
<JuJuBee>
Thanks for the help. I'll be back tomorrow
18:34dhani has left IRC (dhani!b4f8af41@gateway/web/freenode/ip.180.248.175.65, Quit: Page closed)
19:23meamy has left IRC (meamy!~hannes@pd95cdee4.dip0.t-ipconnect.de, Quit: meamy)
19:28autoditac has left IRC (autoditac!~rouven@pd95cdee4.dip0.t-ipconnect.de, Ping timeout: 248 seconds)
19:39autoditac has joined IRC (autoditac!~rouven@brln-4db9fc41.pool.mediaWays.net)
19:39
<vagrantc>
Phantomas, alkisg: i don't know what you mean by '"translator-credits" string'
19:40
just needs the string in the file somewhere? anywhere? what happens with it, or to it?
19:40
<alkisg>
vagrantc: anywhere. It gets in the .pot file
19:40
And then launchpad considers it special,
19:40
<vagrantc>
alkisg: as a header? in comments?
19:40
<alkisg>
and puts the contributors in it
19:40
<Phantomas>
vagrantc: https://translations.launchpad.net/epoptes/trunk/+pots/epoptes/el/1/+translate
19:40
vagrantc: As a normal message string
19:41
<alkisg>
vagrantc: e.g. we could have an "ltsp.sh" file, with just a fake string in it, it doesn't matter where it is in the sources
19:41
Of course it needs to use gettext, to be extracted as a translatable string
19:41
Or _()
19:41
<Phantomas>
You should use it somewhere btw :)
19:41
<alkisg>
It can be inside a shell, .c, or .py file
19:41
Phantomas: why? :)
19:42
<Phantomas>
To show the translators...
19:42
<alkisg>
I don't think we have an appropriate UI for it yet
19:43
<vagrantc>
so it's just a hack that launchpad puts a msgid into the .pot (and .po?) files in order to log the translators somewhere?
19:43
<alkisg>
vagrantc: yup
19:43
it gets it from the .pot, and puts it in the .po's
19:43
*we* put it in the pot
19:45
<vagrantc>
and the simplest way to put it into the .pot is to put: _("translator-credits")
19:45
somewhere into the code?
19:45
<alkisg>
Yup, twice in ltsp, and 3 times in ldm
19:45
<vagrantc>
could put a --credits plugin into ltsp-build-client easily enough
19:45
<alkisg>
So that it goes in each .pot file
19:45
Maybe in some functions.sh file...
19:46
<vagrantc>
ah, yes, in the functions file would be even better, if it picks that up
19:46
how will we get the non-translator credits in there, though? :)
19:48
alkisg: you just wanted my opinions on where to put it, or is it really just me who needs to commit it for some reason?
19:48
<alkisg>
vagrantc: no particular reason, you just were the one that usually updated the .pot files
19:48
And of course your opinion is most valuable :)
19:49
(and if you commit/push your opinion, it'd be even better :P)
19:49* vagrantc kind of balks at all this extra work for the ALTLinux stuff
19:50
<alkisg>
Yeah it felt silly translating those as well...
19:50
<vagrantc>
haven't made a commit in several years, not really even doing LTSP5 (some bizarre 4/5 hybrid?)
19:51
<alkisg>
And we should have cross-distro "config examples"... like we did with lts.conf
19:52
I think we should clean up the translations before sending a mail to ltsp-discuss to ask for more translators
19:53
Maybe even make the usage() ltsp-tool functions translatable first...
20:01autoditac has left IRC (autoditac!~rouven@brln-4db9fc41.pool.mediaWays.net, Ping timeout: 248 seconds)
20:01
<vagrantc>
alkisg, Phantomas: does this look right: http://paste.debian.net/224666/
20:02* vagrantc isn't sure where eval_gettext comes from
20:03
<alkisg>
echo $(eval_gettext "translator-credits") => why not plain eval_gettext "translator-credits" ?
20:04
<vagrantc>
was kind of following the example of all the ltsp-build-client plugins, but sure.
20:05
course i used $() instead of ``
20:05* vagrantc thinks that might be clearer coding-style wise, not sure how much difference it makes otherwise.
20:06
<alkisg>
vagrantc: I think specifically for translations we could use ``
20:06
We don't want nested translations, and it's easier to read
20:06
<vagrantc>
alkisg: one spawns a subshell, the other doesn't?
20:06* vagrantc finds $() far easier to read.
20:06
<alkisg>
vagrantc: I think both do, don't they?
20:06
Example: https://bugs.launchpad.net/ltsp/+bug/1008237
20:07
<vagrantc>
there was some notable difference in the functionality of `` vs. $(), but i forget what it is
20:08* alkisg only remembers the nesting part...
20:09
<vagrantc>
yes, $() is much easier to nest, but there was something more... anyways...
20:09
alkisg: re LP#1008237, i'm not sure `` looks all that much better there.
20:10
<alkisg>
I find it easier to read than parentheses, but no strong feelings either way
20:10
<vagrantc>
overall, i like the idea of a _ helper function
20:10
<alkisg>
(not easier to code/match; just lighter to read the text inside it)
20:11JesseC has joined IRC (JesseC!~JesseCWor@wsip-98-175-20-126.br.br.cox.net)
20:11
<vagrantc>
-b, --base= `_ 'The LTSP base directory. Defaults to "/opt/ltsp".'`
20:11
vs.
20:11
-b, --base= $(_ 'The LTSP base directory. Defaults to "/opt/ltsp".')
20:12
there are so many ' and " that my eyes get lost.
20:12
<alkisg>
Yup. Also, we could get used to ` equals translation
20:12
While $( would be used for any kind of subshell
20:12
True, it's not easier to code it
20:12
<vagrantc>
it wouldn't make me cry, but i prefer $()
20:13
<alkisg>
ok
20:13
vagrantc: another thing is if we prefer shell-style translations vs printf style translations
20:14
echo $(_ "Referencing a $VAR")
20:14
vs
20:14gvy has left IRC (gvy!~mike@altlinux/developer/mike, Quit: Leaving)
20:14
<alkisg>
echo $(_ "Referencing a %s" "$VAR") ...or something
20:15
<vagrantc>
the code you proposed seems to support either?
20:15
the latter is a little cleaner for translators, who might accidentally translate $VAR otherwise?
20:15
<alkisg>
I think it only supports the second case
20:16
But there's a gettext.sh infrastructure which we ignore if we implement it as I propose in the bug report
20:16
<vagrantc>
so we're deviating from a "norm" ?
20:17
<alkisg>
Maybe... I'm not sure I like the "shell norm" though, I do prefer the "all other languages norm"
20:17
<vagrantc>
heh
20:18
<alkisg>
And since we're using multiple languages in the same .pot... it's a bit confusing to have multiple norms
20:18
(languages = shell, c, python)
20:19
<vagrantc>
alkisg: so the $(_ "blah blah %s" "$VAR") is more consinstant across languages?
20:20
<alkisg>
Yes. But I do need to check again that "$VAR" is not considered as a translatable string by xgettext...
20:20
I.e. that it only takes the first $(_ parameter
20:32
<vagrantc>
alkisg: it seems to only take the first parameter.
20:32
alkisg: at least, with xgettext -k_
20:35
<alkisg>
Very nice... so /me proposes that one
20:39
<vagrantc>
sounds good to me
20:42* vagrantc notes that ldm's .pot file generation is totally different...
20:42
<alkisg>
vagrantc: maybe we can stuff the translator-credits string inside ldm.c
20:43
And inside some rc.d script, for ldm-rc.pot
20:46
<Phantomas>
(If a string contains more than 1 variables it is better to leave them with their names in the string, as in some languages it may be needed to change their order)
20:47
<vagrantc>
looks like i could combine the ltsp-cluster-info and rc.d into a single .pot
20:48
not sure how to combine the main ldm.pot
20:48
ah yes, word ordering may matter...
20:50komunista has left IRC (komunista!~slavko@adsl-195-168-235-217.dynamic.nextra.sk, Quit: Leaving.)
20:51
<Phantomas>
vagrantc: Why do you want to combine them?
20:53
<vagrantc>
having to translate multiple .po files, which may even have shared strings, seems redundant
20:53
seems like one .pot per project would be nice
20:54
but maybe that's more messy than useful.
20:56
ugh. regenerating the ldm.pot is not as straightforward as the ltsp.pot
21:00
<alkisg>
coreutils' printf doesn't support argument swapping: http://lists.gnu.org/archive/html/bug-coreutils/2008-12/msg00207.html
21:04leio_ has joined IRC (leio_!~leio@gentoo/developer/leio)
21:06leio has left IRC (leio!~leio@gentoo/developer/leio, Ping timeout: 276 seconds)
21:07
<vagrantc>
alkisg: that's an article from 4 years ago ... and doesn't look to be the same sort of things we're proposing to do?
21:08
ah, but to reorder the arguments...
21:09
<alkisg>
Yup, it reports that at least in 2008 it wasn't supported,
21:09
and some testing here confirms that
21:10
Also neither dash'es nor bash'es internal implemention of print seem to support argument swapping
21:10
<vagrantc>
we can move the %s strings, but they can't be swapped?
21:10
<alkisg>
Right
21:11
I don't think that's too much of a deal though
21:11
<vagrantc>
printf "foo %s bar" "baz" ; printf "foo bar %s"
21:11
<Phantomas>
The simplest way is to include the variable names inside your strings and then you would translate "Hi $name $surname" to "Hi $surname $name"
21:11
<vagrantc>
printf "foo %s bar" "baz" ; printf "foo bar %s" "baz"
21:11
<alkisg>
It should be very rare, and even then, translators should be able to translate withour reordering
21:13
<vagrantc>
it makes it possible for "foo %s" to be translated once, even with an arbitrary number of arguments.
21:13
i.e. "error: %s" "$errormessage"
21:13
i.e. "error: %s" "$another_errormessage"
21:13
would only require a single translation
21:14
of course, if you wanted translated error mesages, that wouldn't work.
21:15* vagrantc marks the days till newer ltsp hits wheezy...
21:15
<alkisg>
Btw, knipwim, I don't think that's translatable: ./server/share/ltsp/plugins/ltsp-build-client/Gentoo/002-set-stage-file: printf "`eval_gettext "Failed finding stage in: ${STAGE_URI}"`\n"
21:16
${STAGE_URI} will be substituted by shell before eval_gettext gets a chance to see it
21:16
So gettext won't be able to substitute the variable, it'll see a string that contains the substituted variable
21:17
<vagrantc>
printf "`eval_gettext "Failed finding stage in: "`${STAGE_URI}\n" ?
21:17
er, less one "
21:17* vagrantc likes $()
21:17
<knipwim>
hmm
21:17
never noticed that before, it probably never happened :)
21:18
<vagrantc>
alkisg: so, with translator-credits, and no en.po ... what happens?
21:19* vagrantc wonders how to get existing translators to show up in translator-credits
21:19
<alkisg>
Meh, let's notify them to work directly on launchpad :P
21:19
<vagrantc>
there's en_ca, en_gb, en_au, but no en_us.
21:19
<alkisg>
vagrantc: I don't think en.po makes any difference
21:20
<vagrantc>
or just plain en
21:20* vagrantc doesn't understand how the translator-credits gets substituted.
21:20
<alkisg>
As I understand it, launchpad takes translator-credits from the .pot
21:20
Then it replaces its (empty) contents with each language translators, as they appear in launchpad internal database
21:20
<vagrantc>
i.e. the translator_credits function would just output "translator-credits" ?
21:21
<alkisg>
If no translator exists, I imagine the translator-credits string remains empty
21:21
<vagrantc>
translator-credits just gets replaced on a per-language basis?
21:21
<alkisg>
Yup
21:21
<vagrantc>
i.e. it wouldn't list all the translators in credits, just the currently affected ones?
21:22
<alkisg>
Once for each .po that launchpad generates
21:22
Right
21:22
<vagrantc>
er, the translations currently in effect.
21:22
<alkisg>
So in epoptes, users with different languages see different translators in about dialog
21:22
<vagrantc>
got it.
21:23
so in order to generate a list of all translators, we'd have to write a script that sets all the locales and spits them out...
21:23
<alkisg>
Or parse the .po files
21:23
<vagrantc>
right
21:24
<alkisg>
./server/share/ltsp/plugins/ltsp-build-client/common/001-set-arch: printf "`eval_gettext "ERROR: --arch %s called on unsupported arch: %s"`", i386 $(uname -m)
21:24
It appears we're already using the %s method
21:24
Even without the "xgettext only reads the first argument" bit
21:27
<otwieracz>
Hello.
21:29
I'm setting up LTSP server on Gentoo.
21:30
And when I'm booting clients I'm getting error:
21:31
mounting /dev on /newroot/dev failed: no such file or directory
21:31
And previously, after >> Mounting root
21:31
udhcpcd: SIOCGIFINDEX: No such device
21:32
<knipwim>
what initramfs builder?
21:32
otwieracz: ^^^
21:32
<otwieracz>
Hm.
21:32
Probably kicktoo.
21:32
If it is it.
21:32
Or, genkernel.
21:33
Probably its kernel does not have selected option "Root on NFS".
21:34
(or something like this)
21:34
<knipwim>
could be
21:34
<otwieracz>
Can I somehow easily recompile client's kernel?
21:35
<knipwim>
sure, ltsp-chroot in the chroot and compile the client kernel there
21:35
<otwieracz>
Hmm.
21:36
But then what if im chrooting from x86_64 into i686?
21:36
Then genkernel is trying to create kernel for x86_64.
21:37
<knipwim>
you can force 32bit in the chrooted environment by prefixing the make commands with linux32
21:37
like linux32 make menuconfig
21:37
<otwieracz>
ok.
21:39
<knipwim>
if the NFS boot doesn't work, you could try the NBD boot option
21:40
http://wiki.gentoo.org/wiki/LTSP#NBD_Boot
21:46
<otwieracz>
What's the correct way of customizing chroot environment?
21:49
<knipwim>
what do you mean, before or after the install?
21:50mikkel has left IRC (mikkel!~mikkel@80-71-132-15.u.parknet.dk, Quit: Leaving)
21:53alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
22:05
<knipwim>
otwieracz: going to bed now
22:05
<otwieracz>
oh.
22:05
<knipwim>
but i'll probably be here tomorrow
22:05
<otwieracz>
After install.
22:06
<knipwim>
use the ltsp-chroot command
22:06
<otwieracz>
Sorry, I've missed the notification.
22:06
<knipwim>
do the customizations
22:06
and exit
22:06
ltsp-chroot --arch=i686 --mount-all
22:07
and you can emerge what you like
22:28Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Ping timeout: 252 seconds)
22:45Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
22:48
<otwieracz>
It doesn't helped.
22:48
(Still the same errors)
22:49
And /dev/nfs is working all the time.
22:49
beucase I have different errors with /dev/nfs and /dev/foobar
22:54xet7 has left IRC (xet7!~xet7@a91-156-233-189.elisa-laajakaista.fi, Quit: Lähdössä)
23:00dead_inside has left IRC (dead_inside!~dead_insi@76.75.3.174, Quit: Computer has gone to sleep.)
23:14andygraybeal_ has joined IRC (andygraybeal_!~andy@h44.218.213.151.dynamic.ip.windstream.net)
23:23risca has left IRC (risca!~risca@h240n6-n-a31.ias.bredband.telia.com, Quit: Lämnar)
23:25DIoX|DaZ has left IRC (DIoX|DaZ!~KaKa@server.civicclub.lt, Ping timeout: 256 seconds)
23:25DIoX|DaZ has joined IRC (DIoX|DaZ!~KaKa@server.civicclub.lt)
23:36bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Ping timeout: 240 seconds)
23:46vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Quit: leaving)