without just literally exporting whatever desktop i happen to be logged into (or not) at the moment
that's not really LTSP ... that'd be VNC, NX or RDP.
LTSP can be used to log into servers using those protocols.
|00:05||* vagrantc just recently discovered xprintidle!|
xprintidle is so useful for detecting idle sessions and getting them to log out.
well, it shows you how long a session has been idle, and then you can write scripts that do things based on that.
maybe a "Log out idle users" option in epoptes would be useful...
but I'm not sure
Phantomas: not a terrible idea...
although you'd want it to be configurable somehow
vagrantc: VNC itself doesn't seem to support anything too sophisticated.. it seems something would have to wrap around it
vagrantc: It's not difficult, we could have an idleMinutes option in our config file
vagrantc: btw, I also wanted to add support for scheduled operations (shut down/WoL after 5 pm everyday e.g.) but alkisg didn't like the idea :P
Phantomas: ah well :)
|00:23||msignor has left IRC (email@example.com, Ping timeout: 240 seconds)|
|00:28||vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Quit: leaving)|
|00:48||Phantomas has left IRC (Phantomas!~Phantomas@unaffiliated/phantomas)|
I'm trying to setup a fat-client under mythbuntu 12.04 and the client has recently stopped booting with the error 'PXE-E32: TFTP open timeout'
when I search for this error all the results seem very old and out of date
If I restart the tftpd-hpa it starts working
But it doesn't work when I reboot the server
even though tftp-hpa is running when the server is started
Do I have to somehow tell tftp-hpa to wait until everything is loaded before starting up?
|00:57||msignor has joined IRC (firstname.lastname@example.org)|
|01:09||Phantomas has joined IRC (Phantomas!~Phantomas@unaffiliated/phantomas)|
I've got a bit of a hack to get around this problem, I just create a /etc/init.d/tftpd-hpa2: http://pastie.org/3998858
it's very hacky and I would like to fix this problem up.
|01:21||Phantomas has left IRC (Phantomas!~Phantomas@unaffiliated/phantomas, Quit: Leaving.)|
|01:34||rhorstkoetter has joined IRC (rhorstkoetter!~rhorstkoe@opensuse/member/rhorstkoetter)|
|01:46||rhorstkoetter has left IRC (rhorstkoetter!~rhorstkoe@opensuse/member/rhorstkoetter, Quit: Leaving.)|
|01:52||andygraybeal has joined IRC (email@example.com)|
|01:56||msignor has left IRC (firstname.lastname@example.org, Ping timeout: 245 seconds)|
|02:15||andygraybeal has left IRC (email@example.com, Ping timeout: 246 seconds)|
|02:23||Parker955 is now known as Parker955_Away|
|02:24||[GuS] has joined IRC ([GuS]!~firstname.lastname@example.org)|
|02:24||[GuS] has joined IRC ([GuS]!~gustavo@unaffiliated/gus/x-663402)|
|02:41||dgeary2 has left IRC (email@example.com, Quit: ĝis la)|
|03:05||vagrantc has joined IRC (firstname.lastname@example.org)|
|03:05||vagrantc has joined IRC (vagrantc!~vagrant@freegeek/vagrantc)|
|03:13||adrianorg__ has left IRC (email@example.com, Ping timeout: 260 seconds)|
|03:49||vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Ping timeout: 260 seconds)|
|03:49||vagrantc has joined IRC (vagrantc!~vagrant@freegeek/vagrantc)|
|03:55||sha_ has joined IRC (firstname.lastname@example.org)|
|03:57||sha has left IRC (email@example.com, Read error: Operation timed out)|
|04:02||highvoltage has left IRC (highvoltage!~highvolta@ubuntu/member/highvoltage, Ping timeout: 260 seconds)|
|04:10||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|04:22||monteslu_ has joined IRC (firstname.lastname@example.org)|
|04:26||monteslu has left IRC (email@example.com, Ping timeout: 260 seconds)|
vagrantc: hi, does that seem good enough to put in ltsp-update-image? http://paste.debian.net/172064/
I don't think EXCLUDED_MOUNTS will be used much, so I don't think it should be an ltsp-update-image option, but just an ltsp-update-image.conf variable...
|05:02||jimjimovich has joined IRC (firstname.lastname@example.org)|
|05:09||shogunx has left IRC (email@example.com, Ping timeout: 246 seconds)|
vagrantc: whenever you got some time, test if it includes /boot now for you
Btw here's my packaging tree with the 2 extra commits I needed to get it building on launchpad: https://code.launchpad.net/~alkisg/+junk/ltsp
|05:24||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)|
|05:38||alkisg has joined IRC (firstname.lastname@example.org)|
|05:38||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
knipwim: is the test wiki going to be moved to the new location, or a new one will be installed? Can I start writing content there?
|05:39||andygraybeal has joined IRC (email@example.com)|
|05:39||* alkisg tried mediawiki a bit yesterday and liked its template syntax very much|
|05:40||* Hyperbyte high fives alkisg|
|05:40||* alkisg high, low, and side-fives Hyperbyte|
The fckeditor for mediawiki is fine as well
Ah I think its new name is just ckeditor
That sounds... kinda... ehm... something...
I think that's ^ why they changed the name :P
So when do we have a site meeting again?
I honestly am not sure.
We could get together tonight.
But I haven't made progress with success story form... it's been hectic weeks, and I've come down with a quite nasty headache.
|05:58||andygraybeal has left IRC (firstname.lastname@example.org, Ping timeout: 246 seconds)|
|06:00||* vagrantc wavves|
i'm only partially here... my attention is mostly elsewhere
mhm... does anyone know how Gnome determines which applications to send a terminate signal on logout?
I have an windowed application (let's call it A) which is started during Gnome login, and that application in turn launches a daemon (B) which does some of the behind-the-scenes work for the application.
On Gnome logout, A is terminated, but B keeps running in the background.
I know Gnome watchdog can probably 'solve' this problem for me, but I'd rather get it fixed permanently for everyone. So I'm talking to the developers, and they tell me they've now configured the daemon to handle SIGHUP, SIGTERM and SIGINT.
I'm wondering - is this enough? Will Gnome send this program a signal at all, since it's windowless and it wasn't Gnome that started it?
Hyperbyte: apps are required to tell gnome that they want to block logout, here's a starting point for your google adventure: http://people.gnome.org/~mccann/gnome-session/docs/gnome-session.html#org.gnome.SessionManager.Inhibitor
|06:21||jimjimovich has left IRC (email@example.com, Ping timeout: 265 seconds)|
|06:24||jimjimovich has joined IRC (firstname.lastname@example.org)|
does anyone have a link to instructions for building a fat client image with Ubuntu 12.04? I can't get the ones on the wiki to work.
!ltsp-pnp | echo jimjimovich
jimjimovich 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
alkisg, mmm... so the best solution here would be to make application A (which is registered in the session) send a terminate signal to the UI-independant daemon B.
Hyperbyte: yes, if the deamon doesn't know how to register itself in the session, someone should send it a terminate signal, might as well be app A
Hyperbyte: actually I read through your question too fast previously, so you weren't looking for inhibitors but for session registration, but the same ^ conclusion applies
I.e. you don't want the daemon to block logout but just to receive a termination signal, but again for this to occur the daemon must know how to register itself in the session
E.g. the user dbus isn't a graphical app but does register itself in the session
(while evolution-data-center doesn't, and remains after logout)
Session registration is not gnome-specific, session inhibiting is DE specific
I've sent my mail to the developers now anyways, so... :)
|07:32||F-GT has joined IRC (F-GTemail@example.com)|
alkisg: so ... ltsp-client-core installed makes my machine unbootable :(
vagrantc: what's your cmdline like?
Does it contain "ltsp" anywhere?
vagrantc: err basically let's start with "how far it goes", it doesn't run any init-ltsp.d bits , right?
|07:36||* alkisg fires up his debian/lxde vm...|
i didn't troubleshoot it too far yet... but the symptop was it wasn't detectinng lvm in the initramfs ... i managed to boot with an older kernel, purged ltsp-client-core, and it rebooted fine...
i don't see how it could be interfering with that...
vagrantc: cat /proc/cmdline then, what's the output, maybe we'll get some idea what part of the initramfs hangs with it
was ldm changed so one can't choose a machine if it's load balanced?
alkisg: no "ltsp" on my boot commandline.
vagrantc: I'm firing up my debian sid vm to test, you don't have a repo somewhere right? I'll use the ltsp-pnp ppa...
alkisg: i could push my current packages somewhere ... note that this is before any recent changes
vagrantc: I think that'd be better, I can copy a few files to update them then
well, before the most recent changes.
(hopefully you mean "a few days/weeks ago", not "a few months ago" :))
|07:42||Steve_The_Pirate has joined IRC (Steve_The_Pirate!~Gary@cpc16-haye17-2-0-cust169.haye.cable.virginmedia.com)|
alkisg: up to 2272
vagrantc: nah leave it I'll try with the PPA
|07:45||jimjimovich has left IRC (firstname.lastname@example.org)|
apt-get dist-upgrade => "453 MB need to be downloaded"... wow I only haven't updated it for a couple of weeks :-/
it's been busy
ok, so i'm going to install ltsp-client-core again, and see if that breaks it...
alkisg: my guess ... it has MODULES=netboot
Why would additional modules break anything?
alkisg: for the initramfs ... which means no disks.
alkisg: it means "only netboot" modules.
Missing lvm module might be the reason, yeah
no disks show up
But I wonder why _my_ system boots properly then
so, that's an RC bug that i think is currently present in wheezy.
alkisg: ubuntu or debian?
Ubuntu, let me check "MODULES"...
The debian vm hasn't updated yet
The same, MODULES=netboot
ubuntu might behave differently with those settings
yup, that fixes it
Meh from ifupdown changelog: ...calls the scripts with IFACE=--all
I bet that'll break a lot of if-up scripts...
so, from initramfs.conf manpage: "netboot adds the base and network modules, but skips block devices."
Yeah I wondered about that in the past too
But since it caused no problems on Ubuntu, I ignored it
I think "most" now does include network devices though, does it not?
i don't know if "most" includes network drivers...
Well initramfs-tools does contain a function to install all network modules, maybe we can use that instead
oh, that would be useful.
well, i figured out that dillema for the moment.
more fun another day.
|08:08||* vagrantc waves|
|08:08||vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Quit: leaving)|
...which is already present in hooks/nbd, so we can completely remove conf.d/ltsp at least when we use nbd
|08:21||bobby_C has joined IRC (bobby_Cemail@example.com)|
On a related note, we shouldn't force_load the modules but call manual_add_modules instead, so that we save RAM if they're not used (e.g. we don't need both overlayfs and aufs loaded, and noone of them for local boots)
|08:46||dobber has joined IRC (firstname.lastname@example.org)|
Pushed, that should save some more RAM...
|09:09||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)|
|09:24||ogra_ has left IRC (email@example.com, Excess Flood)|
|09:24||ogra_ has joined IRC (firstname.lastname@example.org)|
|09:35||shogunx has joined IRC (email@example.com)|
|10:17||alexqwesa has left IRC (firstname.lastname@example.org, Quit: Хана X'ам !!!)|
|10:33||Trixboxer has joined IRC (Trixboxer!~Trixboxer@126.96.36.199)|
|10:44||rhorstkoetter has joined IRC (rhorstkoetter!~rhorstkoe@opensuse/member/rhorstkoetter)|
|10:45||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|10:51||alexqwesa has joined IRC (email@example.com)|
|11:03||komunista has joined IRC (firstname.lastname@example.org)|
|11:12||alexqwesa has left IRC (email@example.com, Ping timeout: 252 seconds)|
|11:13||alexqwesa has joined IRC (firstname.lastname@example.org)|
|11:14||bieb has joined IRC (email@example.com)|
alkisg: have you been using local boot ltsp clients yet? I was just doing some testing with grub2 the other day to get it to boot directly from the chroot .img file stored on local hard drive..
Turned out that grub2 in ubuntu 12.04 is too old for that, but newest beta worked..
|11:38||[GuS] has left IRC ([GuS]!~gustavo@unaffiliated/gus/x-663402, Quit: Konversation terminated!)|
vmlintu, do you have something special in image (like wlan settings) or why not just use ipxe in grub?
I was just testing how to get the whole fat client image installed locally..
to reduce network requirements
Ah, I see the point. I use thin like that in my laptop.
have you copied the chroot image on the hard drive?
Yes, and removed any nbd related from initrd.
So it really loads from hdd, not over network.
where do you have vmlinuz and initrd? on the hard drive?
Yes, like usual local system.
As well lts.conf.
so you have /boot directory on the boot partition?
In same as other chroot. Just dropped on hdd, removed nbd scripts from initrd, rebuilded it and installed grub.
do you have i386.img file on the hard drive as a single file?
or files copied to a partition?
No, as /dev/sda1
To a partition.
ok.. what I was testing now was to have the image file as-is in single file and boot directly from it
there was no need to remove nbd scripts either
I think there's no any if it's a fat.
I was just wondering if others have been testing similar setups and what the pros and cons of them would be..
|11:50||Parker955_Away is now known as Parker955|
Mind me. Ofcourse there is.
Because it assumes it still booting over network.
|11:51||mmetzger has left IRC (firstname.lastname@example.org, Ping timeout: 260 seconds)|
Then it makes a nbd connection and continues loading from there.
if you remove the nbd stuff from kernel command line, it doesn't try that
Isn't there any defaults?
Haven't looked much how 12.04 is changed.
|11:53||mmetzger has joined IRC (email@example.com)|
If it doesn't need network during boot, you should able to boot up to ldm without network.
if it gets a root partition mounted properly, it seems to be happy
it still runs dhclient, so it needs network
But dhclient is not in initrd, so it can timeout.
And doesn't affect rest of booting if doing locally.
vmlintu: grub2 booted directly from the .img?!
vmlintu: I think the easiest way would be to netboot the kernel+initrd, and then mount the nbd file from the local disk
|12:00||komunista has left IRC (firstname.lastname@example.org, Quit: Leaving.)|
vmlintu: also, do try with very recent ltsp versions, there have been optimizations about that
|12:00||Parker955 is now known as Parker955_Away|
|12:00||toscalix has joined IRC (email@example.com)|
alkisg: yes, grub2 installed on mbr loads the kernel from the .img
it even finds the .img file from the partition automatically, so the name can change
vmlintu: so what, you just drop the .nbd file in an ext4 partition in /, and it finds it?
But, don't you need a /boot folder for grub anyway?
yes.. or actually I was testing with ext3
/boot needs to be there, but it never needs to be updated as there's no kernel or initrd there
So you have a local installtion, you run update-grub, and it checks every single file in / in case it's a loopback image, without you specifying anything? cool!
no need to run update-grub
grub actually finds the file dynamically
so you just drop a new file and boot, no need to do anything else
vmlintu: you can just install ltsp-client on the local installation, no need to use nbd
...and put a "boot ltsp" additional grub entry
|12:07||komunista has joined IRC (firstname.lastname@example.org)|
alkisg: now I lost you - what do you refer to with installing ltsp-client?
vmlintu: in that local disk of yours, do you have an OS where grub is installed?
So you installed it manually?
From a live cd? Nice. What I'm saying is that you can have local installations that use the ltsp methods to boo,
offering central settings with lts.conf and the other usual ltsp benefits (authentications, /home)
without involving nbd at all. If you want, you can rsync the chroot too, sure
I booted with pxe, partitioned the hard drive, installed grub with grub-install and copied custom grub.cfg and finally copied the i386.img file to the partition as a single file
so the partition has /boot directory with some grub files and a single i386.img file
It's amazing that grub looks inside that file
No kernel in /boot, right?
you need grub 2.00 for that.. 1.99 in ubuntu doesn't work
no kernel in /boot
And it's also weird that it has support for squashfs images
I wonder if it can do that with ntfs too, i.e. to do the same thing from a windows setup
I think there was an ntfs module in grub2
you can actually write all kinds of things in the grub programming language
...then a windows application that rsyncs nbd images is in order :P
Nah just kidding
You can do that while the client boots, rsync the nbd image
|12:25||toscalix has left IRC (email@example.com, Read error: Connection reset by peer)|
|12:33||mmetzger has left IRC (firstname.lastname@example.org, Ping timeout: 245 seconds)|
|12:35||mmetzger has joined IRC (email@example.com)|
|12:36||artista_frustrad has joined IRC (firstname.lastname@example.org)|
|12:37||[GuS] has joined IRC ([GuS]!~MysT@213-117-16-190.fibertel.com.ar)|
|12:37||[GuS] has joined IRC ([GuS]!~MysT@unaffiliated/gus/x-663402)|
|12:38||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)|
|12:41||rhorstkoetter has left IRC (rhorstkoetter!~rhorstkoe@opensuse/member/rhorstkoetter, Quit: Leaving.)|
i'm saying hi, but i really don't have time
15 minutes to finish something
and meeting time after that
Well, i will ask you later
but i'm here tonight, in like 5 hours
Is just that in new ltsp installations, the initramfs is not being created
i dont know if that is normal
|12:51||mealstrom has joined IRC (email@example.com)|
|12:51||rhorstkoetter has joined IRC (rhorstkoetter!~rhorstkoe@opensuse/member/rhorstkoetter)|
hi, has anyone installed nvidia-173 driver at ubuntu 12.04 thin client? any workaround ?
|12:59||adrianorg__ has joined IRC (firstname.lastname@example.org)|
What it doesn't do?
|13:09||lefteris_nik_ has joined IRC (email@example.com)|
mealstrom, tried nvidia-current ?
Meaning, 173 are quite old, because current are 295
muppis: i've got gf5200fx, it isnt supported at nvidia-current drivers
Let me fix it.. Hold a second.
64 or 32?
|13:16||rhorstkoetter has left IRC (rhorstkoetter!~rhorstkoe@opensuse/member/rhorstkoetter, Quit: Leaving.)|
mealstrom, http://apt.muppis.net/nvidia-173_173.14.30-0ubuntu11_i386.modfied.deb or http://apt.muppis.net/nvidia-173_173.14.30-0ubuntu11_am64.modfied.deb I removed that depency because it is unneeded in precise.
|13:22||dead_inside has joined IRC (firstname.lastname@example.org)|
|13:23||toscalix has joined IRC (email@example.com)|
|13:45||highvoltage has joined IRC (firstname.lastname@example.org)|
|13:48||alkisg has joined IRC (email@example.com)|
|13:48||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|13:53||JesseC has joined IRC (JesseC!~JesseCWor@wsip-98-175-20-126.br.br.cox.net)|
|13:59||Phantomas has joined IRC (Phantomas!~Phantomas@unaffiliated/phantomas)|
|14:21||primeministerp has left IRC (firstname.lastname@example.org, Remote host closed the connection)|
|14:26||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Read error: Connection reset by peer)|
|14:28||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|14:28||toscalix has left IRC (email@example.com, Ping timeout: 250 seconds)|
|14:41||darkpixel_ has joined IRC (darkpixel_!~darkpixel@curetheitch/staff/darkpixel)|
alkisg: gnome-fallback: To select gnome-fallback as your default session, put this line in your lts.conf: LDM_SESSION="gnome-session --session=gnome-fallback"
|14:49||mgariepy_ has joined IRC (mgariepy_!mgariepy@nat/revolutionlinux/x-mjcvomkdfqhcvhja)|
alkisg: have there been changes to lts.conf loading since 12.04?
|14:52||mgariepy has left IRC (mgariepy!mgariepy@ubuntu/member/mgariepy, Ping timeout: 245 seconds)|
we've been using the ltsp-cluster lts.conf loading script on 10.04 with puavo ldap backed server part, but that breaks when image is installed locally
I've never even seen ltsp-cluster, no idea about it
we've been just using the script that loads lts.conf over http
I was even thinking of putting lts.conf info in TXT records in DNS..
IMHO lts.conf should be a bit private, e.g. to be able to send LDM_USERNAME and LDM_PASSWORD to a specific client without the other clients knowing about it
no need to run special server process to serve the configurations
The tftp daemon is running anyway, it's not a special process
But in the future I'd like to merge the whole ltsp configuration with lts.conf and ldminfod in a single daemon
we have no actual lts.conf file anywhere as it is created from LDAP dynamically
it's the same concept as in ltsp-cluster
there should be no need for LDM_PASSWORD if you configure PAM on the server to login the user without checking password
|15:02||NeonLicht has left IRC (NeonLicht!~NeonLicht@darwin.ugr.es, Quit: Lost terminal)|
That would allow anyone on the local network to login as that user though
I'd prefer for lts.conf to be transformed into a set of shell files under /etc/ltsp/lts.conf.d
And "expand" ldminfod to serve those along with the rest of the server info
So the scripts would be run when the client requests for the information?
So you'd be able to run code in the server to dynamically decide lts.conf settings per client
That'd allow nice LDAP integration also
Sure, and load balancing based on arbitrary criteria, and lots of other things
With libpam_exec you can create a script that checks that the connection comes from the right client before letting the user login
LDM_PASSWORD can be read by anyone anyway..
It depends on how you're sending it
is there a way to verify the client?
It's the "secure boot" problem...
Normally you'd want some key stored on the client itself
Maybe one could use some motherboard info for that
maybe some locally installed boot component that checks certificates or something like that
instead of using normal PXE, it would run some "trusted pxe"
Right, I think ipxe has some codes about secure boot
I guess I'll dig next in to replacing ldm with lightdm
I really haven't gone through the internals of ldm yet to know what'd need to be done for that
vmlintu: first sbalneav should finish the libpam_sshauth pam module
I'll be using pam_krb5 for authentication
Once that is done, the chroot should be configured to use it, so any pam-aware DM can use ssh as the authentication mechanism
Then, ldminfod should be queried, and /usr/share/xsession/desktop files should be generated for lightdm to see them
Those would appear in the "select session" DM menu
And then the ssh channel generated by libpam_sshauth would be reused to launch the session on the server
It's not a matter of authentication only, you need to have ssh access to the server to launch a session, does pam_krb5 allow that?
once you have the kerberos ticket, you can launch ssh connection without anything being asked
sshd supports GSSAPI
the same ticket can be used to mount home directory over nfs4+krb5
vmlintu: does pam_krb5 also provide nss services?
I.e. getent passwd user locally on the client should return an uid/gid...
nss_ldapd should handle that
it supports also GSSAPI for authentication
Right but ldap isn't easy for everyone, that's why sbalneav tries to handle that with plain ssh
We are actually using those for fat clients with 10.04 already
I too asked him if it would be easier to just use ldap instead, he said no
Well if you manage to somehow automate all that I'm sure it would be welcomed upstream in ltsp
If it works with pam-krb5 + nss-ldapd, it should work with pam-ssh + nss-ssh with pam/nss configuration changes only
Removing (or at least making optional) LDM is one of the goals of LTSP 6 I think
|15:34||NeonLicht has joined IRC (NeonLicht!~NeonLicht@darwin.ugr.es)|
|15:37||dobber has left IRC (firstname.lastname@example.org, Remote host closed the connection)|
School will close on Saturday, so I get to do some hacking..
can i ask you about rcp.idmapd for nfs4 home dir mounting, how it works? it maps email@example.com(client) to firstname.lastname@example.org(server) with different uid/gid ? am i right.
|15:55||mealstrom has left IRC (email@example.com)|
|15:59||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)|
my lts.conf in /var/lib/tftpboot/ltsp/i386/ only has the line LDM_SESSION="gnome-session --session=gnome-fallback" but when the students login they get the unity desktop
|16:04||slackish has left IRC (firstname.lastname@example.org, Ping timeout: 245 seconds)|
|16:05||toscalix has joined IRC (email@example.com)|
bieb: do you have [default] line?
that should be above the LDM line?
yes, it defines that the following settings should be applied to all thin clients
|16:13||mealstrom has joined IRC (firstname.lastname@example.org)|
vmlintu: we have a print payment system for student printing.. there is a "client" that needs to run for each user, how can I set the .sh to run for anyone that logs in?
mealstrom: I haven't used a setup where uid/gids would be different, but that's how it should work, yes
mealstrom: with nfs4/krb5[ip] nfs server is the one doing all acl checks instead of the client (at least this is my experience with it)
in the lts.conf [default] isnt case sensitive is it?
I've always used lowercase
|16:18||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|16:25||F-GT has left IRC (F-GTemail@example.com, Ping timeout: 244 seconds)|
|16:26||slackish has joined IRC (firstname.lastname@example.org)|
|16:44||knipwim has left IRC (email@example.com, Quit: leaving)|
|16:47||knipwim has joined IRC (firstname.lastname@example.org)|
|17:15||lefteris_nik_ has left IRC (email@example.com, Quit: Ex-Chat)|
|17:16||vagrantc has joined IRC (vagrantc!~vagrant@75-150-46-245-Oregon.hfc.comcastbusiness.net)|
|17:16||vagrantc has joined IRC (vagrantc!~vagrant@freegeek/vagrantc)|
|17:16||GodFather has joined IRC (GodFatherfirstname.lastname@example.org)|
Time left to build ltsp on launchpad... 16 minutes... 21 minutes... 24 minutes... shouldn't that be a countdown instead? :(
|17:25||jimjimovich has joined IRC (email@example.com)|
how do I get dhcpd and tftp to start at boot?
jimjimovich: they should start automatically, if not file bugs on their ubuntu packages on launchpad
vagrantc: I presumably fixed any remaining bugs, if you got any more throw them at !pad
!learn pad as http://pad.ubuntu-uk.org/ltsp
alkisg: The operation succeeded.
alkisg, if you look at it from the inside of the datacenter it counts down, its all a matter of perspective :P
Meh the stupid theory of relativity gave excuses even to datacenters now :P
doesn't that only really distort time at extremely high speeds?
Apparently launchpad approaches the speed of light
well, it has passed the 1mio bugs line ...
you never know what are the sideefects of this
vagrantc: we started talking in the meeting room, if you got time...
usually time to lurk
|17:50||_UsUrPeR_ has joined IRC (_UsUrPeR_firstname.lastname@example.org)|
|17:50||* _UsUrPeR_ tips his hat|
Hi _UsUrPeR_, long time no see, how are you?
hey alkis. Doing well thanks
wow, just had a desktop thin client detect a non-existent laptop secondary screen. weird
|18:06||toscalix has left IRC (email@example.com, Remote host closed the connection)|
jimjimovich: yeah, that happens.
some hardware is ... funny.
vagrantc: is there a way to disable that?
you can use XRANDR_* options in lts.conf i think, but i don't really know the exact info off the top of my head
XRANDR_OUTPUT_0="LVDS-1 --off" or similar, I guess
there was a pxelinux.cfg/DEFAULT setting too
exact nomenclature slips my mind.
video:lvds-1:off or something
and of course, that stops working when ltsp-update-kernels is executed
Not if you put BOOTPROMPT_OPTS in update-kernels.conf
alkisg: i should be able to test and prepare an LTSP upload tomorrow ...
alkisg: will you be somewhat available?
also probably need to upload a new LDM as well, due to the ltsp-update-sshkeys changes.
Ah I didn't test changing the IP...
well, the old LDM logs in using IP only.
Indeed, I'm just saying that the code in ldm trunk is untested
we switched to essentially requiring manually adding IP based entries for ltsp-update-sshkeys, or specifying server hostname in HOST_XX and LDM_SERVER options
alkisg: not untested, i do recall building with it at least once
so I am trying out ltsp in 12.04
seems mostly like 11.10
has there been any significant changes?
vagrantc: I removed ssh_hostchecker without checking that defaulting to LDM_SERVER=server actually works. It should, but I haven't tested it (changing the server IP and still being able to login)
sure feels like huge changes, but maybe the user wouldn't notice :)
vagrantc: well it sure installs about the same.
I noticed a new dhcp server, but that was in 11.10 too wasn't it?
_UsUrPeR_: the NBD serve is handled using nbd-server's native mode, rather than spawning from inetd
The LTSP 5.3 announcement
ok, outta here guys.
|18:28||_UsUrPeR_ has left IRC (_UsUrPeR_firstname.lastname@example.org, Quit: Leaving)|
|18:28||litlebuda has joined IRC (email@example.com)|
|18:29||Trixboxer has left IRC (Trixboxer!~Trixboxer@188.8.131.52, Quit: "Achievement is not the end, its the beginning of new journey !!!")|
|18:35||[GuS] has left IRC ([GuS]!~MysT@unaffiliated/gus/x-663402, Quit: Konversation terminated!)|
|18:37||awilliams has left IRC (awilliams!mistik1@unaffiliated/mistik1, Ping timeout: 250 seconds)|
|18:37||[GuS] has joined IRC ([GuS]!~MysT@213-117-16-190.fibertel.com.ar)|
|18:37||[GuS] has joined IRC ([GuS]!~MysT@unaffiliated/gus/x-663402)|
|18:38||awilliams has joined IRC (awilliams!~awilliams@unaffiliated/mistik1)|
|18:46||Parker955_Away is now known as Parker955|
|18:49||Parker955 is now known as Parker955_Away|
|18:51||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)|
|18:51||awilliams has left IRC (awilliams!~awilliams@unaffiliated/mistik1, Ping timeout: 245 seconds)|
|18:53||awilliams has joined IRC (awilliams!mistik1@unaffiliated/mistik1)|
|19:03||awilliams has left IRC (awilliams!mistik1@unaffiliated/mistik1, Ping timeout: 260 seconds)|
|19:07||litlebuda has left IRC (firstname.lastname@example.org, Remote host closed the connection)|
|19:24||hughessd has joined IRC (hughessd!~steve@173-164-117-109-Oregon.hfc.comcastbusiness.net)|
|19:35||Phantomas has left IRC (Phantomas!~Phantomas@unaffiliated/phantomas, Ping timeout: 252 seconds)|
|19:40||[GuS] has left IRC ([GuS]!~MysT@unaffiliated/gus/x-663402, Quit: Konversation terminated!)|
|19:46||GodFather has left IRC (GodFatheremail@example.com, Quit: Leaving)|
|19:51||bieb has left IRC (firstname.lastname@example.org)|
|19:52||Phantomas has joined IRC (Phantomas!~Phantomas@unaffiliated/phantomas)|
|20:08||awilliams has joined IRC (awilliams!mistik1@unaffiliated/mistik1)|
|20:40||Steve_The_Pirate has left IRC (Steve_The_Pirate!~Gary@cpc16-haye17-2-0-cust169.haye.cable.virginmedia.com, Quit: Leaving)|
|20:54||andygraybeal has joined IRC (email@example.com)|
|21:08||alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)|
|21:10||artista_frustrad has left IRC (firstname.lastname@example.org, Quit: Leaving)|
|21:31||komunista has left IRC (email@example.com, Quit: Leaving.)|
|22:02||alexqwesa has left IRC (firstname.lastname@example.org, Remote host closed the connection)|
|22:03||alexqwesa has joined IRC (email@example.com)|
|22:04||bobby_C has left IRC (bobby_Cfirstname.lastname@example.org, Ping timeout: 260 seconds)|
|22:06||alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)|
|22:10||dead_inside has left IRC (email@example.com, Quit: Leaving...)|
|22:11||hughessd has left IRC (hughessd!~steve@173-164-117-109-Oregon.hfc.comcastbusiness.net, Quit: gone for some reason)|
|22:18||risca has joined IRC (firstname.lastname@example.org)|
|22:21||adrianorg_ has joined IRC (email@example.com)|
|22:24||Parker955_Away is now known as Parker955|
|22:25||adrianorg__ has left IRC (firstname.lastname@example.org, Ping timeout: 256 seconds)|
|22:28||darkpixel_ has left IRC (darkpixel_!~darkpixel@curetheitch/staff/darkpixel, Quit: Leaving.)|