IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 25 March 2011 (all times are UTC)
| 00:20 | alkisg (~alkisg@ubuntu/member/alkisg) joined #ltsp. | |
| 00:35 | <dan_l> alkisig: sorry - I meant ltsp-remoteapps
| |
| 00:35 | gnunux (~emmanuel@194.167.18.244) joined #ltsp. | |
| 00:36 | <dan_l> Yes - I know the homedirs are on the server but I need to them to appear on the desktop which requires a refresh from the client side.
| |
| 00:37 | I also need to call gnome-session-save or pkill -U$user to to force a logoff of the client
| |
| 00:41 | Selveste1 (~Selveste1@88.83.86.20.static.dong.customer.smilecontent.dk) joined #ltsp. | |
| 00:41 | <alkisg> dan_l: can you describe the process in more detail?
| |
| 00:42 | The user inserts the card on the fat client, before he logs in?
| |
| 00:51 | abeehc (~bob@d216-232-201-39.bchsia.telus.net) left irc: Ping timeout: 246 seconds | |
| 00:52 | abeehc (~bob@d216-232-201-39.bchsia.telus.net) joined #ltsp. | |
| 01:24 | dobber (~dobber@89.190.197.77) joined #ltsp. | |
| 01:49 | Da-Geek (~Da-Geek@nat/redhat/x-gvcdxecmpmzlvsya) joined #ltsp. | |
| 01:51 | gentgeen__ (~kevin@c-98-236-71-64.hsd1.pa.comcast.net) left irc: Read error: Operation timed out | |
| 01:51 | gentgeen__ (~kevin@c-98-236-71-64.hsd1.pa.comcast.net) joined #ltsp. | |
| 02:42 | alkisg (~alkisg@ubuntu/member/alkisg) left irc: Quit: Leaving. | |
| 02:52 | Trixboxer (~Trixboxer@office.supportdepartment.net) joined #ltsp. | |
| 03:38 | artista_frustrad (~artista_f@189.75.139.149) left irc: Remote host closed the connection | |
| 03:41 | artista_frustrad (~artista_f@189.75.139.149) joined #ltsp. | |
| 03:47 | garymc (~chatzilla@host81-139-133-67.in-addr.btopenworld.com) joined #ltsp. | |
| 04:03 | HeMan (~jimmy@2a00:801:f5:40:10::19) joined #ltsp. | |
| 04:27 | MorningSon (~MorningSo@cpe-70-114-21-95.satx.res.rr.com) joined #ltsp. | |
| 04:34 | Gremble (~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com) joined #ltsp. | |
| 04:37 | Nick change: zz_evil_root -> evil_root | |
| 04:39 | <HeMan> I'm trying to blacklist a module in LTSP 5.2.4 (Ubuntu 10.10) with rebuilding the initrd in a chroot and then running ltsp-update-kernels
| |
| 04:39 | but the module still loads
| |
| 04:52 | garymc (~chatzilla@host81-139-133-67.in-addr.btopenworld.com) left irc: Ping timeout: 255 seconds | |
| 04:57 | <evil_root> HeMan have you tried going into the chroot, mount -t proc /proc, run rmmod the_mod_name and then putting blacklist the_mod_name in /etc/modprobe.d/blacklist then you exit the chroot, ltsp-update-kernels, unmount /opt/ltsp/arch/proc, then finally ltsp-update-image
| |
| 04:57 | i think its actually /etc/modprobe.d/blacklist.conf
| |
| 04:59 | <HeMan> evil_root: the module is not loaded on my server, so there is no module to rmmod
| |
| 04:59 | <evil_root> then try skipping that one step
| |
| 05:00 | and i am not talking about it being loaded on the server, i am talking if its in the chroot
| |
| 05:00 | <HeMan> to clearify, it's the nouveau-module I like to skip to get nvidia to load instead
| |
| 05:00 | umm, the chroot is done on the server, where else?
| |
| 05:01 | <evil_root> if your using i386 then run chroot /opt/ltsp/i386, and then see if you can find the module
| |
| 05:02 | <HeMan> yes, the module is /lib/modules/2.6.35-28-generic/kernel/drivers/gpu/drm/nouveau/nouveau.ko
| |
| 05:02 | and it's still in my initrd as well
| |
| 05:03 | xavierb (~xavier@lns-bzn-32-82-254-16-28.adsl.proxad.net) joined #ltsp. | |
| 05:07 | <evil_root> ok, when you were trying to remove it, where you trying from the server or from in the chroot environment you just got into?
| |
| 05:07 | Gremble (~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com) left irc: Quit: I Leave | |
| 05:08 | <HeMan> what is /usr/share/initramfs-tools/hooks/ltsp_nbd in the chroot for? It has FRAMEBUFFER=y, could it be part of the problem?
| |
| 05:10 | evil_root: I chroot'ed, added blacklist nouveau in /etc/modprobe.d/blacklist.conf, rebuilt initrd, exited and then ltsp-update-kernels from the server
| |
| 05:10 | <evil_root> i am not sure why that is there, but if i had to guess thats probably just a module to get the nbd working for the ltsp, I doubt its causing another module to load though
| |
| 05:10 | and you need to unmout proc and run ltsp-update-image HeMan
| |
| 05:11 | <HeMan> evil_root: why do i need to update the image when I only change initrd?
| |
| 05:11 | <evil_root> its been my experience if you are using nbd and not strait nfs, and you change anything in the chroot, its always a good idea to also update the image
| |
| 05:12 | <HeMan> just realised that I had updated the image also
| |
| 05:12 | <evil_root> again i couldn't say why, just been playing with ltsp for a long time and its a guess i have
| |
| 05:12 | cool
| |
| 05:12 | now reboot the client and cross your fingers :)
| |
| 05:13 | <HeMan> well, it didn't work
| |
| 05:13 | <evil_root> dam
| |
| 05:13 | <HeMan> that's just what I tested
| |
| 05:13 | <evil_root> sorry man, i am out of ideas
| |
| 05:14 | <HeMan> but I just noticed that the module is loaded really late now so it can't be initrd now
| |
| 05:14 | <evil_root> lol i hate ubuntu
| |
| 05:14 | give me gentoo any day
| |
| 05:15 | <mgariepy> good morning everyone
| |
| 05:16 | <evil_root> morning mgariepy
| |
| 05:21 | HeMan have you tried just moving /lib/modules/2.6.35-28-generic/kernel/drivers/gpu/drm/nouveau/nouveau.ko to like /opt/modbak in the chroot so that the kernel can not even find the module even if something is calling it?
| |
| 05:21 | <HeMan> evil_root: I'll try to remove the xorg nouveau-driver to see if it helps
| |
| 05:22 | <evil_root> cool
| |
| 05:36 | alkisg (~alkisg@ubuntu/member/alkisg) joined #ltsp. | |
| 05:39 | <alkisg> HeMan: are you trying to use the proprietary nvidia driver on the chroot? You don't need to remove nouveau for that
| |
| 05:46 | ogra_ (~ogra@p5098ed03.dip0.t-ipconnect.de) left irc: Excess Flood | |
| 05:47 | ogra_ (~ogra@p5098ed03.dip0.t-ipconnect.de) joined #ltsp. | |
| 05:51 | <HeMan> ha! it was the xorg-driver that loaded the nouveau-module
| |
| 05:53 | now I just have to high DPI settings (305x305)...
| |
| 05:59 | primeministerp (~ppouliot@64.119.153.82) left irc: Remote host closed the connection | |
| 06:00 | primeministerp (~ppouliot@64.119.153.82) joined #ltsp. | |
| 06:05 | jtwood (~jtwood@74.81.111.2) joined #ltsp. | |
| 06:05 | <jtwood> Hello everyone.
| |
| 06:06 | <evil_root> morning jtwood
| |
| 06:06 | <jtwood> Under the thin client setup I was using pam_mount and kerberos to auth and mount a cifs share for home directories
| |
| 06:06 | When I swtiched over to fat-client setups it seemed to break - do I just need to configure kerberos and pam_mount in the fat-client chroot?
| |
| 06:06 | alkisg (~alkisg@ubuntu/member/alkisg) left irc: Ping timeout: 246 seconds | |
| 06:06 | <jtwood> The weird thing is...it worked yesterday
| |
| 06:07 | So I just assumed that ldm took care of it all...but no lie, I am not sure how that works
| |
| 06:15 | <pmatulis> jtwood: i guess with a thin client the ldap client config is on the ltsp server and with a fat client it's on the client
| |
| 06:16 | <jtwood> Thats my guess too...guess I can give it a go...
| |
| 06:16 | I just hate to start changing stuff...even backing it up it has a tendency to end up snafu
| |
| 06:16 | bobby_C (~bobby@85-124-22-227.teleworker.xdsl-line.inode.at) joined #ltsp. | |
| 06:18 | <jtwood> pmatulis: I appreciate the suggestion...it never ceases to amaze me how for the Linux has come in the 10 years I wasn't around...HAHA
| |
| 06:19 | *how far
| |
| 06:19 | Haven't dosed myself with enough caffeine yet this morning
| |
| 06:23 | <pmatulis> "drink more coffee - do stupid things faster"
| |
| 06:24 | <jtwood> pmatulis: Is that your own quote?!?!?! I love it...gonna steal that one if I may!
| |
| 06:25 | <pmatulis> jtwood: no, someone put up a poster here
| |
| 06:25 | <jtwood> pmatulis: Apparently the mounting and auth is done by ldm --- cause I rebooted the client and now the mounts are working.
| |
| 06:25 | I always tell my users -- reboot if it doesn't work...then call me
| |
| 06:26 | ogra_ (~ogra@p5098ed03.dip0.t-ipconnect.de) left irc: Excess Flood | |
| 06:27 | ogra_ (~ogra@p5098ed03.dip0.t-ipconnect.de) joined #ltsp. | |
| 06:29 | cliebow (~cliebow@WatchGuard.ellsworth-hs.ellsworth.k12.me.us) joined #ltsp. | |
| 06:31 | drdata (~drdata@pool-173-57-189-158.dllstx.fios.verizon.net) left irc: Ping timeout: 255 seconds | |
| 06:42 | Gremble (~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com) joined #ltsp. | |
| 06:46 | Gadi (~romm@ool-18bbe47a.static.optonline.net) joined #ltsp. | |
| 06:57 | drdata (~drdata@tw2.udallas.edu) joined #ltsp. | |
| 06:59 | bobby_C (~bobby@85-124-22-227.teleworker.xdsl-line.inode.at) left irc: Quit: Goin' down hard | |
| 07:22 | alkisg (~alkisg@ubuntu/member/alkisg) joined #ltsp. | |
| 07:25 | tobias (~quassel@c-a3aae155.94-9-64736c11.cust.bredbandsbolaget.se) joined #ltsp. | |
| 07:25 | Nick change: tobias -> Guest9338 | |
| 07:26 | <alkisg> jtwood: the authentication is the same both for thin and fat clients
| |
| 07:26 | So any pam configuration where ssh works (on the server) should work on thin/fat clients too
| |
| 07:37 | <cliebow> Gadi:sanity check? i pick up last uid from ldap..the variable is uid..i want to increment by one every time i go through a loop..shouldnt..... let "uid+=1" increment $uid?
| |
| 07:37 | <Gadi> $uid++
| |
| 07:38 | <cliebow> coolio! let me try it..
| |
| 07:40 | in bash..../addmultiplestaff: line 155: 14062++: command not found
| |
| 07:41 | it gets 14062 from ldap..inserts it in
| |
| 07:43 | let "uid+=1" always used to work..
| |
| 07:43 | <alkisg> uid=$(($uid+1)) for sh compatibility
| |
| 07:43 | (if that's shell and not something ldap related)
| |
| 07:43 | <cliebow> thank you give that a whirl
| |
| 07:43 | yeah shell
| |
| 07:47 | Roasted_ (~jason@206.82.22.190) joined #ltsp. | |
| 07:49 | <cliebow> ahhhh..found it..
| |
| 07:50 | Gremble (~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com) left irc: Quit: I Leave | |
| 07:53 | <alkisg> Suppose I have *no X* on a server, but I have a fat chroot. And I also want to serve thin clients from that fat chroot, in order to only install programs once.
| |
| 07:53 | If I use "ChrootDirectory /opt/ltsp/i386" in sshd_config, would that be enough?!
| |
| 07:57 | rjune (~rjune@oh-67-77-122-143.sta.embarqhsd.net) joined #ltsp. | |
| 07:58 | <Gadi> alkisg: maybe - if you use NFS_HOME
| |
| 07:58 | <alkisg> Right
| |
| 07:58 | <Gadi> might blow peoples' minds even more than it already does, tho
| |
| 07:58 | :)
| |
| 07:58 | <alkisg> I think I'd also need to add the users in $CHROOT/etc/passwd though
| |
| 07:59 | <Gadi> I dont think so
| |
| 07:59 | it should auth and then put you in jail
| |
| 07:59 | no?
| |
| 07:59 | <alkisg> changed chrootdirectory, ssh'ed to the server, and I have: I have no name!@alkis:/
| |
| 07:59 | <Gadi> ah
| |
| 08:00 | <alkisg> Maintaining 1 chroot for both thin and fat clients sounds interesting. And the server would be more robust, no X, no users...
| |
| 08:00 | (I'd use the pam that allows $CHROOT/etc/passwd to be used for auth)
| |
| 08:08 | GodFather (~rcc@c-98-250-128-114.hsd1.mi.comcast.net) joined #ltsp. | |
| 08:11 | staffencasa (~staffenca@128-193-144-246.oregonstate.edu) joined #ltsp. | |
| 08:13 | Guest9338 (~quassel@c-a3aae155.94-9-64736c11.cust.bredbandsbolaget.se) left irc: Remote host closed the connection | |
| 08:14 | Roasted_ (~jason@206.82.22.190) left irc: Ping timeout: 260 seconds | |
| 08:15 | whatch (~whatch@12.204.172.194) joined #ltsp. | |
| 08:16 | <whatch> does anybody know of documentation that explains how to get LTSP working on a Ubuntu Server, where DHCP is handled by Novell Netware?
| |
| 08:17 | <alkisg> Just provide the boot filename on your dhcp server,
| |
| 08:17 | and set next-server == the ltsp server
| |
| 08:18 | <whatch> the boot filename lives on the Ubuntu server?
| |
| 08:19 | <alkisg> /ltsp/i386/pxelinux.0
| |
| 08:20 | So, to sum up, an alternative way to maintain LTSP thin/fat clients:
| |
| 08:20 | 1) Install the client OS "graphically", so that ltsp-build-client isn't necessary. E.g. with the Ubuntu desktop CD. Wherever; in a USB disk on an arm client, in a separate partition on the server, in a vbox .vmdk disk, in a qemu disk...
| |
| 08:20 | 2) Install ltsp-client on that client OS. That would contain all the LTSP magic.
| |
| 08:20 | 3) Better make that partition with btrfs + lzo compression. That would make squashfs/ltsp-update-image unnecessary.
| |
| 08:20 | 4) Use lib_pamssh. That would make ldm unnecessary.
| |
| 08:20 | 5) Export that partition with nbd for fat clients.
| |
| 08:20 | 6) Thin clients would again boot from that nbd export, but they'd also use it for their sessions, with sshd ChrootDirectory.
| |
| 08:20 | No ltsp-build-client, no ltsp-update-image, no ldm, no console, no need for duplicate apps. Cool! ;) :D
| |
| 08:20 | <whatch> wowie, you really lost me now
| |
| 08:21 | <alkisg> whatch: that was not for you :)
| |
| 08:21 | <whatch> oops, sorry
| |
| 08:24 | dgroos (~dgroos@63.225.132.145) joined #ltsp. | |
| 08:30 | <Gadi> alkisg: definitely an alternative, but you do limit the thin clients a bit more that way. For example, they would only run 32-bit apps on the server even if the server is 64-bit
| |
| 08:31 | <alkisg> Indeed, good thought
| |
| 08:31 | And chroot wouldn't work if it was an arm disk
| |
| 08:31 | <Gadi> right
| |
| 08:31 | arm is a planet away
| |
| 08:31 | :)
| |
| 08:31 | <alkisg> So no way to support 2 arches without duplicate apps. :(
| |
| 08:32 | (for both thin+fat, that is)
| |
| 08:35 | <Gadi> right - well, for obvious reasons
| |
| 08:35 | :)
| |
| 08:36 | <alkisg> Hmmm and probably the ChrootDirectory wouldn't work anyway because there would be no /proc /sys mounted, no system dbus running etc etc
| |
| 08:36 | <Gadi> yeah, true too
| |
| 08:40 | whatch (~whatch@12.204.172.194) left irc: Quit: Leaving | |
| 08:52 | <jtwood> alkisg: Yeah I see that the auth is the same for both...part of ldm---kind of a frontend for auth for the desktop session?
| |
| 08:53 | Just curious --- looked around and info was sparse on ldm
| |
| 08:54 | <Gadi> jtwood: ldm is LTSP Display Manager - our own creation. Basically needed something that would collect username/password and create an ssh tunnel and session for us
| |
| 08:55 | at last hackfest we talked about PAM-ifying the process so we could use any DM
| |
| 08:55 | and not have to maintain our own
| |
| 08:55 | <jtwood> Oh my...I should have known...so, in otherwords...the answer is yes to my question...
| |
| 08:55 | :)
| |
| 08:55 | thunsucker (~thunsucke@209.65.135.30) joined #ltsp. | |
| 08:55 | <Gadi> its part of our ongoing mission to minimize our own work
| |
| 08:55 | :)
| |
| 08:55 | <jtwood> I totally understand
| |
| 08:55 | Thats what LTSP is doing for me...you guys deserve a bit less work load too!
| |
| 08:55 | <thunsucker> Roasted: how is your fat client setup going?
| |
| 08:56 | jtwood: I owe Gadi and Alkisg many many lunches due to the amount of hassle they have saved me
| |
| 08:57 | <jtwood> thunsucker: Yeah, no doubt there...I owe at least a few beers and a lunch or two as well
| |
| 08:57 | <thunsucker> I'm getting ready to try out FF 4 in a fat kiosk to see how it works
| |
| 08:57 | <jtwood> Of course...if you have a good stout or ale in front of you...like a meal in a bottle!
| |
| 08:57 | ;)
| |
| 08:57 | <thunsucker> :P
| |
| 08:58 | <xavierb> I'm looking for tips for running kde4 with ltsp. Without changes it is somewhat horrible!
| |
| 08:58 | <thunsucker> xavierb: you running thin or fat clients?
| |
| 08:59 | <xavierb> thin, mostly geode based (Artec's thincans)
| |
| 08:59 | <thunsucker> xavierb: very cool, I want the boss to buy one of those
| |
| 08:59 | how much ram and cpu you have on the server?
| |
| 09:01 | <xavierb> A lot! ;-P 16GB of ram and 2 AMD opteron
| |
| 09:02 | for 6 to 15 user (but mainly 6)
| |
| 09:02 | About thincans: they are cheap and solid, but geode drivers could be better...
| |
| 09:04 | <Gadi> hehe - aint that the truth
| |
| 09:06 | <xavierb> well, Mart Raudsepp is still working on improving it
| |
| 09:06 | <Gadi> oh, several folks are
| |
| 09:06 | for sure
| |
| 09:07 | with LTSP, tho, your best with LDM_DIRECTX, 16-bit color depth, and messing around with the driver's acceleration
| |
| 09:09 | <xavierb> Yeah, I've done that. I think KDE4 doesn't like it!
| |
| 09:15 | <alkisg> horrible == slow? Or buggy?
| |
| 09:15 | <xavierb> slow and buggy
| |
| 09:15 | <alkisg> Hehe. What bugs?
| |
| 09:15 | And, so it's fast with non-geode clients?
| |
| 09:15 | <xavierb> part of the slowness comes from akonadi, I need to work on it
| |
| 09:16 | display bugs: missing windows decorations mainly
| |
| 09:17 | it's better with non-geode (pc)
| |
| 09:18 | <alkisg> Is it using composition?
| |
| 09:18 | If so, try disabling it for the windows decorations
| |
| 09:18 | <xavierb> no, I've disabled it
| |
| 09:19 | window decoration problem was mentionned on the mailing list: new kde themes use a lot of resources
| |
| 09:19 | <alkisg> Hmm which translates to pixmaps, wasting local X ram?
| |
| 09:20 | How much RAM does those clients have?
| |
| 09:20 | <xavierb> 256MB
| |
| 09:20 | <alkisg> Not too low. It would have to use a lot of pixmaps to waste 100M
| |
| 09:20 | Running `free` or `xrestop` locally wouldn't hurt though
| |
| 09:20 | Roasted_ (~jason@206.82.22.190) joined #ltsp. | |
| 09:21 | <xavierb> I will try to change QT option (-graphicssystem=raster), but it is supposed to use more network bandwith
| |
| 09:22 | gnunux (~emmanuel@194.167.18.244) left irc: Quit: Ex-Chat | |
| 09:26 | vagrantc (~vagrant@c-24-21-191-93.hsd1.or.comcast.net) joined #ltsp. | |
| 09:36 | thunsucker_ (~thunsucke@209.65.135.30) joined #ltsp. | |
| 09:39 | Gremble (~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com) joined #ltsp. | |
| 09:39 | Gremble (~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com) left irc: Remote host closed the connection | |
| 09:40 | thunsucker (~thunsucke@209.65.135.30) left irc: Ping timeout: 240 seconds | |
| 09:41 | dobber (~dobber@89.190.197.77) left irc: Remote host closed the connection | |
| 09:44 | alkisg (~alkisg@ubuntu/member/alkisg) left irc: Ping timeout: 250 seconds | |
| 09:48 | alkisg (~alkisg@ubuntu/member/alkisg) joined #ltsp. | |
| 09:50 | <Roasted_> Hey guys. How does printing work with LTSP? If it's added on the server will it be accessible on the fat clients?
| |
| 09:54 | <alkisg> If you export it from the cups printer properties
| |
| 09:55 | [v] Publish shared printers connected to this system
| |
| 09:58 | <Roasted_> alkisg, so if I just do that in the GUI of the server the clients will see it?
| |
| 09:59 | I don't have to do any terminal work to pass it off to the chroot?
| |
| 09:59 | <alkisg> Yes
| |
| 09:59 | No
| |
| 09:59 | <Roasted_> you the man alkisg
| |
| 09:59 | I'm giving a presentation on this here LTSP stuff monday morning
| |
| 09:59 | <alkisg> Well cups is the man here :)
| |
| 09:59 | Nothing related to LTSP
| |
| 09:59 | <Roasted_> hopefully our cirriculum director is listening
| |
| 10:00 | well it's part of that awesome thing known as open source...
| |
| 10:00 | where everything is just... great
| |
| 10:00 | lives happily ever after, etc.
| |
| 10:00 | drdata (~drdata@tw2.udallas.edu) left irc: Ping timeout: 252 seconds | |
| 10:04 | * vagrantc wonders what faery tale Roasted_ lives in | |
| 10:04 | <thunsucker_> lol
| |
| 10:04 | <Roasted_> uh oh. somebody here not share the same open source enthusiasm?
| |
| 10:04 | :P
| |
| 10:05 | <thunsucker_> i only have one issue with open source at the moment
| |
| 10:05 | <vagrantc> i'm a free software developer, myself :P
| |
| 10:06 | <Roasted_> wow. my fat client is only using 180 meg of ram to run?
| |
| 10:06 | <thunsucker_> I run an environment where I can't choose 100% of what software we run. We run a little bit of everything. Windows, Ubuntu, Mac, Terminal Services. and etc.
| |
| 10:06 | <Roasted_> how is that possible?
| |
| 10:06 | <thunsucker_> I haven't found a desktop yet that I'm happy with managing everything
| |
| 10:07 | Roasted_: ubuntu rocks, thats why :P
| |
| 10:08 | <Roasted_> there's a lot of distros that rock
| |
| 10:08 | but alas
| |
| 10:08 | I have to agree
| |
| 10:08 | <vagrantc> stallman may be a wingnut, but he's my kind of wingnut
| |
| 10:08 | <thunsucker_> is the kiosk bug going to be fixed soon? where there is a missing script that you have to download
| |
| 10:08 | <Roasted_> hmm
| |
| 10:09 | <alkisg> Roasted_: up until the ldm screen, fat clients should boot with 64 MB RAM. After that, the RAM used depends on the DE/programs you run.
| |
| 10:10 | <Roasted_> alkisg, I just expected them to use more ram.
| |
| 10:10 | <alkisg> Most services are disabled by default on ltsp
| |
| 10:10 | <Roasted_> After all, you said if it can run off a live usb, it can probably run as a fat client. I just expected it to eat up more resources :P
| |
| 10:10 | ahh
| |
| 10:10 | what kind of services exactly?
| |
| 10:10 | anything that I may need?
| |
| 10:10 | <alkisg> But once you open firefox you'll need 512 RAM anyway
| |
| 10:10 | <Roasted_> yeah
| |
| 10:11 | so far I do pretty dang good on resources even with gimp and firefox open
| |
| 10:11 | <alkisg> All services except for those defined in the ltsp RC_WHITELIST, some internal variable in ltsp-build-client
| |
| 10:11 | patrickmw (~pwright@ip68-231-4-153.ph.ph.cox.net) left irc: Quit: Ex-Chat | |
| 10:11 | <Roasted_> I have yet to log into these 30 systems at once though
| |
| 10:11 | but they're fat clients
| |
| 10:11 | <alkisg> (on Ubuntu - and I believe that's not a good thing)
| |
| 10:11 | <Roasted_> so I would assume the server would react the same, whether I have 1 client on or 30
| |
| 10:11 | <vagrantc> alkisg: i finally gave up on that with debian
| |
| 10:11 | <alkisg> Yes, server CPU/RAM isn't strained
| |
| 10:12 | vagrantc: I believe you did well to disable the whitelist
| |
| 10:12 | E.g. now we're missing cpufreq etc
| |
| 10:12 | <Roasted_> alkisg, I never got to bond my NICs. It was just too much of a headache for right now when I need to get this lab rolling asap. Think 1 GB NIC on PCIE will do okay for a lab of 30?
| |
| 10:12 | <vagrantc> alkisg: actually, it was insserv that effectively disabled it ... by redoing the rc?.d/* symlinks whenever it ran
| |
| 10:12 | <alkisg> A blacklist would be more appropriate
| |
| 10:12 | <vagrantc> alkisg: so i just gave up on something that never felt quite right anyways
| |
| 10:13 | <alkisg> vagrantc: I think update-rc notifies insserv, no? - instead of just deleting the files...
| |
| 10:13 | <vagrantc> alkisg: it wasn't working when i disabled it
| |
| 10:13 | <alkisg> But in any case I think that's an initramfs job
| |
| 10:13 | Not an ltsp-build-client one, if we ever want to be able to install ltsp-client to any system
| |
| 10:14 | In lts.conf, DISABLE_SERVICES=xxx, and an initramfs script that deletes the init symlinks
| |
| 10:14 | <vagrantc> yeah, handling too much in ltsp-build-client, while easy, doesn't work for long-term maintenance
| |
| 10:14 | alkisg: i've got to try the aufs on NFS sometime...
| |
| 10:15 | <alkisg> It worked for me
| |
| 10:15 | <vagrantc> that would avoid a lot of headache
| |
| 10:15 | <alkisg> Maybe even nfs + cow nbd for swap+tmpfs over aufs
| |
| 10:15 | Instead of nbd swap, that is. Saves more ram
| |
| 10:16 | <vagrantc> alkisg: just using a swap file on the writeable filesystem?
| |
| 10:16 | <alkisg> Right
| |
| 10:16 | And that would also be able to store all the deletions/modifications etc directly on the server, saving ram on the client
| |
| 10:16 | * vagrantc wonders what happened to swapd | |
| 10:16 | Roasted (~jason@unaffiliated/roasted) left irc: Ping timeout: 255 seconds | |
| 10:17 | <alkisg> I've seen RAM=35MB on LDM with that
| |
| 10:17 | *used ram
| |
| 10:17 | Nope sorry that was xdmcp.
| |
| 10:17 | <vagrantc> heh
| |
| 10:20 | Da-Geek (~Da-Geek@nat/redhat/x-gvcdxecmpmzlvsya) left irc: Quit: Leaving | |
| 10:33 | primeministerp (~ppouliot@64.119.153.82) left irc: Read error: Connection reset by peer | |
| 10:33 | dan_l (~chatzilla@63.193.16.168) left irc: Remote host closed the connection | |
| 10:33 | primeministerp (~ppouliot@64.119.153.82) joined #ltsp. | |
| 10:36 | dan_l (~chatzilla@63.193.16.168) joined #ltsp. | |
| 10:36 | xavierb (xavier@lns-bzn-32-82-254-16-28.adsl.proxad.net) left #ltsp ("Konversation terminated!"). | |
| 10:41 | befrenchy (8e2e1f1e@gateway/web/freenode/ip.142.46.31.30) joined #ltsp. | |
| 10:42 | <befrenchy> Need help with a problem with switching between screens
| |
| 10:43 | I have an app running in SCREEN_03, when i switch to SCREEN_01 and switch back, screen is black and only have a cursor showing
| |
| 10:44 | can't figure out why window won
| |
| 10:44 | 'won't display properly
| |
| 10:44 | anybody have any pointers on what i can look at?
| |
| 10:44 | <alkisg> Which graphics card, distro, version? Try adding "nomodeset" in pxelinux.cfg/default.
| |
| 10:44 | * highvoltage does | |
| 10:45 | <befrenchy> ubuntu 10.10
| |
| 10:45 | <highvoltage> befrenchy: 0x4a28213a
| |
| 10:45 | <alkisg> highvoltage was reading too much xkcd... :P :D
| |
| 10:45 | <befrenchy> Client is a dell with intel graphics
| |
| 10:45 | * highvoltage is busted | |
| 10:45 | <alkisg> befrenchy: which intel graphics?
| |
| 10:45 | `lspci -nn -k | grep -A 2 VGA` helps
| |
| 10:46 | On a local xterm
| |
| 10:46 | I.e. `ltsp-localapps xterm`
| |
| 10:46 | <befrenchy> alkisg: intel 945G
| |
| 10:47 | <alkisg> The exact output of lspci would be better. But ok try adding "nomodeset" in /var/lib/tftpboot/ltsp/i386/pxelinux.cfg/default
| |
| 10:47 | Right next to "quiet splash"
| |
| 10:47 | <befrenchy> alkisg: ok
| |
| 10:47 | <alkisg> Also, you don't put SCREEN_01=xx in lts.conf, right?
| |
| 10:48 | It's reserved, use SCREEN_02 or more
| |
| 10:48 | <befrenchy> alkisg: correct
| |
| 10:48 | alkisg: is there by default, just using it to troubleshoot
| |
| 10:49 | tompa82 (~quassel@c-a3aae155.94-9-64736c11.cust.bredbandsbolaget.se) joined #ltsp. | |
| 10:49 | <befrenchy> alkisg: http://pastebin.com/iSw3vRyq
| |
| 10:49 | <alkisg> I didn't understand the last sentence. What is there by default?
| |
| 10:50 | <befrenchy> alkisg: was just saying that i don't include SCREEN_01 in lts.conf but you can which to it to access a shell login
| |
| 10:50 | <alkisg> Right, but unless you unlock the root account you can't use that shell
| |
| 10:50 | Try the nomodeset thing
| |
| 10:50 | patrickmw (~pwright@ip68-231-4-153.ph.ph.cox.net) joined #ltsp. | |
| 10:51 | <befrenchy> alkisg: i'm able to login :)
| |
| 10:51 | <alkisg> You added users to the chroot?!
| |
| 10:51 | <befrenchy> alkisg: no just set root password
| |
| 10:51 | <alkisg> OK, that's why I said "unless" :)
| |
| 10:52 | <befrenchy> alkisg: how to you use the nomodeset option?
| |
| 10:52 | <alkisg> sudo gedit /var/lib/tftpboot/ltsp/i386/pxelinux.cfg/default
| |
| 10:52 | Write "nomodeset" right next to "quiet splash"
| |
| 10:52 | Without the quotes. With a space around.
| |
| 10:52 | Save. Reboot client. Change VTs
| |
| 10:53 | <befrenchy> alkisg: quiet splash?
| |
| 10:53 | <alkisg> append ro initrd=initrd.img quiet splash nbdport=2000
| |
| 10:53 | Don't you have that line there?
| |
| 10:53 | Change it to: append ro initrd=initrd.img quiet splash nomodeset nbdport=2000
| |
| 10:53 | <befrenchy> alkisg: my bad, looking at the wrong file
| |
| 10:54 | alkisg: what does that command do exactly?
| |
| 10:54 | <alkisg> It disables KMS
| |
| 10:54 | Please don't ask me what KMS is
| |
| 10:54 | Google it :)
| |
| 10:54 | <befrenchy> alkisg: lol
| |
| 10:55 | alkisg: ok , gibing it a try
| |
| 10:57 | alkisg: now the app won't start, getting error: Screens found but none have a usable configuration
| |
| 10:58 | alkisg: fatal server error: mp screens found
| |
| 10:58 | alkisg: fatal server error: no screens found
| |
| 10:59 | <alkisg> Try changing "nomodeset" to "i915.modeset=0"
| |
| 10:59 | If that doesn't work as well, pastebin your lts.conf and your pxelinux.cfg/default
| |
| 10:59 | <befrenchy> alkisg: ok
| |
| 11:03 | alkisg: no luck, http://pastebin.com/Ev4QdLgr
| |
| 11:04 | <alkisg> befrenchy: remove that modeset param
| |
| 11:04 | Just for a test, in lts.conf, try: XSERVER=vesa
| |
| 11:04 | See if it works with vesa
| |
| 11:04 | <befrenchy> alkisg: ok
| |
| 11:11 | alkisg: xinit: no such file or directory , unable to connect to X server
| |
| 11:12 | alkisg: content on vmview screen script: http://pastebin.com/bgA4YsrV
| |
| 11:12 | alkisg: not sure if that helps
| |
| 11:15 | alkisg: Xorg.3.log shows this error: VESA: Kernel modesetting driver in use, refusing to load
| |
| 11:18 | otavio (~otavio@187.117.204.84) joined #ltsp. | |
| 11:18 | otavio (~otavio@187.117.204.84) left irc: Changing host | |
| 11:18 | otavio (~otavio@debian/developer/otavio) joined #ltsp. | |
| 11:22 | drdata (~drdata@tw2.udallas.edu) joined #ltsp. | |
| 11:23 | cliebow (~cliebow@WatchGuard.ellsworth-hs.ellsworth.k12.me.us) left irc: Quit: Leaving | |
| 11:27 | * Gadi fears getltscfg.c actually needs a patch | |
| 11:28 | * vagrantc hasn't known Gadi to fear all that much | |
| 11:28 | <Gadi> but getltscfg.c is sacred
| |
| 11:28 | :)
| |
| 11:29 | but, I think when it returns values, it should really single-quote them and not double
| |
| 11:29 | no?
| |
| 11:29 | <vagrantc> oh boy
| |
| 11:29 | i see where the fear comes in
| |
| 11:29 | there may be many configuration files out there relying on one behavior...
| |
| 11:30 | <Gadi> actually
| |
| 11:30 | the single quotes doesn't
| |
| 11:30 | help
| |
| 11:30 | foo='a * *'
| |
| 11:30 | echo $foo
| |
| 11:30 | still exapnds
| |
| 11:30 | *expands
| |
| 11:34 | aha!
| |
| 11:34 | we should be running: /bin/sh -f
| |
| 11:35 | in ltsp_config
| |
| 11:35 | aaron (~aaron@204.191.8.43) joined #ltsp. | |
| 11:36 | Nick change: aaron -> Guest94358 | |
| 11:36 | * vagrantc wonders if we shouldn't upgrade the bzr repository format for ltsp projects | |
| 11:37 | <Guest94358> I have connected a scanner to a thin client, and have installed xsane in my image, the scanner is detected and all is well, however, the device is owned by root:root and users cannot access it
| |
| 11:37 | <Roasted_> I really
| |
| 11:38 | really should have never tried to bond those interfaces
| |
| 11:38 | ever
| |
| 11:38 | clients currently stuck at TFTP
| |
| 11:38 | PXE-E32 timeout
| |
| 11:38 | <Guest94358> if I change the ownership to the device, or run xsane as root, the scanner works perfectly
| |
| 11:39 | any suggestions?
| |
| 11:40 | <Roasted_> I can't imagine updating the image would fix it...
| |
| 11:40 | <Gadi> are the users in the scanner group?
| |
| 11:41 | Guest94358: ^^
| |
| 11:41 | Roasted_: dont touch the image
| |
| 11:41 | <Roasted_> Gadi, I'm about to rm -rf the image.
| |
| 11:41 | <Gadi> Roasted_: it is simply a tftp issue
| |
| 11:41 | not an image issue
| |
| 11:41 | <Roasted_> Gadi, I rebooted the server. and now this happens.
| |
| 11:41 | how does rebooting *cause* issues.
| |
| 11:42 | <Guest94358> the scanner isn't owned by the scanners group, it's owned by root:root
| |
| 11:42 | <Gadi> it doesnt - bonding may have
| |
| 11:42 | <Roasted_> how do I reverse the bonding
| |
| 11:42 | <Guest94358> Gadi: or does that matter?
| |
| 11:42 | <Roasted_> I thought I just had to edit my interface file
| |
| 11:42 | <Gadi> Guest94358: doesnt matter
| |
| 11:43 | Roasted_: what edit did youmake exactly?
| |
| 11:44 | <Roasted_> Gadi, I created a file in /etc/modprobe.d/bonding.conf, added 2 entries there. Then I edited my interface file under /etc/network.
| |
| 11:44 | <Guest94358> alright, I'll try that
| |
| 11:44 | <Roasted_> Gadi, so naturally when I wanted to reverse it, I deleted the bonding.conf file and I edited my interface file to only utilize 1 NIC
| |
| 11:44 | <Gadi> and now it is reversed?
| |
| 11:44 | <Roasted_> Gadi, but now when I reboot, my NIC doesnt come up. I have to restart networking for it to come up, then it works. Then this last time when I rebooted, it came up when I restarted the network service, but now this TFTP thing with this lab.
| |
| 11:44 | <alkisg> befrenchy: you could try *both* the i915.modeset thing to disable kms, AND vesa, to see if it works this way. If so, you'll be sure it's a driver issue, and try updates etc.
| |
| 11:45 | <Roasted_> I had things *perfect* till I tried to bond it.
| |
| 11:45 | Now things just suck.
| |
| 11:45 | <alkisg> Gadi: -f won't help with $(ls) :)
| |
| 11:45 | We should do a "case" instead of using eval
| |
| 11:46 | <abeehc> does something exist in /proc/net/bonding
| |
| 11:46 | <Roasted_> abeehc, what do you mean by "something"?
| |
| 11:46 | <Gadi> alkisg: Im not sure what you are talking about
| |
| 11:46 | <abeehc> something as in, not nothing?
| |
| 11:46 | <alkisg> Gadi: you were talking about pathname expansion in getltscfg ?
| |
| 11:46 | <abeehc> i have nothing
| |
| 11:47 | <Gadi> alkisg: the problem is that eval $(getltscfg) does a pathname expansion on the var
| |
| 11:47 | <alkisg> Gadi: pm
| |
| 11:47 | <Guest94358> Gadi: I'm looking at the system here and I don't even see a scanners group
| |
| 11:48 | <Roasted_> abeehc, you get no such file or dir?
| |
| 11:48 | <Guest94358> Gadi: this webpage describes the problem people have had in the past and what I have tried to do so far. http://tiny.cc/uvp8w
| |
| 11:48 | <abeehc> i don't even have the dir bonding
| |
| 11:48 | <Roasted_> nor do I
| |
| 11:48 | :(
| |
| 11:48 | <abeehc> and the module is unloaded we can assume then
| |
| 11:49 | <Roasted_> not sure what else we can do
| |
| 11:49 | unless I static the other interface, which comes up each time I boot
| |
| 11:49 | gosh. this is bad. 3pm on a friday and now my clients won't boot.
| |
| 11:49 | this goes live on monday
| |
| 11:49 | WHY DID I BOND
| |
| 11:50 | <abeehc> pastebin ur interfaces maybe'
| |
| 11:50 | <Roasted_> http://pastebin.com/qVWNRm7p
| |
| 11:51 | is there anyway I can just like
| |
| 11:51 | <abeehc> do you have ethtool installed does it say anthing interesting
| |
| 11:51 | <Roasted_> re-vamp the network setup
| |
| 11:51 | never even heard of ethtool
| |
| 11:51 | <abeehc> yeah pretty much
| |
| 11:52 | /etc/udev/rules.d/70-persistent-net.rules
| |
| 11:52 | i sometimes find the need to clear that file
| |
| 11:52 | <Roasted_> just remove it?
| |
| 11:52 | <abeehc> well only when device nubmers swithc i guess
| |
| 11:52 | no just look at it
| |
| 11:52 | and maybe comment out the lines
| |
| 11:53 | also worth trying to see what ethtool says
| |
| 11:53 | <Roasted_> I'm not sure what I'm doing in this file
| |
| 11:54 | what am I to do with it
| |
| 11:54 | <befrenchy> alkisg: still no luck... :(, oh well. thanks for trying
| |
| 11:54 | <abeehc> the file assocates mac to dev
| |
| 11:54 | comment out all your nic lines
| |
| 11:55 | it's re-populate when you reboot i think
| |
| 11:55 | <Roasted_> the subsystem lines?
| |
| 11:55 | <abeehc> yeah crap though READ THEM
| |
| 11:55 | mac address, devnice name
| |
| 11:55 | how many do you have?
| |
| 11:55 | <Roasted_> two
| |
| 11:55 | tompa82 (~quassel@c-a3aae155.94-9-64736c11.cust.bredbandsbolaget.se) left irc: Remote host closed the connection | |
| 11:56 | <abeehc> yeah comment them and reboot see if the new ones are different?
| |
| 11:56 | <Roasted_> abeehc, do you think the fact I don't have eth1 in the interface file makes it confused?
| |
| 11:56 | <abeehc> i think that's how it work
| |
| 11:56 | s
| |
| 11:56 | IF
| |
| 11:56 | you have only eht1 in this file you are looking at
| |
| 11:56 | <Roasted_> no
| |
| 11:56 | I have both
| |
| 11:56 | <abeehc> shouldn't be an issue then
| |
| 11:57 | <Roasted_> k
| |
| 11:57 | rebooted it remotely
| |
| 11:57 | going to go over in person and see whats up when it comes online again
| |
| 11:57 | hope it works...
| |
| 11:57 | <Gadi> Roasted_: do you have both nics plugged into the same switch still?
| |
| 11:57 | <Roasted_> no
| |
| 11:57 | eth1 isn't used at all
| |
| 11:58 | it's not in network manager or interface file
| |
| 12:05 | <Guest94358> Gadi: there is no scanner group, any suggestions?
| |
| 12:05 | <vagrantc> gah. i can't push to ltsp-trunk because it's got such an old format or something...
| |
| 12:06 | Gadi (~romm@ool-18bbe47a.static.optonline.net) left irc: Ping timeout: 240 seconds | |
| 12:09 | <vagrantc> well... hopefully this isn't a bad idea, but i'm trying to upgrade ltsp-trunk's bzr format...
| |
| 12:10 | <alkisg> Roasted_: so when you don't have net access, what's wrong? Your IP? Your subnet? Your route?
| |
| 12:17 | <Roasted_> the server booted up and I could ping it without going over to it and physically bringing the interface up. nice.
| |
| 12:17 | alkisg, I have eth0 and eth1. eth1 unused/unplugged, eth0 static.
| |
| 12:17 | alkisg, when I boot up, if I ran ifconfig I'd see lo and eth1, not eth0.
| |
| 12:18 | alkisg, if I reboot net interfaces, it brings up both. But then out of no where when I rebooted the server once, it was hanging to bring the images down @ TFTP...
| |
| 12:18 | alkisg, I did what abeehc recommended, and he nailed it. It did re-generate the 2 entries, but I'mc omparing them and I see nothing off about the previous eth0 entry, so I'm almost skeptical that it "fixed" it when it doesn't appear as if anything was "wrong" based on the differences
| |
| 12:19 | <alkisg> Roasted_: can you paste your current /etc/network/interfaces?
| |
| 12:19 | <Roasted_> alkisg, I did above
| |
| 12:19 | http://pastebin.com/qVWNRm7p
| |
| 12:19 | The only thing it changed was the order. Eth1 used to be on top with Eth0 below it
| |
| 12:19 | <alkisg> Your gateway is wrong
| |
| 12:19 | <Roasted_> now Eth0 is above
| |
| 12:19 | <alkisg> Is on a different subnet
| |
| 12:20 | <Roasted_> whats the command to check gateway
| |
| 12:20 | <alkisg> ?
| |
| 12:20 | It's wrong because it's on a different subnet
| |
| 12:20 | You don't need to check it.
| |
| 12:20 | <Roasted_> I want to check what my laptop is on
| |
| 12:20 | <alkisg> Ah sorry
| |
| 12:20 | <Roasted_> it's not a whack 24
| |
| 12:20 | <alkisg> Your subnet mask may cover it
| |
| 12:20 | Gadi (~romm@ool-18bbe47a.static.optonline.net) joined #ltsp. | |
| 12:20 | <alkisg> Let me check
| |
| 12:23 | thunsucker (~thunsucke@209.65.135.30) joined #ltsp. | |
| 12:24 | <Roasted_> The only other PITA I'm facing is Edubuntu is on the Windows domain. When a user logs in to my other Edubuntu server, it auto authenticates when I try to hit a windows file server. It just KNOWS I have permission. On this recent one (the big one) it asks for my credentials each time...
| |
| 12:25 | <thunsucker> Roasted_: did you set everything up identical, besides one being thin and one being fat?
| |
| 12:25 | <abeehc> likewise?
| |
| 12:25 | <Roasted_> thunsucker, yep. Even used the exact same likewise.bin install I got from their site.
| |
| 12:25 | yes
| |
| 12:25 | likewise open
| |
| 12:25 | version 6 or whatever is on their site
| |
| 12:25 | <thunsucker> Roasted_: same versions of edubuntu on both?
| |
| 12:26 | <Roasted_> yep
| |
| 12:26 | 10 10 64 bit
| |
| 12:26 | <abeehc> if you want a long weekend of work i think kerberos can help
| |
| 12:26 | thunsucker_ (~thunsucke@209.65.135.30) left irc: Ping timeout: 252 seconds | |
| 12:26 | <Roasted_> abeehc, nooooooooo thanks.
| |
| 12:26 | likewise works. perfectly. on the other install.
| |
| 12:26 | <abeehc> haha sorry
| |
| 12:26 | <Roasted_> It WILL work here. :)
| |
| 12:26 | <abeehc> i imagein if you run klist on the one that works stuff will be there
| |
| 12:26 | <thunsucker> i think there is an edubuntu channel here
| |
| 12:27 | <abeehc> where it wont on the nonworking one
| |
| 12:27 | <Roasted_> like if I'm logged in as steve_jobs and I hit the windows file server, it doesn't ask who I am.
| |
| 12:27 | because I'm already logged in as DOMAIN\Steve_Jobs
| |
| 12:27 | but on this one, it asks each time.
| |
| 12:27 | what if I drop it from domain and rejoin?
| |
| 12:27 | <abeehc> also do lw-enum-groups or whateer to see if it's able to pull groups from your domain
| |
| 12:28 | <thunsucker> i would be interested to see /etc/resolv.conf on both solutions
| |
| 12:28 | both servers and both clients
| |
| 12:28 | <Roasted_> let me ssh in and see wuz ^
| |
| 12:28 | <thunsucker> domain steve-jobs.local
| |
| 12:28 | search steve-jobs.local
| |
| 12:29 | lol
| |
| 12:29 | bobby_C (~bobby@188.20.161.210) joined #ltsp. | |
| 12:29 | <Roasted_> the /etc/resolv.conf is identical, except here at the HS there's 2 DNS entries since we have a backup DNS server here
| |
| 12:29 | but I see domain.local listed twice in each one
| |
| 12:29 | xavierb (~xavier@lns-bzn-32-82-254-16-28.adsl.proxad.net) joined #ltsp. | |
| 12:30 | <thunsucker> did you check /etc/resolv.conf of the servers and inside the chroot as well?
| |
| 12:30 | <Roasted_> no
| |
| 12:30 | does that matter?
| |
| 12:30 | would that applyw ith fat vs thin?
| |
| 12:30 | <thunsucker> depends
| |
| 12:30 | are both installs using the same dhcp server?
| |
| 12:30 | <Roasted_> no
| |
| 12:31 | resolv.conf of the problematic server is identical
| |
| 12:31 | <thunsucker> are both dhcp servers setup the exact same way with the same options?
| |
| 12:31 | <Roasted_> to chroot to server
| |
| 12:31 | yes
| |
| 12:31 | <thunsucker> ahh probably fine then
| |
| 12:31 | i know with windows domains that is an issue that has always plagued me
| |
| 12:31 | <Roasted_> I wonder what it's issue is
| |
| 12:31 | what, /etc/resolv.conf?
| |
| 12:31 | <thunsucker> when you go to access the windows server do you use the name?
| |
| 12:31 | <Roasted_> I shouldn't need to use anything
| |
| 12:31 | it should see I am steve_jobs currently logged in
| |
| 12:31 | <thunsucker> i know but what i'm asking is this
| |
| 12:32 | <Roasted_> but it comes up with an authentication box
| |
| 12:32 | <thunsucker> when you try to access a windows share, do you access it by server-name.school.local
| |
| 12:32 | or just server-name
| |
| 12:32 | <Roasted_> smb://server/share
| |
| 12:32 | <thunsucker> again
| |
| 12:32 | is it smb://server/share or smb://server.domain.local/share
| |
| 12:32 | <Roasted_> smb://server/share
| |
| 12:33 | <thunsucker> just for giggles try smb://server.fqdn/share
| |
| 12:33 | so if your domain is steve.com use: smb//server-name.steve.com/share
| |
| 12:33 | <Roasted_> our server is named storage. are you asking smb://storage.domain/share?
| |
| 12:33 | ah
| |
| 12:33 | <thunsucker> whats your domain?
| |
| 12:33 | <Roasted_> solanco
| |
| 12:33 | <thunsucker> dot what?
| |
| 12:33 | .com ?
| |
| 12:33 | <Roasted_> local
| |
| 12:33 | <thunsucker> .local ?
| |
| 12:33 | k
| |
| 12:33 | try
| |
| 12:33 | smb://storage.solanco.local/share
| |
| 12:34 | <Roasted_> couldn't display
| |
| 12:34 | <thunsucker> so with that you get nothing
| |
| 12:34 | fun
| |
| 12:34 | does likewise sync your time?
| |
| 12:35 | <Roasted_> I'm not quite sure. I doubt it since it's off by about 4 hours
| |
| 12:35 | <thunsucker> or are you syncing your time on the ltsp server with the windows domain?
| |
| 12:35 | that could be it
| |
| 12:35 | <Roasted_> should I just set the clock manually on the server?
| |
| 12:35 | to as accurate as possible?
| |
| 12:35 | <thunsucker> worth a try
| |
| 12:35 | <Roasted_> brb
| |
| 12:35 | <thunsucker> but would be better to sync it
| |
| 12:35 | <Roasted_> good thought
| |
| 12:35 | <thunsucker> how do your computers get time
| |
| 12:36 | <Roasted_> the win boxes sync
| |
| 12:36 | <thunsucker> you have a dc setup to be a network time source?
| |
| 12:36 | <Roasted_> brb...
| |
| 12:36 | <thunsucker> however you're doing it, sycn your ltsp server's time with the domain
| |
| 12:36 | <abeehc> off by 4 hours haha
| |
| 12:37 | that's some time
| |
| 12:38 | thunsucker_ (~thunsucke@209.65.135.30) joined #ltsp. | |
| 12:40 | <Roasted_> thunsucker, you're on to something here
| |
| 12:40 | the server had correct time
| |
| 12:40 | so I logged in as myself to the server and hit a share
| |
| 12:40 | worked fine
| |
| 12:40 | so the CLIENTS time is the issue
| |
| 12:41 | thunsucker (~thunsucke@209.65.135.30) left irc: Ping timeout: 276 seconds | |
| 12:41 | <thunsucker_> :)
| |
| 12:41 | windows trys to find any reason not to hand shake
| |
| 12:41 | <Roasted_> well, it makes sense
| |
| 12:41 | I should have called that
| |
| 12:41 | but I suck and didn't thinjk of it
| |
| 12:42 | <thunsucker_> nahhh it's a process
| |
| 12:42 | i think i saw a guide on how to set your clients to use ntp back to the server
| |
| 12:42 | sec let me try and find it
| |
| 12:42 | <Roasted_> how do I log out from terminal?
| |
| 12:43 | <thunsucker_> Roasted_: https://help.ubuntu.com/community/UbuntuLTSP/InstallNTPServer
| |
| 12:43 | type: exit
| |
| 12:43 | <Roasted_> that just exits term
| |
| 12:43 | I want to log OUT of the session
| |
| 12:43 | <thunsucker_> oh you want to log out of yoru X session?
| |
| 12:43 | <Roasted_> my gizmo keeps crashing
| |
| 12:43 | under the local admin account on the clients
| |
| 12:43 | I ha dto log in as admin for root access to change time
| |
| 12:43 | Trixboxer (~Trixboxer@office.supportdepartment.net) left irc: Quit: "Achievement is not the end, its the beginning of new journey !!!" | |
| 12:43 | <Roasted_> and now Im stuck, haha
| |
| 12:43 | Il just reboot the box
| |
| 12:43 | <thunsucker_> this might work
| |
| 12:43 | gnome-session-save --kill --silent
| |
| 12:43 | <Roasted_> I have 29 others here
| |
| 12:44 | dangit
| |
| 12:44 | Im still getting the old time
| |
| 12:44 | <abeehc> kerberos is amazing sensitive to time float which is especialyl interesting in virtual environments
| |
| 12:44 | <thunsucker_> Roasted_: https://help.ubuntu.com/community/UbuntuLTSP/InstallNTPServer
| |
| 12:44 | <abeehc> i think greater than 4 minute difference to your dc and its game over
| |
| 12:44 | <Roasted_> reading over it now
| |
| 12:44 | I just wonder why I dont have this issue in the other school
| |
| 12:45 | <thunsucker_> because the time syncs correctly :)
| |
| 12:45 | <Roasted_> but that setup is identical to here
| |
| 12:45 | :(
| |
| 12:45 | <thunsucker_> whichmeans you got lucky
| |
| 12:45 | <abeehc> the likewise admin document is exhastive and reads well
| |
| 12:45 | <thunsucker_> and eventually the same issue will arise there
| |
| 12:45 | <Roasted_> I don't get this
| |
| 12:45 | if the server syncs to the windows box
| |
| 12:45 | why doesn't the client sync to the ltsp server
| |
| 12:45 | <abeehc> http://www.likewise.com/resources/documentation_library/manuals/open/likewise-open-guide.html#id2757412
| |
| 12:46 | <thunsucker_> Roasted_: remember all the processes turned off to make fat clients awesome? well that is probably one of them
| |
| 12:46 | <Roasted_> alkisg, can you comment on this?
| |
| 12:46 | <thunsucker_> i know it's been an issue for me in the past
| |
| 12:46 | xavierb (~xavier@lns-bzn-32-82-254-16-28.adsl.proxad.net) left irc: Read error: Operation timed out | |
| 12:46 | <thunsucker_> but the ntp guide is the basic solution
| |
| 12:47 | ntp = network time protocol, so use that guide and 15 minutes later the issue is resolved
| |
| 12:47 | <dan_l> alkisg: I've put a full description the problem we are working with regarding executing a program on the client from a remote host here http://pastebin.com/S8iwZdpv
| |
| 12:47 | <Roasted_> 15 minutes later eh?
| |
| 12:47 | its 347, Im outa here in 13 minutes
| |
| 12:47 | unacceptable! :P
| |
| 12:47 | <thunsucker_> lol
| |
| 12:47 | never know
| |
| 12:47 | if you're fast you can have it done faster
| |
| 12:47 | depends on how long it takes for your fat client image update
| |
| 12:47 | <Roasted_> ehh
| |
| 12:47 | it isn't TOO bad
| |
| 12:47 | we'll give it a shot
| |
| 12:47 | <thunsucker_> i know mine takes longer than 15 minutes but i'm in a test environment right now in vbox
| |
| 12:48 | <Roasted_> now wait
| |
| 12:48 | <abeehc> once you get ntp setup you should set TIMESERVER in lts.conf?
| |
| 12:48 | <thunsucker_> i'm getting ready to test the kiosk setup with firefox 4
| |
| 12:48 | <Roasted_> is this done on server or chroot?
| |
| 12:48 | <alkisg> dan_l: will look at it later on, you may also send it to the ltsp mailing list if you want
| |
| 12:49 | Roasted_: i didn't follow the conversation, what's the question?
| |
| 12:49 | <thunsucker_> Roasted_: read the guide the first command is done on the server
| |
| 12:49 | the rest are in the chroot
| |
| 12:49 | <Roasted_> alkisg, my clients don't have the same time as my server. I'm wondering why.
| |
| 12:49 | <thunsucker_> but that guide is kinda rough and assumes that you are an advanced user
| |
| 12:49 | <Roasted_> alkisg, fat clients, that is.
| |
| 12:50 | <alkisg> Roasted_: time sync is the same for thin and fat clients
| |
| 12:50 | <Roasted_> thunsucker_, I'm a terminal junkie, so I'll be all right. I didn't relaize it was split between server/chroot.
| |
| 12:50 | <alkisg> No difference
| |
| 12:50 | <Roasted_> alkisg, my fat clients are 4 hours behind my server.
| |
| 12:50 | <alkisg> Roasted_: ok. I'm just saying that if you boot them as thin, they'll be 4 hours behind too.
| |
| 12:50 | Is it a timezone problem or a timeserver problem?
| |
| 12:50 | <Roasted_> gotcha
| |
| 12:50 | <vagrantc> Gadi: ah, i see now why you were exploring bugs in getltscfg parsing
| |
| 12:51 | <Roasted_> uh
| |
| 12:51 | let me check
| |
| 12:51 | <abeehc> sounds like timezone
| |
| 12:51 | <vagrantc> Gadi: the evil hack to handle CRONTAB_*
| |
| 12:51 | <alkisg> If you change the hour of your server, do the clients get it (even with 4 hours difference?)
| |
| 12:51 | <thunsucker_> Roasted_: are the minutes the same?
| |
| 12:51 | <Roasted_> yes
| |
| 12:51 | <thunsucker_> if the minutes are the same most likely it's a timezone issue
| |
| 12:51 | <Roasted_> how's that work? If I set the time zone on 1 client user will it update all?
| |
| 12:51 | <alkisg> You set it in lts.conf
| |
| 12:52 | In which timezone are you?
| |
| 12:52 | <Roasted_> eastern
| |
| 12:52 | <thunsucker_> oh cool i didkn't know you could set it in lts.conf
| |
| 12:53 | thunsucker_ (~thunsucke@209.65.135.30) left irc: Quit: Leaving | |
| 12:53 | <Roasted_> alkisg, under default I take it?
| |
| 12:53 | TIME_ZONE=eastern?
| |
| 12:53 | thunsucker (~thunsucke@209.65.135.30) joined #ltsp. | |
| 12:55 | <alkisg> /usr/share/zoneinfo/right/Europe/Athens
| |
| 12:55 | man lts.conf
| |
| 12:55 | ==> TIMEZONE
| |
| 12:55 | E.g.: TIMEZONE=Europe/Athens
| |
| 12:55 | See your timezone from the dir above
| |
| 12:55 | /usr/share/zoneinfo/right
| |
| 12:56 | <Roasted_> haaa
| |
| 12:56 | <alkisg> EST ? Or EEST?
| |
| 12:56 | <Roasted_> if unset, the client just uses bios time
| |
| 12:56 | GUESS WHO HAS THE WRONG BIOS TIME
| |
| 12:56 | EST
| |
| 12:56 | <alkisg> So TIMEZONE=EST in lts.conf
| |
| 12:56 | <vagrantc> localtime is your enemy
| |
| 12:56 | <Roasted_> just for funsies I want to see if this works
| |
| 12:56 | with bios time
| |
| 12:56 | thunsucker (~thunsucke@209.65.135.30) left irc: Client Quit | |
| 12:57 | <Roasted_> alkisg, is this on chroot or server
| |
| 12:57 | <alkisg> In lts.conf
| |
| 12:58 | /var/lib/tftpboot/ltsp/i386/lts.conf
| |
| 12:58 | Server
| |
| 12:58 | <Roasted_> gotcha
| |
| 12:58 | yeah the bios change did nothing. :P Let me do this quick
| |
| 12:58 | under default or doesnt it matter?
| |
| 12:59 | thunsucker (~thunsucke@209.65.135.30) joined #ltsp. | |
| 12:59 | komunista (~slavko@adsl-195-098-015-248.dynamic.nextra.sk) joined #ltsp. | |
| 13:00 | thunsucker (~thunsucke@209.65.135.30) left irc: Client Quit | |
| 13:00 | xavierb (~xavier@lns-bzn-32-82-254-16-28.adsl.proxad.net) joined #ltsp. | |
| 13:02 | Nick change: evil_root -> zz_evil_root | |
| 13:08 | <Roasted_> uhm
| |
| 13:08 | still seems to be 11 am here
| |
| 13:11 | There's also no man entry for lts.conf
| |
| 13:11 | or else I'd read it
| |
| 13:11 | :/
| |
| 13:11 | <alkisg> http://manpages.ubuntu.com/lts.conf
| |
| 13:12 | Install the ltsp-docs package
| |
| 13:12 | Also pastebin your lts.conf
| |
| 13:12 | <Roasted_> not much to pastebin
| |
| 13:12 | its just
| |
| 13:12 | [Default]
| |
| 13:12 | TIMEZONE=EST
| |
| 13:12 | <alkisg> What about timeserver?
| |
| 13:12 | <Roasted_> I need both?
| |
| 13:12 | <alkisg> Yes
| |
| 13:23 | <stgraber> vagrantc: what just happened to ltsp-trunk ?
| |
| 13:24 | vagrantc: I got a notification from LP that you apparently renamed ltsp-trunk to ltsp-trunk-old
| |
| 13:24 | <Roasted_> alkisg, this is still out of whack.
| |
| 13:24 | <vagrantc> stgraber: i'm trying to upgrade the format
| |
| 13:24 | <Roasted_> I'm seeing 1122 am, 324 pm, etc
| |
| 13:24 | and yet I have timezone and timeserver set.
| |
| 13:24 | <vagrantc> stgraber: and it seems easier to just push a new branch than upgrade it remotely
| |
| 13:24 | <stgraber> vagrantc: ah, ok
| |
| 13:24 | <Roasted_> I thought the clients would just look at the server for time, which considering the server is syncing properly.. would make sense :/
| |
| 13:27 | <stgraber> vagrantc: ok, seems to work fine. I just did a "bzr upgrade" locally, then did a pull and it got some remaining changes from ltsp-trunk
| |
| 13:27 | <vagrantc> hopefully it doesn't send an email for each of 1811 commits
| |
| 13:28 | stgraber: good.
| |
| 13:28 | <stgraber> vagrantc: not sure what happens if soemone doesn't "bzr upgrade" locally first. Might be worth sending something to ltsp-devel
| |
| 13:28 | <vagrantc> stgraber: yeah.
| |
| 13:30 | <Guest94358> I'm trying to get a scanner working that is connected to a thin client locally (not the server)
| |
| 13:32 | I've tried applying this guide from within the chroot and updating my image http://my.oltrans.org/en/ubuntu/faqs/44-ltsp-and-thin-client-technology/88-access-scanner-on-ltsp-server-from-thin-client.html
| |
| 13:32 | Nick change: zz_evil_root -> evil_root | |
| 13:32 | <Guest94358> but in order to get it to work I have to restart udev and re-plug in the scanner
| |
| 13:33 | mgariepy (~mgariepy@ubuntu/member/mgariepy) left irc: Remote host closed the connection | |
| 13:33 | <vagrantc> Guest94358: that's probably because udev didn't know about the new scanner group
| |
| 13:33 | Guest94358: so it needed to be restarted in order to set the correct permissions
| |
| 13:33 | a similar bug happens with fuse
| |
| 13:33 | udev doesn't check for new group names
| |
| 13:34 | <Guest94358> what I"ve done though
| |
| 13:35 | is set the rule to change group ownership to the tsuers group
| |
| 13:35 | not sure if that's a legit way to do it, but that's the route I took
| |
| 13:39 | so yeah.. any way to fix this?
| |
| 13:40 | <Roasted_> I have ntpdate already installed on chroot. I have my timezone and timeserver entry set up properly. Yet I am STILL not getting the proper time on these fat clients. Is there hope?
| |
| 13:41 | <alkisg> Paste your lts.conf. Do the clients have access to the timeserver?
| |
| 13:41 | <abeehc> does that lts.conf param TIMESERVER require ntpdate package in the chroot?
| |
| 13:42 | <Roasted_> [Default]
| |
| 13:42 | TIMEZONE=EST
| |
| 13:42 | TIMESERVER=10.52.1.8
| |
| 13:42 | <alkisg> Roasted_: and you have a time server running in 10.52.1.8?
| |
| 13:42 | <Roasted_> yep
| |
| 13:42 | befrenchy (8e2e1f1e@gateway/web/freenode/ip.142.46.31.30) left irc: Ping timeout: 252 seconds | |
| 13:42 | <alkisg> Can you get a local SCREEN_02=shell terminal on a thin client?
| |
| 13:42 | To try some commands as root there?
| |
| 13:43 | <Roasted_> yeah, I'm in as local admin now on a client
| |
| 13:43 | <alkisg> [Default]
| |
| 13:43 | Ah ok
| |
| 13:43 | How? Ssh?
| |
| 13:43 | <Roasted_> I'm physically in a lab
| |
| 13:43 | <alkisg> Or unlocked password?
| |
| 13:43 | *root
| |
| 13:43 | <Roasted_> I'm SSH'd into the server on my laptop
| |
| 13:43 | I just logged into the client as the admin account
| |
| 13:43 | who has root
| |
| 13:44 | <alkisg> So `hostname` is ltsp123?
| |
| 13:44 | Or hostname == the server?
| |
| 13:44 | <Roasted_> ltsp2870
| |
| 13:44 | <alkisg> OK
| |
| 13:45 | getltscfg -a
| |
| 13:45 | What does this tell you?
| |
| 13:45 | <Roasted_> TIMEZONE="EST" TIMESERVER="10.52.1.8" export TIMEZONE export TIMESERVER
| |
| 13:45 | <alkisg> Nice. Next: eval $(getltscfg -a)
| |
| 13:46 | Then: date
| |
| 13:46 | Then: ntpdate $TIMESERVER; hwclock --systohc --${HWCLOCK:-"utc"} --noadjfile
| |
| 13:46 | Then: date
| |
| 13:46 | Output?
| |
| 13:46 | <Roasted_> the eval command does nothing
| |
| 13:46 | <alkisg> Yes, no output there
| |
| 13:46 | <Roasted_> k
| |
| 13:46 | cody_ (~cody@216-102-79-229.scoe.org) joined #ltsp. | |
| 13:47 | <Roasted_> okay
| |
| 13:47 | at the ntpdate command
| |
| 13:47 | Command ntpdate is available in usr bin ntdate. could not be located blah blah
| |
| 13:47 | should I be CD'd into it
| |
| 13:47 | <alkisg> whoami
| |
| 13:47 | What does this tell you?
| |
| 13:47 | <Guest94358> vagrantc: are you familiar with udev enough to help me with a workaround on this?
| |
| 13:47 | <Roasted_> administrator
| |
| 13:48 | <alkisg> Can you get a root shell?
| |
| 13:48 | So that whoami returns root?
| |
| 13:48 | <cody_> I've installed a brand new LTSP server, most of the clients boot just fine, only our older Wyse terminals are having an issue. Basically the client just sits on the Ubuntu screen the "dots" are moving but nothing happens. When I hit esacpe the only message pertains to DHCP lease...I remember an issue like this before where I inserted nomodset into pxelinux.cfg/default but this does not seem to be working for this issue...any ideas?
| |
| 13:48 | <alkisg> Roasted_: `sudo -i` or whatever else works for you
| |
| 13:48 | <Roasted_> I'm a little confused
| |
| 13:48 | when I do sudo -i
| |
| 13:48 | it errors out on the pw
| |
| 13:49 | yet I'm typing it right...
| |
| 13:49 | <alkisg> I'm confused on how you got ssh to work there
| |
| 13:49 | <Roasted_> sudo -i works on my ssh tho
| |
| 13:49 | <alkisg> You modified the chroot to join your domain?
| |
| 13:49 | <Roasted_> no
| |
| 13:49 | I just used likewise open to join the domain on the server
| |
| 13:49 | then I built the fat image
| |
| 13:49 | er
| |
| 13:49 | join the server on the domain
| |
| 13:50 | <alkisg> The fat image has only 1 user. Root. And it doesn't have a password, i.e. it's locked. So you shouldn't be able to login on the client.
| |
| 13:50 | What did you do to be able to login to the fat client *locally* ?
| |
| 13:50 | <Roasted_> when I installed edubuntu, my user I created was administrator
| |
| 13:50 | that's the user I use at the server
| |
| 13:50 | and that's the user I'm using now, on the client
| |
| 13:50 | <alkisg> That's on the server, not on the chroot
| |
| 13:50 | <Roasted_> when I'm in the chroot I just sudo chroot /opt/ltsp/i386
| |
| 13:50 | and I do my thing
| |
| 13:50 | <alkisg> OK let's go back
| |
| 13:51 | That thing about SCREEN_02=shell that I said?
| |
| 13:51 | You need it
| |
| 13:51 | So. open lts.conf
| |
| 13:51 | Under [default], add:
| |
| 13:51 | SCREEN_02=shell
| |
| 13:51 | SCREEN_07=ldm
| |
| 13:51 | Reboot client. Press Alt+ctrl+f2. Then we continue :)
| |
| 13:52 | <Roasted_> I thought setting time on clients would be easy.
| |
| 13:52 | :/
| |
| 13:52 | <alkisg> It is. Troubleshooting it when it goes wrong, isn't.
| |
| 13:52 | <Roasted_> I wonder what could be wrong with it
| |
| 13:52 | BIOS time is right
| |
| 13:52 | <Gadi> guys, this was just addressed on the mailing list, too
| |
| 13:52 | <Roasted_> server time is right
| |
| 13:53 | rebooting client...
| |
| 13:53 | okay
| |
| 13:53 | alt ctrl f2'd in
| |
| 13:53 | <alkisg> Roasted_: are you on 10.10?
| |
| 13:53 | <Roasted_> yes
| |
| 13:53 | <alkisg> Hm ok lets see if something's regressed there
| |
| 13:54 | Again, the commands above:
| |
| 13:54 | eval $(getltscfg -a)
| |
| 13:54 | date
| |
| 13:54 | ntpdate $TIMESERVER; hwclock --systohc --${HWCLOCK:-"utc"} --noadjfile
| |
| 13:54 | date
| |
| 13:55 | cody_: try removing "quiet splash" from /var/lib/tftpboot/ltsp/i386/pxelinux.cfg/default so that you better see any error messages
| |
| 13:55 | <Roasted_> okay
| |
| 13:55 | its saying 15:55
| |
| 13:55 | thats off by 1 hour
| |
| 13:55 | <Gadi> yknow, we just changed the clocks
| |
| 13:55 | :)
| |
| 13:55 | <alkisg> On both times the date said 15:55 ?
| |
| 13:56 | Before and after ntpdate?
| |
| 13:56 | <Roasted_> no
| |
| 13:56 | it did change
| |
| 13:56 | 11:53 to 15:55
| |
| 13:56 | <Gadi> Roasted_: we aren't in EST
| |
| 13:56 | it is daylight savings time now
| |
| 13:56 | :)
| |
| 13:56 | <cody_> alkisg: Ok...did that and the boot stops at Being: Mounting root file system... ... Being: Running /scripts/nfs-top ... Done.
| |
| 13:57 | <Roasted_> alkisg, so.. are we good now?
| |
| 13:57 | <Gadi> use: America/New_York
| |
| 13:57 | <alkisg> Roasted_: no idea, is the time correct?
| |
| 13:57 | <Roasted_> Gadi, for what
| |
| 13:57 | <Gadi> for TIMEZONE
| |
| 13:57 | <Roasted_> alkisg, it's 4:57 here... so that's 16:57 in military time right?
| |
| 13:57 | this thing says 15:55
| |
| 13:57 | so uh
| |
| 13:57 | no
| |
| 13:57 | wait
| |
| 13:57 | yeah... it's off by 1
| |
| 13:57 | <Gadi> nobody listens to me
| |
| 13:57 | <alkisg> Roasted_: read what gadi said
| |
| 13:57 | <Gadi> :)
| |
| 13:58 | <Roasted_> so I DONT use TIMEZONE=EST
| |
| 13:58 | ?
| |
| 13:58 | <abeehc> that's right you are not in EST
| |
| 13:58 | <Gadi> not when we are in daylight savings time
| |
| 13:58 | :)
| |
| 13:58 | didn't you notice?
| |
| 13:58 | say, about 2 weeks ago
| |
| 13:58 | or were you always late?
| |
| 13:58 | <abeehc> you'd think your watch would even be wrong at this point
| |
| 13:58 | <Roasted_> I don't wear a watch
| |
| 13:58 | my cell phone does that thing automagically
| |
| 13:59 | <abeehc> im sad
| |
| 13:59 | <Gadi> as will your thin client
| |
| 13:59 | <Roasted_> Gadi, if I have it set to america/new_york will it change automatically?
| |
| 13:59 | <alkisg> cody_: try this: at the end of pxelinux.cfg/default, insert a new line, and put in it this:
| |
| 13:59 | IPAPPEND 3
| |
| 13:59 | Then reboot your client
| |
| 13:59 | <Gadi> if you use America/New_York
| |
| 13:59 | yup
| |
| 13:59 | <Roasted_> gadi
| |
| 13:59 | where were you earlier
| |
| 13:59 | <Gadi> preoccupied with winding my clock
| |
| 13:59 | :P
| |
| 13:59 | <Roasted_> gadi
| |
| 13:59 | I just had a legit "LOL"
| |
| 13:59 | <alkisg> Roasted_: it seems that you still have another problem, aside from timezone
| |
| 13:59 | <Roasted_> thought you should know
| |
| 14:00 | <Gadi> :D
| |
| 14:00 | <Roasted_> how so alkisg ?
| |
| 14:00 | <alkisg> I.e. that 10.10 wasn't calling ntpdate
| |
| 14:00 | <abeehc> you were four hours off earlier somehow
| |
| 14:00 | <alkisg> It only synced when you called it manually
| |
| 14:00 | <Gadi> that was what I was just discussing on the mailing list
| |
| 14:00 | <Roasted_> alkisg, keep in mind it was only the clients. the server has been right all along if I physically log into it
| |
| 14:00 | ogra_ (~ogra@p5098ed03.dip0.t-ipconnect.de) left irc: Read error: Operation timed out | |
| 14:00 | <Gadi> we have to stop setting the time in the background
| |
| 14:00 | <Roasted_> alkisg, just one of those "for what it's worth" things
| |
| 14:01 | oh man
| |
| 14:01 | <alkisg> Gadi, you think that's the problem? Why would it fail in the backgruond?
| |
| 14:01 | <Roasted_> these clients are 17:00
| |
| 14:01 | that's 5:00 PM
| |
| 14:01 | <Gadi> but, I suppose mailing lists are old school by now
| |
| 14:01 | <Roasted_> it's 5:00 PM
| |
| 14:01 | <Gadi> 5PM here, too
| |
| 14:01 | how about that?
| |
| 14:02 | <Roasted_> now why is it still requiring me to authenticate to the file server...
| |
| 14:02 | I'm about to just redo this flippin server
| |
| 14:02 | I thought fixing the time would make it work, since it works on the server and the server has laways been on the correct time
| |
| 14:03 | ogra_ (~ogra@p5098ed03.dip0.t-ipconnect.de) joined #ltsp. | |
| 14:03 | <cody_> alkisg: that worked! WIll that affect the clients that are booting?
| |
| 14:04 | <alkisg> cody_: distro/version?
| |
| 14:04 | And dhcp server?
| |
| 14:04 | <cody_> ubuntu/10.04.2
| |
| 14:04 | default dhcp server installed with ltsp-server
| |
| 14:05 | <alkisg> There's a bug in udhcp about the client identifier etc, was solved in maverick
| |
| 14:05 | <vagrantc> Guest94358: are you talking about udev on the server, or udev on the thin client?
| |
| 14:06 | Guest94358: you should just have to reboot once for the changes to take effect.
| |
| 14:06 | <alkisg> cody_: IPAPPEND 3 means to reuse the IP the clients get in the pxe stage, so it works around that bug
| |
| 14:06 | cody_: but you should'nt be affected if you use the default dhcpd, so I'm not sure why the client stack there, it might be a different udhcpc bug
| |
| 14:06 | <cody_> alkisg: Great! Weird that issue just surfaced though
| |
| 14:07 | <alkisg> Are you using ltsp-cluster? If not, then IPAPPEND 3 shouldn't cause you any problems
| |
| 14:07 | <cody_> Nope, no ltsp-cluster
| |
| 14:08 | Thanks for the help! I'm eternally grateful. People helping people...it's powerful stuff
| |
| 14:08 | cody_ (~cody@216-102-79-229.scoe.org) left irc: Quit: cody_ | |
| 14:12 | <Gadi> so, what dya think, 9 crontab lines enough for everybody?
| |
| 14:13 | <vagrantc> could always have a hook to just copy a file into place, no?
| |
| 14:13 | if you needed more than that
| |
| 14:14 | <Gadi> well, I'm going to rehash the code and I am not cycling through 99 times
| |
| 14:14 | :)
| |
| 14:15 | <alkisg> Gadi, grepping $(getltscfg -a) with case CRONTAB[0-9][0-9] ?
| |
| 14:15 | Wouln't that allow for 100 crontab lines without needless looping?
| |
| 14:15 | <Gadi> its already in the env at that point
| |
| 14:15 | and grepping and seding takes precious time
| |
| 14:16 | I could case out env, I suppose
| |
| 14:18 | Roasted_ (~jason@206.82.22.190) left irc: Ping timeout: 255 seconds | |
| 14:19 | _UsUrPeR__ (~jsass@fw.acurrus.com) joined #ltsp. | |
| 14:19 | _Kicer_ (~Kicer86@host-5db0eeee.sileman.net.pl) joined #ltsp. | |
| 14:19 | _UsUrPeR_ (~jsass@fw.acurrus.com) left irc: Read error: Operation timed out | |
| 14:20 | ogra_ (~ogra@p5098ed03.dip0.t-ipconnect.de) left irc: Read error: Operation timed out | |
| 14:20 | Kicer86 (~Kicer86@host-5db0eeee.sileman.net.pl) left irc: Read error: Operation timed out | |
| 14:20 | * alkisg would just do `env | egrep ^CRONTAB[0-9][0-9] >> /etc/cronwhatever. It does take 0.001 secs though :) | |
| 14:20 | ogra_ (~ogra@p5098ed03.dip0.t-ipconnect.de) joined #ltsp. | |
| 14:20 | <Gadi> hehe
| |
| 14:20 | fine
| |
| 14:20 | :P
| |
| 14:22 | <alkisg> Eval is about 100 times faster. But it's just one call instead of 100 calls to eval ;)
| |
| 14:22 | And I didn't count the loop required for eval
| |
| 14:25 | Gadi: env | sed -n 's/^CRONTAB_[0-9][0-9]=//p'
| |
| 14:28 | <Gadi> done
| |
| 14:29 | have a good weekend, all
| |
| 14:29 | Gadi (romm@ool-18bbe47a.static.optonline.net) left #ltsp. | |
| 14:29 | <alkisg> bb
| |
| 14:31 | _Kicer_ (~Kicer86@host-5db0eeee.sileman.net.pl) left irc: Quit: KVIrc Insomnia 4.0.2, revision: 4740, sources date: 20100627, built on: 2010-08-08 18:29:00 UTC http://www.kvirc.net/ | |
| 14:36 | Roasted (~jason@unaffiliated/roasted) joined #ltsp. | |
| 14:39 | <alkisg> Hmm that `set +f` should probably go a little above, in line 59
| |
| 14:39 | mistik1 (mistik1@unaffiliated/mistik1) left irc: Read error: Connection reset by peer | |
| 14:45 | <alkisg> dan_l: here?
| |
| 14:45 | mistik1 (mistik1@unaffiliated/mistik1) joined #ltsp. | |
| 14:47 | jtwood (~jtwood@74.81.111.2) left irc: Ping timeout: 240 seconds | |
| 15:19 | <dan_l> alkisg: Someonne offered me cake - I'm back now.
| |
| 15:19 | <alkisg> Hey
| |
| 15:19 | <dan_l> I'll send you some cake
| |
| 15:20 | <alkisg> I got a little confused by your description. Are you using something like the kiosk mode, where everything runs locally?
| |
| 15:20 | (thanks for the cake offer :))
| |
| 15:21 | <dan_l> hmm not sure what you mean by kiosk mode but I think no.
| |
| 15:21 | <alkisg> Or complete fat session?
| |
| 15:21 | Are you using thin or fat clients?
| |
| 15:21 | I.e. are the processes running on the server, or on the client locally?
| |
| 15:21 | <dan_l> We are just using fatclient ltsp and faking a login so we can do authentication against our library catalog
| |
| 15:22 | <alkisg> Where is the user inserting his card? In a device on the client?
| |
| 15:22 | <dan_l> they type it into a login window on the client
| |
| 15:23 | <alkisg> Ah so you don't have any hardware for reading cards
| |
| 15:23 | They just enter a pin?
| |
| 15:23 | <dan_l> No they enter their library card number and a pin to gain access
| |
| 15:24 | <alkisg> OK. Do you have a separate authentication machine? Or is it just <fat client> and <ltsp server>?
| |
| 15:25 | <dan_l> yes. the auth server sends a query to the catalog server. If the card/pin are valid the login window is replaced with a confirmation window. That is then replaced with a countdown timer on the desktop
| |
| 15:26 | <alkisg> Do the users have persistent data on their homes?
| |
| 15:26 | Or is everyone a guest with no persistent files?
| |
| 15:27 | <dan_l> The application runs on the server. it is called using ltsp-remoteapps. A connection is made with web app on the auth server which opens the login window on the client.
| |
| 15:27 | There no persistent data in the home dir. The users data must be purged at the end of the session.
| |
| 15:28 | <alkisg> OK. So.
| |
| 15:28 | 1) Why go from fat client to ltsp server to catalog server to authenticate, instead of getting from fat client directly to catalog server? I.e. put the web app in the chroot?
| |
| 15:28 | 2) Why not have a local user and a local home then? Why bother with sshfs etc?
| |
| 15:29 | And about the cleanup process, if I got your problem right, you wouldn't need many of that if you had a local user
| |
| 15:29 | <dan_l> Each branch library has it's own ltsp server. Login time is coordinated across all locations so we need a central database to store the login and figure the elapsed time
| |
| 15:30 | <alkisg> That's what the catalog server does, right?
| |
| 15:30 | But why put the ltsp server inbetween?
| |
| 15:30 | (for authentication, i mean)
| |
| 15:30 | <dan_l> No. We only access teh catalog server to verify tath the library card is valid.
| |
| 15:31 | <alkisg> So the web app in the ltsp server logs the logins?
| |
| 15:31 | <dan_l> Teh catalog server is also an Dec Alpha running VMS so not anxious to run the app there
| |
| 15:32 | <alkisg> I meant to run the web app locally on the fat client
| |
| 15:32 | vmlintu (~vmlintu@nblzone-240-143.nblnetworks.fi) left irc: Ping timeout: 276 seconds | |
| 15:32 | <dan_l> Thinking - that might be a possibility
| |
| 15:33 | I would mean loading the Ruby libraries on the fatclient image though
| |
| 15:33 | <alkisg> Or at least, if it's a web app, to have it run on the browser on the fat client like http://server/path/to/web/app
| |
| 15:34 | <dan_l> The browers is not availlable at login time.
| |
| 15:34 | <alkisg> You can put it there, it's easy, e.g. a restricted web app with prism
| |
| 15:35 | Tell me about the second part. If you don't want to store anything permanately, why have /home on the server?
| |
| 15:36 | <dan_l> You are making me rethink this though - that's good. Perhaps this could function ore like a captive portal. I would have to rethink the lock down
| |
| 15:37 | <alkisg> About the lock down. If you only want to be able to run 2-3 specific apps with no DE, are you sure you want gnome?
| |
| 15:37 | <dan_l> We allow users to dwonload files and use other apps like openoffice
| |
| 15:37 | <alkisg> The downloaded files get deleted when they log out, right?
| |
| 15:38 | <dan_l> gnome - hate it. I could go back to icewm or fluxbox and that would probably resolve all my issues.
| |
| 15:38 | <xavierb> use window maker
| |
| 15:38 | <dan_l> Yes - the homedir get a complete refresh.
| |
| 15:38 | <HrdwrBoB> alkisg: we do tht
| |
| 15:38 | we run 3 apps + gnome
| |
| 15:38 | <dan_l> My hesitance about dropping gnome is the level of integration with ltsp.
| |
| 15:39 | I will be leaving this job at the end of the year and want people to feel comfortable supporting ltsp
| |
| 15:40 | Just to simplify a bit ...
| |
| 15:42 | When the user logs off or is logged off, the auth server calls a script on the ltsp server. We acn put anything there. I was hoping that we could call gnome-session-save on the client and then rsync the associated homedir
| |
| 15:43 | otavio_ (~otavio@debian/developer/otavio) joined #ltsp. | |
| 15:44 | otavio (~otavio@debian/developer/otavio) left irc: Ping timeout: 240 seconds | |
| 15:44 | <alkisg> OK, but that's not simpler, it's more complicated. You could be running most things on the fat client so you wouldn't need any synching/refreshing/remote running
| |
| 15:45 | Is that script ran as the current user? Or it's ran from some daemon?
| |
| 15:46 | <dan_l> Yes - I get that. The app on the ltsp server requires gtk and ruby. I have to assess the increase to the image size if I run these on the fatclient.
| |
| 15:47 | The script is running as root on the ltsp server.
| |
| 15:49 | xavierb (xavier@lns-bzn-32-82-254-16-28.adsl.proxad.net) left #ltsp ("Konversation terminated!"). | |
| 15:53 | <alkisg> The login takes place and a count down timer is placed on the users desktop. ==> is that a remote app?
| |
| 15:54 | <dan_l> alkisg: jsut did a ps. the process is owned by the fatclient not root. I assume because it is started using ltsp-remoteapps
| |
| 15:54 | <alkisg> If it's started with ltsp-remoteapps, then it can use ltsp-localapps to run something locally
| |
| 15:54 | <dan_l> The countdown timer is running on the ltsp server - started by ltsp-remoteapps
| |
| 15:55 | <alkisg> Wouldn't it be easier if it ran locally? OK, not the webapp, just the countdown?
| |
| 15:55 | Also, what prevents the user from killing that countdown app?
| |
| 15:57 | <dan_l> the timer is always on top an dcan't be closed or killed by the user since they no cli available
| |
| 15:58 | <alkisg> Instead of running on the server, and when it finishes it would call ltsp-localapps gnome-session-logout etc,
| |
| 15:58 | it could call `ltsp-localapps countdown` at the beginning, and run locally, and just call gnome-session-logout at the end
| |
| 15:59 | That would also save network bandwidth
| |
| 15:59 | I think your setup needs some redesign though, with local /home, local user etc
| |
| 15:59 | Time for bed... night all, cu tomorrow
| |
| 15:59 | Faithful (~Faithful@202.189.73.144) joined #ltsp. | |
| 15:59 | <dan_l> Yes. Running the timer locally makes sense that way. I will have to rethink how I message teh auth server to knwo that a logout has occured.
| |
| 16:00 | Alksig: Many thanks for you time. I will rethink the timer portion - makes sense.
| |
| 16:01 | <alkisg> Also think about a local home and local user - if we're both around tomorrow I'll tell you how to make them local and you can avoid the remote cleanup+syncing
| |
| 16:01 | bb
| |
| 16:01 | alkisg (~alkisg@ubuntu/member/alkisg) left irc: Quit: Leaving. | |
| 16:03 | Nick change: otavio_ -> otavio | |
| 16:28 | vvinet (~vince@2001:470:8829:1000:227:eff:fe25:ee64) left irc: Read error: Operation timed out | |
| 16:40 | staffencasa (~staffenca@128-193-144-246.oregonstate.edu) left irc: Read error: Connection reset by peer | |
| 16:41 | otavio (~otavio@debian/developer/otavio) left irc: Quit: leaving | |
| 16:45 | vvinet (~vince@gw-sherb.rlnx.com) joined #ltsp. | |
| 16:47 | MorningSon (~MorningSo@cpe-70-114-21-95.satx.res.rr.com) left irc: Quit: WeeChat 0.3.0 | |
| 16:49 | MorningSon (~MorningSo@cpe-70-114-21-95.satx.res.rr.com) joined #ltsp. | |
| 16:53 | drdata (~drdata@tw2.udallas.edu) left irc: Ping timeout: 255 seconds | |
| 17:03 | bobby_C (~bobby@188.20.161.210) left irc: Read error: Operation timed out | |
| 17:09 | Faithful (~Faithful@202.189.73.144) left irc: Ping timeout: 250 seconds | |
| 17:11 | Faithful (~Faithful@202.189.73.144) joined #ltsp. | |
| 17:30 | komunista (~slavko@adsl-195-098-015-248.dynamic.nextra.sk) left irc: Quit: Leaving. | |
| 17:42 | chupacabra (~chupacabr@cpe-70-112-10-45.austin.res.rr.com) joined #ltsp. | |
| 18:10 | _LoveStorm (storm@thepcrepair.info) left irc: Ping timeout: 248 seconds | |
| 18:11 | _LoveStorm (storm@thepcrepair.info) joined #ltsp. | |
| 18:24 | Nick change: evil_root -> zz_evil_root | |
| 19:16 | Adriannom (~adrian@cpc10-cwma7-2-0-cust15.7-3.cable.virginmedia.com) left irc: Quit: Ex-Chat | |
| 19:18 | nothingman (~joshua@32.138.178.148) joined #ltsp. | |
| 19:18 | <nothingman> hi, all
| |
| 19:20 | I was just wondering what methods are best to use for accessing a squashfs system image from a kernel loaded via pxe
| |
| 19:55 | Nick change: zz_evil_root -> evil_root | |
| 19:57 | Nick change: evil_root -> zz_evil_root | |
| 20:08 | nothingman (~joshua@32.138.178.148) left irc: Ping timeout: 264 seconds | |
| 20:08 | nothingman (~joshua@32.143.124.183) joined #ltsp. | |
| 20:58 | nothingman (~joshua@32.143.124.183) left irc: Ping timeout: 248 seconds | |
| 21:09 | Roasted (~jason@unaffiliated/roasted) left irc: Ping timeout: 246 seconds | |
| 21:18 | MorningSon (~MorningSo@cpe-70-114-21-95.satx.res.rr.com) left irc: Quit: WeeChat 0.3.0 | |
| 21:27 | alkisg (~alkisg@ubuntu/member/alkisg) joined #ltsp. | |
| 21:28 | alkisg (~alkisg@ubuntu/member/alkisg) left irc: Client Quit | |
| 21:28 | alkisg (~alkisg@ubuntu/member/alkisg) joined #ltsp. | |
| 22:25 | Roasted (~jason@unaffiliated/roasted) joined #ltsp. | |
| 22:30 | Faithful (~Faithful@202.189.73.144) left irc: Ping timeout: 246 seconds | |
| 23:05 | Kicer86 (~Kicer86@host-5db0eeee.sileman.net.pl) joined #ltsp. | |
| 23:49 | vagrantc (~vagrant@c-24-21-191-93.hsd1.or.comcast.net) left irc: Quit: leaving | |
| 00:00 | --- Sat Mar 26 2011 | |