IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 5 October 2017 (all times are UTC)
| 00:17 | Statler has left IRC (Statler!~Georg@mail.lohn24.de, Ping timeout: 240 seconds) | |
| 00:17 | Statler has joined IRC (Statler!~Georg@gwrz3.lohn24.de) | |
| 00:22 | Statler has left IRC (Statler!~Georg@gwrz3.lohn24.de, Remote host closed the connection) | |
| 01:51 | adrianor1 has joined IRC (adrianor1!~adrianorg@189.58.237.30.dynamic.adsl.gvt.net.br) | |
| 01:54 | adrianorg has left IRC (adrianorg!~adrianorg@177.156.229.204, Ping timeout: 248 seconds) | |
| 02:00 | ABIX_Adamj_ has left IRC (ABIX_Adamj_!~quassel@abix-vps.abix.info.pl, Quit: No Ping reply in 180 seconds.) | |
| 02:02 | ABIX_Adamj has joined IRC (ABIX_Adamj!~quassel@2a01:7c8:aab0:3d9:5054:ff:fed3:602b) | |
| 05:16 | Statler has joined IRC (Statler!~Georg@p579FE60B.dip0.t-ipconnect.de) | |
| 05:18 | ricotz has joined IRC (ricotz!~ricotz@p5B2A8714.dip0.t-ipconnect.de) | |
| 05:18 | ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz) | |
| 05:43 | mikkel has joined IRC (mikkel!~mikkel@mail.dlvs.dk) | |
| 06:44 | ogra_ has left IRC (ogra_!~ogra_@p5098ed03.dip0.t-ipconnect.de, Ping timeout: 258 seconds) | |
| 06:48 | ogra_ has joined IRC (ogra_!~ogra_@p5098ed03.dip0.t-ipconnect.de) | |
| 07:05 | eu^78-13-190stat has joined IRC (eu^78-13-190stat!4e570dbe@gateway/web/freenode/ip.78.87.13.190) | |
| 07:08 | eu^78-13-190stat has left IRC (eu^78-13-190stat!4e570dbe@gateway/web/freenode/ip.78.87.13.190, Client Quit) | |
| 08:05 | Statler has left IRC (Statler!~Georg@p579FE60B.dip0.t-ipconnect.de, Remote host closed the connection) | |
| 08:26 | wim1 has joined IRC (wim1!~Thunderbi@WEGC203033.UNI-GRAZ.AT) | |
| 08:37 | Statler has joined IRC (Statler!~Georg@gwrz3.lohn24.de) | |
| 09:40 | GodFather has joined IRC (GodFather!~rcc@47.33.250.142) | |
| 09:44 | GodFather has left IRC (GodFather!~rcc@47.33.250.142, Read error: Connection reset by peer) | |
| 10:00 | <sebd> Hello, we found out that our debian 9 LTSP amd64 fat client chroot got 2 problems : /etc/timezone was not created and the ssh service was not registered with systemd. Purging and reinstalling openssh-server fixed this.
| |
| 10:01 | I am not sure whether this is a problem with ltsp-build-client or our specific server setup.
| |
| 10:57 | <alkisg> Not many people use ltsp-build-client nowadays, maybe try the mailing list or a bug report in debian...
| |
| 11:06 | adrianorg has joined IRC (adrianorg!~adrianorg@179.187.26.60.dynamic.adsl.gvt.net.br) | |
| 11:08 | adrianor1 has left IRC (adrianor1!~adrianorg@189.58.237.30.dynamic.adsl.gvt.net.br, Ping timeout: 260 seconds) | |
| 11:25 | TFC has left IRC (TFC!b9416d65@gateway/web/freenode/ip.185.65.109.101, Ping timeout: 260 seconds) | |
| 12:18 | lucascastro has joined IRC (lucascastro!~lucas@201.182.221.14) | |
| 12:24 | <sebd> what do people use ? A standard debootstrap in a chroot then installation of package ltsp-client ?
| |
| 12:29 | <alkisg> !ltsp-pnp | echo sebd
| |
| 12:29 | <ltsp> sebd ltsp-pnp: ltsp-pnp is the recommented method to install and maintain LTSP for "usual" setups. Since it doesn't involve chroots, it requires little to no command line to maintain it. It automatically supports both thin and fat ltsp clients. https://help.ubuntu.com/community/UbuntuLTSP/ltsp-pnp
| |
| 12:29 | <alkisg> !ltsp-manager | echo sebd
| |
| 12:29 | <ltsp> sebd ltsp-manager: LTSP Manager is a GUI tool that makes LTSP maintenance easy. It's the recommended way to install LTSP in common setups. More info: http://wiki.ltsp.org/wiki/Ltsp-manager
| |
| 12:30 | <alkisg> So no chroots at all
| |
| 12:30 | The server root is a template for the clients
| |
| 12:32 | There are so many methods to create chroots and VMs that there's really no point for ltsp to maintain an additional one
| |
| 12:33 | And, we have an `ltsp-update-image --cleanup` phase and an "init-ltsp.d" boot phase to configure ltsp, so no need for additional ltsp-build-client handlers...
| |
| 12:53 | <sebd> I am so oldschool
| |
| 13:42 | mikkel has left IRC (mikkel!~mikkel@mail.dlvs.dk, Quit: Leaving) | |
| 13:53 | ben_roose has joined IRC (ben_roose!~roose@roose.cs.wichita.edu) | |
| 14:28 | mmarconm has joined IRC (mmarconm!~mmarconm@191.217.95.61) | |
| 16:03 | <bennabiy> alkisg: Do you have any experience with LTSP server run from a headless server with a graphics card installed, but not hooked up to a monitor ?
| |
| 16:03 | <alkisg> bennabiy: what's the difference then? :)
| |
| 16:03 | I had servers that teachers removed the monitors, sure
| |
| 16:03 | wim1 has left IRC (wim1!~Thunderbi@WEGC203033.UNI-GRAZ.AT, Quit: wim1) | |
| 16:03 | <bennabiy> alkisg: well, I need the graphics processing power, but am not going to have someone running from the server directly
| |
| 16:04 | <alkisg> I'm not sure what you mean with "graphics processing power"
| |
| 16:04 | How is this going to reach the clients?
| |
| 16:04 | <bennabiy> for thin clients
| |
| 16:04 | <alkisg> opencl apps?
| |
| 16:04 | Thin clients don't get server-side acceleration at all
| |
| 16:04 | In any case
| |
| 16:04 | <bennabiy> possibly
| |
| 16:04 | ah
| |
| 16:05 | <alkisg> Well,unless you do something custom with virtualgl
| |
| 16:05 | But not in standard ltsp
| |
| 16:05 | And that would only be for 3d...
| |
| 16:06 | <bennabiy> Basically, we are upgrading our lab, and are going to be putting the server in a closet because of the noise, but would like to make sure that it is going to provide for all the clients hooked up
| |
| 16:06 | <alkisg> Well you don't care about graphics, only about cpu and ram
| |
| 16:06 | Check cpubenchmark.net
| |
| 16:07 | And make sure your server has more score than N*client cpu
| |
| 16:07 | Your clients are so old that you use them as thins?
| |
| 16:07 | <bennabiy> I am looking at a dual Xeon X5660 with about 128-144GB Ram
| |
| 16:07 | my clients are N280 atom
| |
| 16:08 | Hp 5742
| |
| 16:08 | <alkisg> score?
| |
| 16:10 | <bennabiy> 286
| |
| 16:10 | Accordingly, the Xeons would be the equiv of 54 thin clients
| |
| 16:10 | with ram to match
| |
| 16:11 | <alkisg> cool
| |
| 16:11 | <bennabiy> I am pretty hard on the system though, and so I account for me using up many resources but I do not want to affect the lab
| |
| 16:11 | <alkisg> and how many clients?
| |
| 16:11 | <bennabiy> We have 5 right now, but are possibly going to have up to 12 with some power users
| |
| 16:12 | Not a classroom environment
| |
| 16:12 | more of an office
| |
| 16:13 | On a server like that, would it make sense to run the lab within a VM from it?
| |
| 16:13 | I have done it a couple different ways, but am still not settled on which way is best
| |
| 16:13 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
| 16:14 | <bennabiy> My current lab is an i5-2600K with 14.6GB Ram and we max that out pretty easily
| |
| 16:17 | <alkisg> I only use VMs for servers when I need snapshots
| |
| 16:17 | Otherwise,if backups are OK, I use real installations
| |
| 16:19 | <bennabiy> I am planning on having the whole thing on ZFS, but I also plan on running VMs on it, but the lab does not need to participate in that, so I thought I would just put the lab in a VM as well, which could make for me migrating to newer distros easy, because I could get the new one going and then just a quick switchover would have clients rebooted to new lab
| |
| 16:20 | ZFS makes for good snapshotting
| |
| 16:27 | I am going to try a remote install on a machine I already have and see how it performs. (with the onboard graphics etc)
| |
| 16:29 | <mmarconm> my server is a intel i3 with 4gb ram :) 30 fat clients 10/100 with hubs =(
| |
| 16:29 | running ltsp and ldap server
| |
| 16:30 | <bennabiy> I use LDAP as well
| |
| 16:30 | we are looking into fat clients, but at the moment we have many thin clients available, and they suit most people's needs
| |
| 16:31 | clients are Atom N280 with 2GB ram, and as a fat client, they would run pretty slow
| |
| 16:31 | <alkisg> bennabiy: I think in your case I would use thins with localapps
| |
| 16:31 | <mmarconm> bennabiy good, for me with a 4gb of ram server its impossible to run in thin client, and 10/100 ethernet speed as well
| |
| 16:32 | <alkisg> mmarconm: can't afford a single switch?!
| |
| 16:32 | <bennabiy> Yes, I have Gig ethernet to each client
| |
| 16:32 | <mmarconm> alkisg, how i explain, governemant do not have money to afford
| |
| 16:33 | * mmarconm <bennabiy> Yes, I have Gig ethernet to each client "jealous" | |
| 16:33 | <alkisg> OK, although buying one switch to get clients 10 times faster...
| |
| 16:33 | <mmarconm> yes, and turning the server 100/1000 too
| |
| 16:33 | <bennabiy> localapps works pretty good, is there a way to specify certain apps for localapps on all clients?
| |
| 16:33 | <alkisg> The server probably is already gigabit
| |
| 16:33 | <vagrantc> hard to imagine an i3 that doesn't have gigabit
| |
| 16:34 | <alkisg> bennabiy: yes, man lts.conf, there are settings for that
| |
| 16:34 | <mmarconm> alkisg, when i boot above 4 machine at same time i got a error, its possible to controle the boot time ?
| |
| 16:34 | <bennabiy> vagrantc: Hi!
| |
| 16:34 | * vagrantc waves | |
| 16:34 | <alkisg> mmarconm: what error?
| |
| 16:34 | Hi vagrantc
| |
| 16:34 | <mmarconm> let me see
| |
| 16:34 | <bennabiy> How is LTSP and Stretch?
| |
| 16:34 | <mmarconm> i took a picture of that
| |
| 16:34 | <alkisg> mmarconm: you should be able to boot all clients together
| |
| 16:34 | <vagrantc> bennabiy: admittedly, probably rougher than i'd like
| |
| 16:34 | * alkisg tried NFS + overlayfs in ubuntu 16.04 raspberry pi, it worked ok... | |
| 16:35 | <alkisg> Kernel 4.10 if it makes a difference
| |
| 16:35 | <vagrantc> bennabiy: i haven't had a real lab for years, so ... i can test on my little virtual lab, but it doesn't get the testing it used to
| |
| 16:35 | <bennabiy> vagrantc: understand!
| |
| 16:35 | <vagrantc> alkisg: overlayfs, or overlay fs ?
| |
| 16:35 | <bennabiy> I was considering moving the lab to it, but then thought against it
| |
| 16:36 | <mmarconm> alkisg, uploading the image
| |
| 16:36 | <bennabiy> I was just going to do a test install on Jessie (ish) system
| |
| 16:36 | <mmarconm> alkisg, https://imgur.com/a/TDWqp this error, if i try to boot 2 or 3 machines above
| |
| 16:36 | <alkisg> vagrantc: the kernel one
| |
| 16:36 | <vagrantc> bennabiy: don't mess with jessie at this point ... just use stretch, which released a few months ago
| |
| 16:37 | <mmarconm> Lubuntu 16.04 x64
| |
| 16:37 | <vagrantc> alkisg: i've not seen it work in ages
| |
| 16:37 | <bennabiy> vagrantc: well, I have a VM with a Jessie equiv running on smartos which I was going to test
| |
| 16:37 | <vagrantc> alkisg: well, it works well enough to be annoying ... e.g. some files write, but some don't
| |
| 16:37 | <bennabiy> not for production
| |
| 16:38 | lunch time, then launch time!
| |
| 16:38 | <vagrantc> bennabiy: well, there are lots of known bugs in jessie's ltsp :P
| |
| 16:38 | * bennabiy waves | |
| 16:38 | <bennabiy> ah
| |
| 16:38 | then I will just need to do an Ubuntu variety
| |
| 16:39 | <vagrantc> rkwesk has been testing ltsp on debian lately and updating the wiki page a bit
| |
| 16:53 | <mmarconm> vagrantc, did you seed the pic ?
| |
| 16:53 | any idea whats is this error ?
| |
| 16:54 | <vagrantc> mmarconm: do you have multiple dhcp servers? is it sometimes getting the wrong dhcp response?
| |
| 16:55 | mmarconm: e.g. it's trying to connect to nbd-server on your router rather than your LTSP server
| |
| 16:56 | <mmarconm> vagrantc, hunmnm i have other dhcp server yes
| |
| 16:57 | <vagrantc> mmarconm: are you using "ipappend 3" in pxelinux.cfg/default ?
| |
| 16:58 | <mmarconm> vagrantc, i didnt any config about this option in pxelinux.cf/default
| |
| 16:58 | its a default config
| |
| 16:59 | i will open the virtual machine here, has the save config thats i use on my lab
| |
| 16:59 | <vagrantc> mmarconm: default config is ipappend 2
| |
| 17:00 | mmarconm: so it's something you'll want to change
| |
| 17:00 | <mmarconm> vagrantc, what this ipappend 3 doees ?
| |
| 17:00 | <vagrantc> mmarconm: it passes the dhcp settings received on the initial dhcp request on to the kernel, and doesn't make a second dhcp request
| |
| 17:01 | !ipappend
| |
| 17:01 | <ltsp> ipappend: o temporarily solve DHCP problems in the initramfs, try putting IPAPPEND 3 after the APPEND line in /var/lib/tftpboot/ltsp/i386/pxelinux.cfg/default. More info: http://www.syslinux.org/wiki/index.php/SYSLINUX#IPAPPEND_flag_val_.5BPXELINUX_only.5D
| |
| 17:01 | <mmarconm> thank you, i wll change on lab later and tell you if works
| |
| 17:01 | * mmarconm grab a beer for vagrantc | |
| 17:02 | <vagrantc> mmarconm: once you change it there, you'll need to change it in another place to make sure it sticks
| |
| 17:02 | !update-kernels.conf
| |
| 17:02 | <ltsp> Error: "update-kernels.conf" is not a valid command.
| |
| 17:04 | <mmarconm> ltsp-update-kernel ?
| |
| 17:08 | <vagrantc> well, you need to edit update-kernels.conf
| |
| 17:09 | but first, test that editing pxelinux.cfg/default actually solves your issue
| |
| 17:09 | if it already has "ipappend 3" then you have a different issue
| |
| 17:10 | <mmarconm> vagrantc, nope, ipappend was 2, now i will change to 3 to see if the issue continues =)
| |
| 17:15 | <vagrantc> do not run ltsp-update-kernels after manually editing it
| |
| 17:15 | just reboot a client
| |
| 17:18 | <mmarconm> ok
| |
| 17:30 | Statler has left IRC (Statler!~Georg@gwrz3.lohn24.de, Remote host closed the connection) | |
| 18:04 | mmarconm has left IRC (mmarconm!~mmarconm@191.217.95.61, Quit: Leaving) | |
| 18:32 | lucascastro has left IRC (lucascastro!~lucas@201.182.221.14, Remote host closed the connection) | |
| 20:56 | Ghostdog has joined IRC (Ghostdog!5210e409@gateway/web/freenode/ip.82.16.228.9) | |
| 21:07 | Eric4 has left IRC (Eric4!~eric@sdi.iut-valence.fr, Ping timeout: 240 seconds) | |
| 21:12 | <Ghostdog> Hello, I have followed the instructions to install LTSP-Manager using Ubuntu MATE.( Steps 1 to 5). Am a bit stuck in terms of how I install the client software?ware
| |
| 21:14 | <vagrantc> Ghostdog: you install whatever software you want on the server
| |
| 21:14 | Ghostdog: then you use ltsp-manager to build an image out of your installed server
| |
| 21:16 | <Ghostdog> OK I guess thats step 4 (/Update image: create the virtual disk for the clients)?
| |
| 21:16 | <vagrantc> sounds about right
| |
| 21:17 | <Ghostdog> Do I need to copy the image from the server and install the image on the client workstation?
| |
| 21:17 | <vagrantc> typically no, they just boot off of the network
| |
| 21:18 | <Ghostdog> the client workstations are Windows XP, so do I need to install the client image first?
| |
| 21:19 | <vagrantc> you just need to configure PXE boot in BIOS
| |
| 21:19 | <Ghostdog> ok
| |
| 21:19 | Appreciate your help.
| |
| 21:20 | <vagrantc> might also be called network boot, etc.
| |
| 21:21 | <Ghostdog> ok, thanks for that
| |
| 21:21 | <vagrantc> !pxe
| |
| 21:21 | <ltsp> pxe: Preboot execution environment boots computers using a network interface and image. See http://en.wikipedia.org/wiki/Preboot_Execution_Environment
| |
| 21:21 | <vagrantc> !ipxe
| |
| 21:21 | <ltsp> ipxe: iPXE is the successor to the etherboot/gPXE project, and can be used to netboot clients that don't have a NIC ROM with a PXE stack. To add it to grub, see !grub-ipxe. To add it to the Windows boot loader, see !win32-loader. To download floppy, CD or USB images, visit http://ipxe.org or install the ipxe package.
| |
| 21:22 | <vagrantc> you can use ipxe if your bios doesn't have built-in network boot ... but most computers have it these days
| |
| 22:00 | Ghostdog has left IRC (Ghostdog!5210e409@gateway/web/freenode/ip.82.16.228.9, Ping timeout: 260 seconds) | |
| 22:05 | lucascastro has joined IRC (lucascastro!~lucas@138.68.106.79) | |
| 22:05 | ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving) | |
| 23:29 | lucascastro has left IRC (lucascastro!~lucas@138.68.106.79, Remote host closed the connection) | |
| 23:32 | ben_roose has left IRC (ben_roose!~roose@roose.cs.wichita.edu, Remote host closed the connection) | |