IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 28 June 2017 (all times are UTC)
| 03:40 | cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, Ping timeout: 240 seconds) | |
| 03:43 | lucascastro has joined IRC (lucascastro!~lucas@186.227.186.18) | |
| 03:53 | cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg) | |
| 05:13 | cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, Ping timeout: 268 seconds) | |
| 05:19 | cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg) | |
| 05:26 | Statler has joined IRC (Statler!~Georg@p579FEA3F.dip0.t-ipconnect.de) | |
| 05:47 | ABIX_Adamj_ is now known as ABIX_Adamj | |
| 06:02 | ricotz has joined IRC (ricotz!~ricotz@p5B2A9B02.dip0.t-ipconnect.de) | |
| 06:02 | ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz) | |
| 06:42 | <alkisg> highvoltage: now that we don't have much manpower in ltsp, is there any point in keeping so many ldm-themes around, instead of having a single common one for all ltsp distros/flavors?
| |
| 06:46 | <highvoltage> alkisg: not sure, depends what people want. I think a single theme will indeed be better.
| |
| 06:46 | <alkisg> highvoltage: if you're ok with that, and vagrant's ok with that, I don't care much about what people will want. They can come here and maintain them if they want. :)
| |
| 06:47 | <highvoltage> alkisg: +1
| |
| 06:47 | <alkisg> A nice theme with an LTSP logo instead of ubuntu/debian etc is fine for me!
| |
| 06:47 | Cool, I'll bring this up with vagrantc
| |
| 06:47 | <highvoltage> alkisg: I was wondering what's happening with the login screen plans now that lightdm is being abandonded
| |
| 06:47 | alkisg: so I guess good old ldm will have to live on a bit longer?
| |
| 06:48 | <alkisg> I don't see much work being done for ltsp6...
| |
| 06:48 | Personally I don't have the time to pull this off by myself
| |
| 06:48 | In which case, we'll stick with plain old ldm a while longer...
| |
| 06:49 | <highvoltage> I still have an old wishlist for LTSP I'll probably never get to
| |
| 06:49 | Eric4 has joined IRC (Eric4!~eric@sdi.iut-valence.fr) | |
| 06:49 | <alkisg> I will start dropping much unmaintained code though
| |
| 06:49 | E.g. nbd-proxy or altlinux
| |
| 06:49 | <highvoltage> but I care more about general netboot infrastructure than remote desktops via ssh anyway
| |
| 06:50 | what's altlinux?
| |
| 06:50 | <alkisg> Haha a distro that has a lot of code in ltsp but hasn't been touched for 8 years or so
| |
| 06:50 | <highvoltage> ah
| |
| 06:50 | ah right
| |
| 06:51 | * alkisg also cares more about fat clients and netbooting infrastructure than about thin clients... | |
| 06:53 | <bcg> ldm works fine as it is.
| |
| 06:55 | I'm using nfs-mounted chroot, most of the stuff running on servers (two per site), firefox as localapps, ldm, ...
| |
| 06:55 | <alkisg> Why localapps instead of ltsp fat clients?
| |
| 06:56 | If the client can run a browser, I imagine it can run a full session...
| |
| 06:56 | <bcg> I started with firefox in server too with smaller clients, but things have evolved.
| |
| 06:57 | Switching to localapps-firefox was smaller task than switching to full session.
| |
| 06:58 | We'll probably rething now as we're switching to rhel7 and new NUCs
| |
| 06:58 | * alkisg prefers the "either thin or fat", and preferably fat; localapps/remoteapps have inherent issues... | |
| 07:00 | <bcg> yeah, fat-testing is on my long todo.
| |
| 07:03 | and as there was (is :-) no real rhel-support the fat-clients are in bit bad shape, i guess.
| |
| 07:03 | <alkisg> bcg: if you use ltsp-pnp instead of ltsp-build-client, you shouldn't need to do much
| |
| 07:04 | <bcg> ltsp-pnp was not available in the older ltsp that I was using
| |
| 07:04 | <alkisg> ltsp-pnp clones the server installation into a squashfs image, so the basic thing to do is to disable automatic ip assignment at boot
| |
| 07:04 | Yeah I think I introduced it in 2012...
| |
| 07:04 | <bcg> Yeah, too new. :-)
| |
| 07:05 | Having separate chroot is not a problem, and we actually prefer different settings there.
| |
| 07:06 | I've also written yum ltsp-plugin, which updates chroot when server gets updates.
| |
| 07:06 | <alkisg> If you have a sysadmin/coder, sure maintaining a chroot isn't an issue
| |
| 07:06 | <bcg> So it's basically zero work after install.
| |
| 07:06 | <alkisg> But tell that to persons trying to install dkms modules in a chroot with different kernel version
| |
| 07:07 | While maintaining a template installation via GUI is the normal thing for most persons
| |
| 07:07 | So I think ltsp-build-client will die in a few years...
| |
| 07:09 | <bcg> Also our (current, old) chroot is 32bit while server is 64bit.
| |
| 07:10 | There was no 64bit support in ltsp's rhel code.
| |
| 07:10 | <alkisg> Our rule there is, "if you have at least one 32bit client, install 32bit ltsp-pnp. If all of your clients are 64bit, install 64bit ltsp-pnp".
| |
| 07:10 | <bcg> Sure, of course.
| |
| 07:11 | We've finally replaced all old 32bit clients with 64bit ones.
| |
| 07:12 | anyway, time to go now or I'll miss my flight.
| |
| 07:13 | <alkisg> Have fun with your vacations!
| |
| 07:13 | bb
| |
| 07:35 | forum has joined IRC (forum!~Thunderbi@93-82-98-209.adsl.highway.telekom.at) | |
| 07:43 | <alkisg> !ldm-bug
| |
| 07:43 | <ltsp`> I do not know about 'ldm-bug', but I do know about these similar topics: 'ltsp-bug'
| |
| 07:47 | vervelak has left IRC (vervelak!~vervelak@139.91.200.185, Remote host closed the connection) | |
| 08:12 | Statler has left IRC (Statler!~Georg@p579FEA3F.dip0.t-ipconnect.de, Remote host closed the connection) | |
| 08:26 | vervelak has joined IRC (vervelak!~vervelak@139.91.200.185) | |
| 08:52 | Statler has joined IRC (Statler!~Georg@mail.lohn24.de) | |
| 08:57 | JerryT has left IRC (JerryT!~jerry@wsip-70-165-106-163.om.om.cox.net, Ping timeout: 240 seconds) | |
| 08:58 | JerryT has joined IRC (JerryT!~jerry@wsip-70-165-106-163.om.om.cox.net) | |
| 10:13 | markus_e92 has left IRC (markus_e92!~markus_e9@91-115-17-251.adsl.highway.telekom.at, Ping timeout: 260 seconds) | |
| 10:16 | markus_e92 has joined IRC (markus_e92!~markus_e9@62-46-99-33.adsl.highway.telekom.at) | |
| 10:19 | GodFather has joined IRC (GodFather!~rcc@47.33.250.142) | |
| 10:47 | forum has left IRC (forum!~Thunderbi@93-82-98-209.adsl.highway.telekom.at, Ping timeout: 255 seconds) | |
| 10:52 | GodFather has left IRC (GodFather!~rcc@47.33.250.142, Ping timeout: 260 seconds) | |
| 11:29 | gvy has joined IRC (gvy!~mike@altlinux/developer/mike) | |
| 11:34 | forum has joined IRC (forum!~Thunderbi@93-82-98-209.adsl.highway.telekom.at) | |
| 12:28 | freddy has joined IRC (freddy!4edec986@gateway/web/freenode/ip.78.222.201.134) | |
| 12:29 | freddy is now known as Guest82540 | |
| 12:33 | <Guest82540> Hello, I'm trying to setup dual screen on thin clients booting LTSP 5.5.1 server installed on top of Ubuntu 14.04. I've difficulties to find good documentation on that point. Does anyone know where I may find documentation on this ? xrandr does not access to hardware on thin clients and report kind of virtual graphic device... How can I do in order xrandr to see local hardware while LOCALDEV is set to true in my lts.conf ?
| |
| 12:39 | <alkisg> Guest82540: first of all, it'd be best to use some newer ltsp version; but anyways, what's the output of `xrandr`, when you run it on the client?
| |
| 12:39 | Better yet, try this:
| |
| 12:39 | ltsp-localapps xterm
| |
| 12:39 | and from that xterm: xrandr | nc termbin.com 9999
| |
| 12:39 | This will put the xrandr results to termbin so that we see them too...
| |
| 12:41 | silvia_ has joined IRC (silvia_!4e0d6928@gateway/web/freenode/ip.78.13.105.40) | |
| 12:42 | silvia_ has left IRC (silvia_!4e0d6928@gateway/web/freenode/ip.78.13.105.40, Client Quit) | |
| 13:21 | <Guest82540> alkisg: I've added xterm to local_apps_items in my lts.conf. xrandr report : Screen0 and a "default" monitor, but nothing about hardware output VGA and HDMI...
| |
| 13:27 | alkisg: I see everywhere ltsp-localapps but on my installation there is only a ltsp-localappsd ... is this normal ?
| |
| 13:38 | forum has left IRC (forum!~Thunderbi@93-82-98-209.adsl.highway.telekom.at, Ping timeout: 240 seconds) | |
| 13:43 | exebat has left IRC (exebat!555e6071@gateway/web/freenode/ip.85.94.96.113, Ping timeout: 260 seconds) | |
| 13:46 | forum has joined IRC (forum!~Thunderbi@93-82-98-209.adsl.highway.telekom.at) | |
| 13:58 | <Guest82540> alkisg: I can see my thin clients are using /var/run/ltsp-xorg.conf which contain what I'm seeing with xrandr isn't it ? What would be the best way to update this config in order dual monitors to be taken in account ? Here is the default setup:Section "Screen" Identifier "Screen0" SubSection "Display" EndSubSection EndSection Section "Device" Identifier "Device0" Option "NoAccel" EndSection
| |
| 14:22 | <alkisg> Guest82540: if you managed to open a local xterm, there, `hostname` should tell you "ltsp123" instead of "server". Is that the case?
| |
| 14:22 | Guest82540: if you have noaccel, then you're using a subset of the driver needed
| |
| 14:22 | This comes from lts.conf settings like X_OPTION_01=xxx
| |
| 14:23 | To see the hardware, you'd run something like this: lspci -nn -k | grep -A 2 VGA
| |
| 14:40 | <Guest82540> alkisg: I can confirm I'm logged on thin client (using telnet to join thin clients from server)
| |
| 14:41 | <alkisg> OK, go on with the lspci command
| |
| 14:41 | <Guest82540> alkisg: I already tryed disable hw accel
| |
| 14:41 | alkisg: 00:02.0 VGA compatible controller: Intel Corporation Device 22b1 (rev 35)
| |
| 14:43 | <alkisg> And put the output of xrandr to pastebin
| |
| 14:43 | !pastebin
| |
| 14:43 | <ltsp`> pastebin: try !paste
| |
| 14:43 | <alkisg> !paste
| |
| 14:43 | <ltsp`> paste: the LTSP pastebin is at http://ltsp.pastebin.com. Please paste all text longer than a line or two to the pastebin, as it helps to reduce traffic in the channel. Don't forget to paste the URL of the text here.
| |
| 14:44 | <Guest82540> alkisg: I've to leave a couple of hour. I'll be back to debug this...
| |
| 14:45 | <alkisg> Guest82540: no problem, but note that it's summertime, so it might be a few weeks before you get help ;)
| |
| 14:45 | Keep trying.
| |
| 14:54 | ben_roose has joined IRC (ben_roose!~roose@roose.cs.wichita.edu) | |
| 15:19 | Freejack has left IRC (Freejack!~quassel@unaffiliated/freejack, Ping timeout: 246 seconds) | |
| 15:21 | Freejack has joined IRC (Freejack!~quassel@unaffiliated/freejack) | |
| 15:24 | forum has left IRC (forum!~Thunderbi@93-82-98-209.adsl.highway.telekom.at, Read error: Connection reset by peer) | |
| 15:24 | forum has joined IRC (forum!~Thunderbi@93-82-98-209.adsl.highway.telekom.at) | |
| 15:29 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
| 15:32 | <vagrantc> alkisg: heya!
| |
| 15:32 | alkisg: i seem to recall you mentioned a bug regarding PAE on i386, but i'm not finding any obvious references to this in email
| |
| 15:36 | IT-goober has joined IRC (IT-goober!4a7b4a42@gateway/web/freenode/ip.74.123.74.66) | |
| 15:40 | Guest82540 has left IRC (Guest82540!4edec986@gateway/web/freenode/ip.78.222.201.134, Ping timeout: 260 seconds) | |
| 15:45 | IT-goober has left IRC (IT-goober!4a7b4a42@gateway/web/freenode/ip.74.123.74.66, Ping timeout: 260 seconds) | |
| 16:26 | <alkisg> vagrantc: https://bugzilla.kernel.org/show_bug.cgi?id=196157
| |
| 16:26 | echo 1 > highmem_is_dirtyable is a workaround
| |
| 16:30 | <vagrantc> alkisg: thanks!
| |
| 16:30 | <alkisg> np! We're making some progress with ltsp-manager, today it was the first time we've seen it run on debian ;)
| |
| 16:31 | <vagrantc> alkisg: we were running a mixture of amd64 and i386 kernels on i386 builds for reproducible-builds.org, which caught some reproducibility issues ... but to work around whatever issue we're experiencing have switchted to amd64 kernels in all cases
| |
| 16:31 | alkisg: maybe it's related to that bug you mentioned :)
| |
| 16:31 | <alkisg> It could be. To verify, I was measuring dd'ing a small file before and after the issue
| |
| 16:32 | Before => 200 MB/sec, after => 2 MB/sec
| |
| 17:31 | vagrantc, stgraber, ogra: I think declaring ldm and ltspfs as "series and milestones" of ltsp in https://launchpad.net/ltsp is not correct, and we should declare them as separate projects instead, e.g. launchpad.net/ldm and launchpad.net/ltspfs... what do you think?
| |
| 17:31 | highvoltage: ^
| |
| 17:31 | (and same for ltsp-manager now...)
| |
| 17:32 | <highvoltage> alkisg: sounds sensible yes
| |
| 17:34 | <alkisg> ty - if the others agree, I'll create those projects
| |
| 17:35 | <vagrantc> i'm not sure what series and milestones are even used for
| |
| 17:35 | <alkisg> I think they're a method for supporting e.g. ltsp 5.5.9 for ubuntu 16.04, and ltsp 5.5.5 for ubuntu 14.04, etc
| |
| 17:36 | <vagrantc> ah, yes, that would make sense
| |
| 17:36 | <alkisg> So that one can have ltsp 5.5.9-1, -2, -3 etc, all in launchpad
| |
| 17:36 | <vagrantc> i don't really understand all the structure of launchpad
| |
| 17:36 | <alkisg> So we're abusing it in order to have "subprojects" for ldm/ltspfs
| |
| 17:36 | <vagrantc> i do like having ldm and ltspfs as subprojects of ltsp
| |
| 17:36 | <alkisg> But they're not subprojects that way
| |
| 17:37 | <vagrantc> agreed
| |
| 17:37 | <alkisg> It's just one way to abuse series/milestones to get ldm to be translatable etc
| |
| 17:37 | <vagrantc> by your proposed changes, is that going to change VCS urls and such?
| |
| 17:37 | <alkisg> Yes
| |
| 17:37 | <vagrantc> that would be really annoying
| |
| 17:37 | <alkisg> It'll be lp:ldm, not lp:ltsp/ldm
| |
| 17:37 | And I don't know about translations
| |
| 17:38 | I hope those can be moved...
| |
| 17:38 | <vagrantc> while it's clearly weird, i'm a little inclined to not mess with it
| |
| 17:38 | but i don't know all the ramifications ... i do know that it's frustrating to have all the packages in the debian repository reference invalid URLs
| |
| 17:39 | weather that's actually the case ... not sure
| |
| 17:39 | for ldm i'm currently using: git+ssh://vagrantc@git.launchpad.net/~ltsp-upstream/+git/ldm
| |
| 17:40 | <alkisg> ldm: Vcs-Git: https://anonscm.debian.org/git/collab-maint/ldm.git -b dmaster
| |
| 17:40 | Vcs-Browser: https://anonscm.debian.org/cgit/collab-maint/ldm.git/
| |
| 17:40 | Didn't you say you'd switch those to launchpad anyway?
| |
| 17:40 | <vagrantc> and debian/copyright still references bazaar...
| |
| 17:41 | * vagrantc sighs | |
| 17:41 | <vagrantc> it's all a big mess
| |
| 17:41 | <alkisg> ~ltsp-upstream/+git/ldm ==> I don't think that will change
| |
| 17:43 | <vagrantc> that's good
| |
| 17:45 | <alkisg> Although it's wrong, it's currently https://code.launchpad.net/~ltsp-upstream/ltsp/ldm
| |
| 17:45 | I.e. you missed the /ltsp there
| |
| 17:45 | But we can make it +git/ldm to match what you already have :D
| |
| 17:46 | (which makes more sense)
| |
| 17:51 | Since ltsp-manager is new, I think I'll switch it to a separate project and see how that goes
| |
| 17:51 | forum has left IRC (forum!~Thunderbi@93-82-98-209.adsl.highway.telekom.at, Quit: forum) | |
| 17:51 | <alkisg> I'll reuse ogra_'s project page; hopefullly I'll have enough access there..
| |
| 17:51 | gvy has left IRC (gvy!~mike@altlinux/developer/mike, Quit: Leaving) | |
| 17:52 | sebd has left IRC (sebd!~seb@aditu.ldd.fr, Ping timeout: 260 seconds) | |
| 18:01 | sebd has joined IRC (sebd!~seb@aditu.ldd.fr) | |
| 18:10 | <vagrantc> alkisg: it works, so i'm not sure how wrong it is
| |
| 18:11 | alkisg: i think the git and bzr URLs don't necessarily follow the same conventions... to keep things very interesting :)
| |
| 18:12 | Eric4 has left IRC (Eric4!~eric@sdi.iut-valence.fr, Ping timeout: 258 seconds) | |
| 18:14 | <vagrantc> or, at least, it did work at one point
| |
| 18:14 | as that's where i updated the repository from
| |
| 18:14 | this gets back to my frustration with the URLs moving around all the time
| |
| 18:15 | now neither work
| |
| 18:48 | ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving) | |
| 19:03 | Statler has left IRC (Statler!~Georg@mail.lohn24.de, Remote host closed the connection) | |
| 19:15 | gp has joined IRC (gp!~gp@96.71.93.13) | |
| 20:23 | freddy has joined IRC (freddy!4ef63284@gateway/web/freenode/ip.78.246.50.132) | |
| 20:24 | freddy is now known as Guest76315 | |
| 20:28 | <Guest76315> hello, I'm having trouble using thin client with dual monitors. Thin clients boot up an i386 image from a 5.5.1 LTSP deployed on top of an Ubuntu 14.04. xrandr on thin client does not report about local hardware. Does the xorg server require to run on thin clients in order local hardware to be used ? LOCALDEV is set to true in my lts.conf file.
| |
| 20:29 | <alkisg> Guest76315: that's a copy/paste of your previous question without mentioning anything about the progress
| |
| 20:29 | <Guest76315> I have a console access to thin clients, system and xorg logs.
| |
| 20:29 | <alkisg> For example, now you know how to run xrandr locally, you know the hardware because of lspci, etc etc
| |
| 20:29 | Put the xrandr output to pastebin
| |
| 20:31 | <Guest76315> alkisg: I can't run it as I do not have access to desktop, I only have a console access on thin client. Would you like to see xorg logs output instead of xrandr output ?
| |
| 20:31 | <alkisg> Guest76315: do you have root access to the client via telnet?
| |
| 20:32 | Or user access?
| |
| 20:32 | <Guest76315> alkisg: yep
| |
| 20:32 | <alkisg> OK, run these:
| |
| 20:32 | apt-get update; apt-get install epoptes-client
| |
| 20:32 | export $(/usr/share/epoptes-client/get-display)
| |
| 20:32 | xrandr
| |
| 20:32 | Then you'll have access to xorg...
| |
| 20:32 | <Guest76315> alkisg: ok I try it now
| |
| 20:42 | alkisg: https://pastebin.com/zSub4ejS
| |
| 20:42 | <alkisg> OK, so you have one output with those resolutions available
| |
| 20:42 | And what do you want to do with that?
| |
| 20:43 | <Guest76315> i'd like xrandr to report local output available on thin client, HDMI, VGA
| |
| 20:44 | in order to setup dual screen, one monitor left of the other for example
| |
| 20:46 | <alkisg> Isn't that the output of the client?
| |
| 20:46 | Did you run the commands on the server?!
| |
| 20:46 | <Guest76315> no the command ran on the thin client
| |
| 20:47 | <alkisg> OK, then you only have a "default" monitor there
| |
| 20:47 | Put your client xorg.log to pastebin as well
| |
| 20:47 | <Guest76315> ok
| |
| 20:47 | <alkisg> cat /var/log/Xorg* | nc termbin.com 9999
| |
| 20:47 | If you're using a "bad" xorg.conf, remove it and reboot the client
| |
| 20:48 | E.g. if you blacklisted the intel driver or used nomodeset or noaccel etc
| |
| 20:49 | <Guest76315> alkisg: what is termbin.com ? will I get an URL back using this cmd ?
| |
| 20:49 | <alkisg> Yes
| |
| 20:51 | Anyways, time to go... goodnight guys
| |
| 20:52 | <Guest76315> alkisg: thanks for your support ! good night
| |
| 20:53 | lucascastro has left IRC (lucascastro!~lucas@186.227.186.18, Ping timeout: 260 seconds) | |
| 20:59 | <Guest76315> As you can see, thin client kernel does see the HDMI hardware at boot time : Jun 28 17:28:12 ltsp22 kernel: [ 10.414625] input: HDA Intel HDMI as /devices/ pci0000:00/0000:00:1b.0/sound/card0/input11
| |
| 21:22 | Here is the log of the xorg server running on thin client : https://pastebin.com/KjBxmac3
| |
| 21:42 | Guest76315 has left IRC (Guest76315!4ef63284@gateway/web/freenode/ip.78.246.50.132, Quit: Page closed) | |
| 21:43 | <vagrantc> alkisg: what do you think about getting rid of ltsp-client-core and ltsp-server-standalone?
| |
| 21:44 | alkisg: basically, transforming ltsp-client into ltsp-client-core, adding recommends. similar for adding recommends for the ltsp-server-standalone package.
| |
| 21:45 | that didn't come out as clearly in words as in my mind...
| |
| 21:45 | e.g. we'd only have ltsp (the new one you proposed), ltsp-client and ltsp-server packages, and whatever optional packages would only be listed in recommends...
| |
| 21:46 | might be stuck with having transitional packages for a whole release cycle though
| |
| 21:49 | * vagrantc will follow-up on that tomorrow | |
| 22:58 | quinox has left IRC (quinox!~quinox@ghost.qtea.nl, Quit: WeeChat 1.7) | |
| 22:59 | quinox has joined IRC (quinox!~quinox@ghost.qtea.nl) | |
| 23:10 | ben_roose has left IRC (ben_roose!~roose@roose.cs.wichita.edu, Remote host closed the connection) | |