IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 18 November 2015 (all times are UTC)
| 00:28 | telex has left IRC (telex!teletype@94.247.40.156, Remote host closed the connection) | |
| 05:15 | Fenuks|2 has joined IRC (Fenuks|2!~Fenuks@91.192.173.88) | |
| 05:51 | Fenuks|2 has left IRC (Fenuks|2!~Fenuks@91.192.173.88, Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/) | |
| 05:52 | Fenuks|2 has joined IRC (Fenuks|2!~Fenuks@91.192.173.88) | |
| 05:52 | Fenuks|2 is now known as Fenuks | |
| 05:52 | work_alkisg is now known as alkisg | |
| 06:30 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
| 06:49 | vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Ping timeout: 240 seconds) | |
| 06:59 | alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.) | |
| 07:01 | alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg) | |
| 07:18 | vagrantc has joined IRC (vagrantc!~vagrant@2601:1c2:200:d15e:86a6:c8ff:fe87:ac53) | |
| 07:18 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
| 07:20 | <alkisg> vagrantc: have you seen https://bugs.launchpad.net/ltsp/+bug/693151 ?
| |
| 07:21 | ricotz has joined IRC (ricotz!~ricotz@p5B2A9758.dip0.t-ipconnect.de) | |
| 07:21 | ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz) | |
| 07:24 | uXus has left IRC (uXus!~uXus@217.77.222.72, Remote host closed the connection) | |
| 07:25 | navgou has joined IRC (navgou!c23fefeb@gateway/web/freenode/ip.194.63.239.235) | |
| 07:29 | uXus has joined IRC (uXus!~uXus@217.77.222.72) | |
| 07:33 | <vagrantc> alkisg: i've seen the bug ... never had a dual-monitor setup to test with
| |
| 07:35 | <alkisg> The person that sent the patch seems tired, like the devs are not even responding to patches etc
| |
| 07:52 | mikkel has joined IRC (mikkel!~mikkel@mail.dlvs.dk) | |
| 08:04 | navgou has left IRC (navgou!c23fefeb@gateway/web/freenode/ip.194.63.239.235, Quit: Page closed) | |
| 08:20 | vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving) | |
| 09:26 | alkisg is now known as work_alkisg | |
| 10:25 | gchaos has joined IRC (gchaos!~administr@187.87.208.114) | |
| 10:48 | Faith has joined IRC (Faith!~paty_@unaffiliated/faith) | |
| 11:06 | BuddyButterfly has joined IRC (BuddyButterfly!~BuddyButt@81.169.136.31) | |
| 11:36 | lbssousa has joined IRC (lbssousa!~laercio@177.143.61.251) | |
| 11:39 | vickymnt has joined IRC (vickymnt!c23fefeb@gateway/web/freenode/ip.194.63.239.235) | |
| 13:05 | Phantomas has joined IRC (Phantomas!~ftsamis@ubuntu/member/phantomas) | |
| 13:11 | BuddyButterfly has left IRC (BuddyButterfly!~BuddyButt@81.169.136.31, Ping timeout: 252 seconds) | |
| 13:28 | BuddyButterfly has joined IRC (BuddyButterfly!~BuddyButt@h1359005.stratoserver.net) | |
| 13:40 | professor_ has joined IRC (professor_!baf86132@gateway/web/freenode/ip.186.248.97.50) | |
| 13:41 | professor_ has left IRC (professor_!baf86132@gateway/web/freenode/ip.186.248.97.50, Client Quit) | |
| 13:47 | Fenuks has left IRC (Fenuks!~Fenuks@91.192.173.88, Ping timeout: 240 seconds) | |
| 13:54 | schlady has joined IRC (schlady!~schlady@141-53-209-146.ip.uni-greifswald.de) | |
| 13:55 | <schlady> Hi there!
| |
| 13:55 | Has anyone of you used Dropbox on LTSP FatClients?
| |
| 13:57 | I have the problem that the dropbox database seems to be corrupted everytime the FatClient is restarted and therefore always forced the user to login into dropbox again. Is that a problem with dropbox and ssfs?!
| |
| 13:58 | sshfs
| |
| 13:59 | schlady has left IRC (schlady!~schlady@141-53-209-146.ip.uni-greifswald.de, Remote host closed the connection) | |
| 14:00 | schlady has joined IRC (schlady!~schlady@141-53-209-146.ip.uni-greifswald.de) | |
| 14:11 | <bennabiy> vagrantc, alkisg: The issue with that bug is it is two different issues people are running into. It is one I experience because of the thin client I use. Somehow it picks up a ghost display which is not present, which needs to be disabled (hence the LVDS fix) but the original poster was having issues with an actual dual display having the login split between windows.
| |
| 14:11 | Same issue, but two different issues
| |
| 14:57 | ben_roose has joined IRC (ben_roose!~roose@roose.cs.wichita.edu) | |
| 15:20 | mikkel has left IRC (mikkel!~mikkel@mail.dlvs.dk, Quit: Leaving) | |
| 15:40 | work_alkisg is now known as alkisg | |
| 15:40 | <alkisg> bennabiy: I think the main issue is that noone is working on upstreaming that bug :)
| |
| 15:40 | *patch
| |
| 15:41 | <bennabiy> alkisg: In a few weeks I might be able to, but I do not have a dual monitor setup where I am at to confirm, but I know I can confirm it with the client I have (the second of the effects of the bug)
| |
| 15:42 | <alkisg> Can you confirm that the patch solves the second issue?
| |
| 15:42 | Also, it's quite easy to arrange a multi-monitor setup in a vbox client
| |
| 15:42 | <bennabiy> Ah yes, I forgot about that
| |
| 15:43 | I will test it and let you know
| |
| 15:43 | <alkisg> That'd be cool, thanks
| |
| 15:43 | A comment in the bug report "works for me, solves the second issue with no regressions" should be enough to get it upstream...
| |
| 15:45 | <bennabiy> I do not remember, did the fat client password hash make it upstream into 14.04?
| |
| 15:47 | <alkisg> I think it's post-14.04
| |
| 15:47 | Maybe even post-15.10
| |
| 15:47 | It's available in the Greek schools PPA though
| |
| 15:48 | m3741 has left IRC (m3741!8c20b7fe@gateway/web/freenode/ip.140.32.183.254, Quit: Page closed) | |
| 15:58 | <teknkik> now that you guys mentioned it
| |
| 15:58 | the login screen being split across 2 displays shows up in the setup i administer as well
| |
| 15:58 | ubuntu 14.04
| |
| 15:58 | <alkisg> teknkik: are you able to test the fix?
| |
| 15:59 | that's proposed in that bug report?
| |
| 16:01 | <lbssousa> Phantomas, alkisg, I've just seen Epoptes' latest revisions (457 and 458). I presume my auto-reconnect branch is no longer needed, right?
| |
| 16:02 | <alkisg> lbssousa: yup, I had to reject it, thank you though!
| |
| 16:02 | <teknkik> well it's kinda an deployment system
| |
| 16:02 | but we're doing some maintenance on it on 12th of december, i'll add that to the list
| |
| 16:05 | <alkisg> lbssousa: some of the issues are hard to solve correctly, that's why we don't work on them for a long time... some of your branches unfortunately target such issues... e.g. we'll probably do the same for the multi-seat branch, once ubuntu starts using systemd for the user sessions
| |
| 16:05 | So your branches are welcome; but unfortnately some of them aren't commit-able...
| |
| 16:06 | teknkik: I'll try to have an ldm build in one of the PPAs till then, so that you can test and comment more easily.
| |
| 16:10 | <lbssousa> alkisg, Phantomas, OK. I'll test latest epoptes-client code and see if auto-reconnection works for me. I'll tell you if I find something relevant.
| |
| 16:10 | <alkisg> Thank you lbssousa :)
| |
| 16:11 | <lbssousa> Any comment about my x11vnc-wrapper branch?
| |
| 16:11 | It's the last one now.
| |
| 16:16 | <alkisg> lbssousa: it's a bit "hackish" to have to slow down the mouse, that's why we aren't committing it without exploring other options first
| |
| 16:17 | And it's low priority, so that one will unfortunately have to wait too, sorry :-/
| |
| 16:17 | <lbssousa> alkisg: OK :)
| |
| 16:17 | <alkisg> It may be the best solution available, and thank you for that, but I don't think we can commit it without fully exploring it first (also for possible regressions)
| |
| 16:18 | Thanks! :)
| |
| 16:20 | <lbssousa> BTW Please keep track of https://github.com/LibVNC/x11vnc/issues/18
| |
| 16:22 | <alkisg> Nice!
| |
| 16:34 | <bennabiy> alksig: testing on my client right now...
| |
| 16:34 | Phantomas has left IRC (Phantomas!~ftsamis@ubuntu/member/phantomas) | |
| 16:35 | <bennabiy> patch does not fix second issue
| |
| 16:36 | If I do not have the video=LVDS-1:d then I get the split screen issue (with no second screen)
| |
| 16:37 | any other testing I should do before I remove the image?
| |
| 16:39 | is there a way to only target ltsp-update-image to build from one chroot, and not everything in the /opt/ltsp directory?
| |
| 16:39 | alkisg ^
| |
| 16:48 | afmorro has joined IRC (afmorro!bffd2f65@gateway/web/freenode/ip.191.253.47.101) | |
| 16:49 | <afmorro> hi guys. just a simple question. Is LTSP dead?
| |
| 16:49 | haven't seen any modifications/updates on the website in a couple years now
| |
| 16:50 | schlady has left IRC (schlady!~schlady@141-53-209-146.ip.uni-greifswald.de, Remote host closed the connection) | |
| 16:50 | <alkisg> afmorro: no it's not dead, but it only has a couple of developers working on it now
| |
| 16:51 | <afmorro> so sad...
| |
| 16:51 | <alkisg> And outside contributions are now minimal
| |
| 16:51 | Well, you could maintain the site if you wanted :)
| |
| 16:51 | <afmorro> to many things going on now. :)
| |
| 16:52 | <alkisg> For all of us :)
| |
| 16:52 | <afmorro> for sure.
| |
| 16:52 | <alkisg> So at least it's still working fine
| |
| 16:52 | Even if the documentation is not updated frequently
| |
| 16:53 | <afmorro> yeap, just cuz' usually we se lots and lots of updates on pretty much everything, and ltsp seemed to be stoped somewhere in 2013
| |
| 16:53 | <alkisg> bennabiy: so you now have a ghost monitor and the login is at the virtual center of those 2 monitors?
| |
| 16:53 | <afmorro> and since been a longe time I don't work with it, and now got a project for a public school, made me go searching documentations again
| |
| 16:53 | m3741 has joined IRC (m3741!8c20b7fe@gateway/web/freenode/ip.140.32.183.254) | |
| 16:53 | <alkisg> Some persons left the project then, yes
| |
| 16:53 | The guy that was working with the wiki included
| |
| 16:53 | <bennabiy> on my actual thin client, yes, the main monitor is showing half a screen, with the logo pushed over to the right hand side.
| |
| 16:54 | <alkisg> bennabiy: could you comment on the bug report?
| |
| 16:54 | <afmorro> allright. Thank you for your help. Back to building up my server and making some students happy again.
| |
| 16:54 | <alkisg> bennabiy: ltsp-update-image chroot?
| |
| 16:54 | afmorro: which distro/version will you be using?
| |
| 16:55 | <afmorro> Maybe CentOS, which I like, but ubuntu and its derivates seem to be more user friendly, specially for people migrating from windows
| |
| 16:55 | <bennabiy> alkisg: I thought I tried that, but I probably tried ltsp-update-image -a i386 or something like that
| |
| 16:55 | <alkisg> afmorro: ltsp isn't maintained on centos, but it's working fine for ubuntu if you use our ppa
| |
| 16:55 | bennabiy: -a won't work, right
| |
| 16:55 | <bennabiy> yes
| |
| 16:56 | <maldridge> afmorro: highly recommend ubuntu, the ppa gets updates way faster than any other source
| |
| 16:56 | <alkisg> !greek-schools-ppa
| |
| 16:56 | <ltsp> greek-schools-ppa: https://launchpad.net/~ts.sch.gr/+archive/ppa/ supports LTS Ubuntu releases with newer LTSP versions, bug fixes etc
| |
| 16:56 | <bennabiy> alkisg: without the -a it works
| |
| 16:56 | thank you
| |
| 16:56 | <alkisg> We're using that one in 1000+ schools
| |
| 16:56 | <afmorro> I'll try. Will its derivates work fine also?
| |
| 16:57 | <maldridge> afmorro: I'm running stuff here on xubuntu, haven't noticed any issues thus far
| |
| 16:57 | <alkisg> afmorro: some of them. Gnome-flashback is the recommented one, lubuntu and mate will also work fine
| |
| 16:57 | Thanks maldridge for the xubuntu input
| |
| 16:57 | <maldridge> np
| |
| 16:57 | <afmorro> awesome.
| |
| 16:57 | <bennabiy> afmorro: works fine in linuxmint
| |
| 16:57 | <alkisg> Personally I don't recomment linux mint, its packaging sucks (unrelated to ltsp)
| |
| 16:57 | * bennabiy sighs | |
| 16:57 | <alkisg> E.g. apt-get dist-upgrade => base files complains about /etc/os-release
| |
| 16:58 | <maldridge> alkisg: can you suggest any sources for school management software in linux?
| |
| 16:58 | <alkisg> The mint developers don't even know to at least use dpkg-divert...
| |
| 16:58 | <afmorro> linuxmint... its... didn't like it
| |
| 16:58 | <alkisg> maldridge: school management, as in timetable etc?
| |
| 16:58 | Or like epoptes, sch-scripts, i.e. computer lab administration?
| |
| 16:58 | <afmorro> allright. Time to download xubuntu. Thank you alkisg and bennabiy
| |
| 16:59 | <bennabiy> afmorro: good luck!
| |
| 16:59 | <maldridge> alkisg: as in grade management, degree planning, transcript generation etc. I was tasked to find a replacement to a program that was EOL'd back in 2005 with XP
| |
| 16:59 | <afmorro> thank you all
| |
| 16:59 | <alkisg> bb
| |
| 16:59 | maldridge: sorry, I did see a couple of related apps but I haven't tried any of them, as we're using an in-house solution from our ministry of education
| |
| 16:59 | web-based
| |
| 17:00 | <maldridge> ah, np; I note edubuntu but it seems like its more based torwards student facing stuff
| |
| 17:05 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
| 17:11 | * alkisg has been restarting epoptes and epoptes-client for half an hour, and the clients are still connected... success! :) | |
| 17:11 | <vagrantc> alkisg: epoptes and ltsp made it into jessie-backports
| |
| 17:11 | <alkisg> Nice!!!
| |
| 17:11 | * vagrantc hasn't pushed ldm yet, and there's no real reason to push ltspfs | |
| 17:12 | <alkisg> We'll make a new 0.6 release for epoptes, which will be the last one that won't depend on systemd
| |
| 17:12 | <vagrantc> arguably, the ldm fixes might be worth pushing to jessie directly
| |
| 17:12 | <alkisg> I'll test and commit the dual-screen issue with ldm
| |
| 17:12 | You might want to wait for that
| |
| 17:12 | <vagrantc> that's a notable feature
| |
| 17:14 | <alkisg> gym time!
| |
| 17:14 | alkisg is now known as work_alkisg | |
| 17:19 | <maldridge> so epoptes 0.6 is the version to fork, noted
| |
| 17:40 | <lbssousa> alkisg, I've just tested latest epoptes code here. It seems auto-reconnection is working as expected. I only needed to change my server FQDN (resolved by my local DNS) with its IP address in /etc/default/epoptes-client.
| |
| 17:42 | I only have a lillte suggestion: make socat timeout configurable in /etc/default/epoptes-client. Some users my want to give it a greater or smaller value, depending on their network characteristics.
| |
| 18:01 | danau111 has joined IRC (danau111!~durban@66.251.57.114) | |
| 18:04 | danau111 has left IRC (danau111!~durban@66.251.57.114) | |
| 18:08 | <lbssousa> work_alkisg, I'll open a bug report to keep track of this.
| |
| 18:20 | <afmorro> Hello again, quick question. I saw a few of my diskless clientes have a NVidia card on it, will it be possible to use its capabilities?
| |
| 18:21 | <maldridge> afmorro: if you build your image with the correct drivers, you should be able to use it just fine
| |
| 18:21 | <vagrantc> you'll need to use LTSP fatclients instead of thinclients to get the most out of it
| |
| 18:22 | <afmorro> fatclients?
| |
| 18:22 | <vagrantc> !fatclient | echo afmorro
| |
| 18:22 | <ltsp> afmorro I do not know about 'fatclient', but I do know about these similar topics: 'fatclients', 'fatclient-printers'
| |
| 18:22 | <vagrantc> !fatclients | echo afmorro
| |
| 18:22 | <ltsp> afmorro fatclients: You may find some info about the Ubuntu/LTSP implementation of fat clients at https://help.ubuntu.com/community/UbuntuLTSP/FatClients
| |
| 18:23 | <vagrantc> network booted, network root filesystem, but runs applications locally
| |
| 18:23 | <afmorro> just got there.
| |
| 18:23 | nice... will give it a go
| |
| 18:23 | but seems I'll need more then one image
| |
| 18:24 | between thing and fat clients.
| |
| 18:24 | <vagrantc> ah, you might need to write hooks to switch the mesa libraries at boot ...
| |
| 18:24 | <afmorro> nope, sorry, just found a few arguments... FAT_RAM_THRESHOLD=800
| |
| 18:24 | it can be enough
| |
| 18:25 | <vagrantc> worked on supporting a mixed nvidia/other environment until it turned out that the noveua drivers actually performed better on that hardware anyways...
| |
| 18:27 | <afmorro> got it... but its okay. I'm pretty sure it will be easier to manage come pieces in and out, so I can have all of the "commom area" with same specs, and leave more powerfull ones to specific jobs
| |
| 18:27 | some* pieces
| |
| 18:33 | danau11 has joined IRC (danau11!~durban@66.251.57.114) | |
| 18:41 | danau11 has left IRC (danau11!~durban@66.251.57.114) | |
| 19:04 | lbssousa has left IRC (lbssousa!~laercio@177.143.61.251, Quit: lbssousa) | |
| 19:10 | work_alkisg is now known as alkisg | |
| 19:13 | <alkisg> (07:19:35 μμ) maldridge: so epoptes 0.6 is the version to fork, noted ==> aren't you using xubuntu?!
| |
| 19:13 | lbssousa: in which case is the timeout not suitable?
| |
| 19:14 | afmorro: this is probably the easiest way to install and maintain ltsp:
| |
| 19:14 | !ltsp-pnp
| |
| 19:14 | <ltsp> ltsp-pnp: ltsp-pnp is an alternative (upstream) method to maintain LTSP installations for thin and fat clients that doesn't involve chroots: https://help.ubuntu.com/community/UbuntuLTSP/ltsp-pnp
| |
| 19:22 | <alkisg> vagrantc: I hit this problem with KERNEL_DEVICE: on vfat partitions, the vfat module isn't available to modprobe it, so we can't access the partition in order to update the initramfs
| |
| 19:22 | A solution would be to move the script to the initramfs, where the vfat module is available,
| |
| 19:23 | but then I thought of something else... if the squashfs file system doesn't have /lib/modules/$(uname -r), then I think it's best to copy it from the initramfs, so that at least some of the modules are available
| |
| 19:23 | It will waste some swap space, but it sounds better than having only built-in modules available, what do you say?
| |
| 19:26 | (to sum up, I'm suggesting to cp -r /lib/modules/kernel-version from the initramfs to /rootmnt, from init-bottom/ltsp)
| |
| 19:30 | Hmmm although ntfs seems to work fine, as it's based on fuse...
| |
| 19:36 | <vagrantc> alkisg: why isn't the module available?
| |
| 19:38 | Faith has left IRC (Faith!~paty_@unaffiliated/faith, Quit: Leaving) | |
| 19:38 | <alkisg> vagrantc: because the kernel was upgraded+purged and then ltsp-update-image was ran
| |
| 19:39 | And the kernel/initrd was loaded locally from some hard disk
| |
| 19:39 | (this is used in raspberry pi and in cases where ipxe doesn't support the card)
| |
| 19:40 | I wonder why it worked when I was testing with Pi, and it didn't work in my tests today with an ipxe-unsupported card...
| |
| 19:40 | vfat in both cases
| |
| 19:42 | * alkisg checks if it has to do with 3.2 vs 3.13 kernel... | |
| 19:47 | <vagrantc> kernel might have had vfat as a built-in?
| |
| 19:47 | alkisg: why not just add vfat to the initramfs modules?
| |
| 19:48 | <alkisg> vagrantc: the script currently is in init-ltsp.d
| |
| 19:48 | vfat is already present as an initramfs module
| |
| 19:48 | <vagrantc> alkisg: so load it from the initramfs :)
| |
| 19:48 | <alkisg> So I imagine that if I copied the modules dir, it would work
| |
| 19:49 | That's the easy solution, sure
| |
| 19:49 | But if the modules dir is missing, then I imagine that more things won't work
| |
| 19:49 | <vagrantc> just force-load it or whatever
| |
| 19:49 | <alkisg> Like mice
| |
| 19:49 | And if we do copy it from the initramfs, they'll work
| |
| 19:49 | It's one `cp` away. Should I do that?
| |
| 19:49 | <vagrantc> this is just done if it detects the update is needed?
| |
| 19:50 | <alkisg> There are 2 checks
| |
| 19:50 | The first one is to check if the modules dir for the current version is missing
| |
| 19:50 | That's the one I'm talking about
| |
| 19:50 | In that case, we can make things a bit better by copying the modules dir from the initramfs to the real file system
| |
| 19:51 | So that at least all the modules included in the initramfs, are also available later on when the system continues to boot
| |
| 19:51 | That part has nothing to do with kernel updating
| |
| 19:51 | <vagrantc> at least in debian, it's not uncommon for the modules dir to be identical, even though the kernel versions are different ... though i think in theory they should be compatible at that point
| |
| 19:52 | <alkisg> I think modprobe refuses to load modules tagged with a different kernel version
| |
| 19:52 | And it doesn't even find them, it doesn't search in dirs other than $(uname -r)
| |
| 19:52 | <vagrantc> yes, on debian, "uname -r" will match for multiple kernel versions, when it's ABI compatible.
| |
| 19:53 | the ABI rarely changes during a stable cycle, for example, but security updates and new upstream point releases come out all the time
| |
| 19:54 | <alkisg> OK, but we're talking about different $(uname -r) now, right? Even if that happens more rarely in debian, it's still affected...
| |
| 19:57 | <vagrantc> well, i'd rather not have old modules copied into the directory which may have security vulnerabilities
| |
| 19:58 | even if it's ABI compatible, it should only be done if needed.
| |
| 19:58 | <alkisg> vagrantc: I don't get it. Old kernel without modules (i.e. broken) is better than old kernel with its own modules?
| |
| 19:58 | Ah, here's the difference. CONFIG_FAT_FS was "m" in 3.2, and it's "y" in 3.13.
| |
| 19:58 | Probably due to efi nowadays
| |
| 20:03 | <vagrantc> alkisg: ok, i didn't quite catch you were talking about when the modules dir is missing entirely only
| |
| 20:03 | <alkisg> Yup
| |
| 20:04 | <vagrantc> in that case, i guess it makes sense
| |
| 20:04 | <alkisg> About the local kernel upgrades, 3.13+ can read vfat and ntfs without the modules dir present, so np there
| |
| 20:04 | Let me see the size of the modules dir in the initramfs...
| |
| 20:05 | $ du -sh 3.19.0-33-generic/
| |
| 20:05 | 31M 3.19.0-33-generic/
| |
| 20:06 | Hmmm that's a lot though, swap won't always be available....
| |
| 20:06 | And I'm guessing if I run mount --bind, then initramfs won't be freed
| |
| 20:20 | schlady has joined IRC (schlady!~schlady@ip1f111304.dynamic.kabel-deutschland.de) | |
| 20:36 | <vagrantc> fwiw, i seem to be using u-boot with rpi2 just fine...
| |
| 20:37 | guess i haven't tried network boot
| |
| 20:45 | <maldridge> alkisg: I run xubuntu on one of the production clusters, but the LUG's cluster is SystemD-free
| |
| 20:45 | actually all the production clusters run xubuntu, we haven't got the dev cluster stable yet
| |
| 21:02 | <alkisg> maldridge: why the mention about epoptes 0.6 (==last systemd version) then?
| |
| 21:03 | vagrantc: did you self-compile u-boot or did you download it from somewhere?
| |
| 21:04 | <vagrantc> alkisg: i added it as a target to the debian source package ...
| |
| 21:04 | alkisg: i haven't even tested video yet, though
| |
| 21:04 | <alkisg> Ah
| |
| 21:05 | It'd be nice to have an easy way to select between netbooted ltsp client, standalone ltsp client, or normal local boot...
| |
| 21:05 | <vagrantc> video doesn't seem to work
| |
| 21:05 | <alkisg> ...i'll probably end up writing a menu inside the initramfs for this...
| |
| 21:07 | <vagrantc> and a custom kernel built with rpi's patches and debian's packaging
| |
| 21:07 | and a really sloppy kernel config ...
| |
| 21:07 | <maldridge> alkisg: honestly, I'm very curious what it could possibly be using that is systemd related
| |
| 21:07 | schlady has left IRC (schlady!~schlady@ip1f111304.dynamic.kabel-deutschland.de, Remote host closed the connection) | |
| 21:08 | schlady has joined IRC (schlady!~schlady@ip1f111304.dynamic.kabel-deutschland.de) | |
| 21:09 | <maldridge> also, a custom menu inside initramfs is probably going to be easier than working with uboot, I've found its not very intuitive to get things done inside u-boot, and you can make a much nicer menu in initramfs
| |
| 21:11 | <alkisg> maldridge: about epoptes, we'll be shipping only a systemd service instead of a sysvinit one, and we'll be using systemd-logind for session registration etc
| |
| 21:11 | <maldridge> ah, any reason not using pam_tally?
| |
| 21:12 | the service one I can refactor easily, but the logind one is going to be much harder to synthesize
| |
| 21:12 | <alkisg> Not for counting, but for registering the app inside the user session
| |
| 21:12 | schlady has left IRC (schlady!~schlady@ip1f111304.dynamic.kabel-deutschland.de, Ping timeout: 252 seconds) | |
| 21:12 | <alkisg> So that it exits when the user logs out
| |
| 21:12 | non-xorg apps have an issue there
| |
| 21:13 | Currently we're using xprop for this, to test if xorg is still available
| |
| 21:13 | Why would you need to refactor it?
| |
| 21:13 | <maldridge> ah, yes I've run into that before, wound up setting up a dm hook to pkill -u on exit
| |
| 21:13 | <alkisg> xubuntu is using systemd...
| |
| 21:14 | <maldridge> alkisg: because I'm slowly trying to stand up all of ltsp/epoptes on a distro that has no SystemD
| |
| 21:14 | <alkisg> Which distor?
| |
| 21:14 | *distro
| |
| 21:14 | <maldridge> Void Linux (init is runit)
| |
| 21:15 | I've got a lot of the base stuff up, now its just down to porting over some of the compiled bits, unfortunately I have more irons in the fire than I can tend at the moment
| |
| 21:15 | <alkisg> OK, I'll try to make it easy to keep using it without systemd wherever I can
| |
| 21:17 | <maldridge> many thanks, I'm not oppossed to needing helper software since it would seem sans systemd is rare now, but refactoring C++ is not my strongsuit
| |
| 21:23 | afmorro has left IRC (afmorro!bffd2f65@gateway/web/freenode/ip.191.253.47.101, Ping timeout: 246 seconds) | |
| 21:26 | alkisg is now known as work_alkisg | |
| 21:37 | ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving) | |
| 22:33 | cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, Ping timeout: 240 seconds) | |
| 22:42 | cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg) | |
| 23:45 | ben_roose has left IRC (ben_roose!~roose@roose.cs.wichita.edu, Remote host closed the connection) | |