IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 30 May 2019 (all times are UTC)
| 00:24 | vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving) | |
| 02:48 | adrianorg has joined IRC (adrianorg!~adrianorg@177.18.102.123) | |
| 02:51 | adrianor1 has left IRC (adrianor1!~adrianorg@177.18.173.116, Ping timeout: 258 seconds) | |
| 05:23 | kjackal has joined IRC (kjackal!~quassel@2a02:587:3106:1000:a895:839f:87b0:61d8) | |
| 06:04 | kjackal has left IRC (kjackal!~quassel@2a02:587:3106:1000:a895:839f:87b0:61d8, Ping timeout: 258 seconds) | |
| 06:25 | kjackal has joined IRC (kjackal!~quassel@ppp141237188022.access.hol.gr) | |
| 08:17 | woernie has joined IRC (woernie!~werner@p57A0E866.dip0.t-ipconnect.de) | |
| 10:14 | profe01 has joined IRC (profe01!d500570e@gateway/web/freenode/ip.213.0.87.14) | |
| 10:32 | kjackal has left IRC (kjackal!~quassel@ppp141237188022.access.hol.gr, Ping timeout: 244 seconds) | |
| 10:47 | kjackal has joined IRC (kjackal!~quassel@2a02:587:3106:1000:a895:839f:87b0:61d8) | |
| 12:57 | bengoa has left IRC (bengoa!~alberto@194.50.55.200, Quit: Leaving) | |
| 12:58 | bengoa has joined IRC (bengoa!~alberto@194.50.55.200) | |
| 12:58 | bengoa has left IRC (bengoa!~alberto@194.50.55.200) | |
| 12:58 | bengoa has joined IRC (bengoa!~alberto@194.50.55.200) | |
| 13:07 | Faith has joined IRC (Faith!~Paty_@unaffiliated/faith) | |
| 13:29 | mgariepy has left IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy, Quit: Leaving) | |
| 14:12 | adrianor1 has joined IRC (adrianor1!~adrianorg@179.187.27.248.dynamic.adsl.gvt.net.br) | |
| 14:14 | adrianorg has left IRC (adrianorg!~adrianorg@177.18.102.123, Ping timeout: 248 seconds) | |
| 14:39 | GodFather has joined IRC (GodFather!~rcc@143.59.184.72) | |
| 15:47 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
| 16:24 | mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy) | |
| 17:11 | kjackal has left IRC (kjackal!~quassel@2a02:587:3106:1000:a895:839f:87b0:61d8, Ping timeout: 252 seconds) | |
| 17:15 | woernie has left IRC (woernie!~werner@p57A0E866.dip0.t-ipconnect.de, Ping timeout: 248 seconds) | |
| 17:15 | woernie has joined IRC (woernie!~werner@p57A0E866.dip0.t-ipconnect.de) | |
| 17:57 | kjackal has joined IRC (kjackal!~quassel@ppp-2-86-50-83.home.otenet.gr) | |
| 18:17 | <alkisg> Heh, I installed debian in a laptop, went to a trip, whoops, firmware missing for the wifi card. Meh, thankfully I also had windows to download the .deb :D
| |
| 18:17 | <mwalters> ops
| |
| 18:20 | <alkisg> Which of those sound better? ltsp-kernel, ltsp-image, ltsp-initrd, ltsp-config... or ltsp kernel, ltsp image, ltsp initrd, ltsp config?
| |
| 18:21 | And, /etc/ltsp/client.conf (previously lts.conf) or /etc/ltsp/ltsp-client.conf? I can't say I like that extra ltsp- there, even though the man page should be man 5 ltsp-client, I guess...
| |
| 18:21 | *man 5 ltsp-client.conf
| |
| 18:21 | <mwalters> so there'd be an 'ltsp' executable that you pass commands to?
| |
| 18:22 | <alkisg> Yup
| |
| 18:22 | I already support both of those methods; it's just a question for packaging at this point, what users would prefer
| |
| 18:22 | <mwalters> not sure I care from an end user standpoint for the first part... for the second one, the extra ltsp- seems pointless
| |
| 18:22 | (for the conf)
| |
| 18:22 | /etc/ltsp/client.conf seems fine
| |
| 18:22 | <alkisg> Great, i like that answer, ty :)
| |
| 18:23 | <mwalters> :)
| |
| 18:26 | <vagrantc> alkisg: if we manage to ship commandline completions that handle the subcommands, then "ltsp" seems better.
| |
| 18:26 | and even without, then it can be on a TODO list to ship the commandline completions ...
| |
| 18:27 | <alkisg> vagrantc: would the users know to `man ltsp-kernel` for ltsp kernel?
| |
| 18:28 | <vagrantc> alkisg: git manages to make "man git commit" work as expected
| |
| 18:28 | commandline completions again
| |
| 18:28 | or rather, not as expected, but as useful :)
| |
| 18:28 | * vagrantc was quite surprised | |
| 18:29 | <alkisg> Great then, less packaging work, we won't have to ship the /usr/sbin/ltsp-kernel => /usr/share/ltsp/ltsp.sh symlinks :)
| |
| 18:29 | I even made ltsp-initrd-bottom work that way, as an "ltsp applet"...
| |
| 18:30 | ltsp.sh sources all generic functions, then the applet, along with vendor/user overrides etc, so the final applet code is minimal
| |
| 18:31 | E.g. this is ltsp-initrd, which creates ltsp.img: https://github.com/eellak/gsoc2019-ltsp/blob/master/ltsp/applets/initrd/55-initrd.sh
| |
| 18:32 | Btw, ltsp config dnsmasq --options will need 3 words, but I guess that's ok too
| |
| 18:33 | <vagrantc> alkisg: this is just creating the stuff appended to the initrd, not the whole initrd?
| |
| 18:33 | <alkisg> Right, we don't touch the original initrd
| |
| 18:33 | <vagrantc> so it's really not creating an initrd... not a big deal, but a bit of a mental disconnect for me
| |
| 18:35 | <alkisg> vagrantc: ehm, how so? It's a .cpio.gz file, an additional initrd, that is added to the file system that the client sees
| |
| 18:35 | <vagrantc> alkisg: in my mind, an initrd is a complete initrd ... this is more of an initrd add-on.
| |
| 18:35 | alkisg: like i said, not a big deal
| |
| 18:36 | but someone might get the idea they only have to pass that part of the initrd or something...
| |
| 18:36 | <alkisg> The client init ram disk is comprised from 2 cpios; one the stock one, one the ltsp and the user code
| |
| 18:36 | <vagrantc> right, i get that
| |
| 18:36 | <alkisg> Oh, he'd have to run ltsp config ipxe or something, it wouldn't be adviceable to try a simple pxelinux etc
| |
| 18:37 | <vagrantc> it's not a technical disconnect, it's more of a mental image about what "ltsp initrd" generates.
| |
| 18:37 | <alkisg> So as to make it work with uefi and all
| |
| 18:37 | Any better name for it? ltsp client?
| |
| 18:37 | <vagrantc> not really, hence not a big deal and documentation can solve it
| |
| 18:37 | <alkisg> As it zips the client code into that additional initrd?
| |
| 18:38 | Btw, ltsp-update-kernels (and update-kernels) created pxelinux and all,
| |
| 18:39 | now we need ipxe.efi, ipxe.lkrn, boot.ipxe, grub.efi etc, things that usually are created once, then the user may manually edit boot.ipxe,
| |
| 18:39 | ...does that sound like an "ltsp config ipxe" command, or a separate "ltsp ipxe" or "ltsp tftp --ipxe" command?
| |
| 18:40 | It might be possible to use just syslinux or just grub in the future, but the initial version will use ipxe (+grub only for booting 32bit clients under uefi)
| |
| 18:42 | <vagrantc> from what you mentioned, grub.efi is the only thing that makes "ltsp config ipxe" not quite make sense
| |
| 18:43 | <alkisg> and memtest, and whatever else we might want to put there as the default;
| |
| 18:43 | while the user then can manually add entries for netbooting cds or whatever, by reading the wiki
| |
| 18:43 | that grub is just a helper for ipxe, it doesn't get a real menu
| |
| 18:49 | <vagrantc> alkisg: i'm leaning towards "ltsp config ipxe" given that it's primarily setting up ipxe related files
| |
| 18:50 | <alkisg> Thanks; I was thinking that all the other ltsp-config commands create configuration files for the server, while this messes with TFTP, like ltsp-initrd, ltsp-kernel etc, so ltsp tftp or something sounded a bit better to me... but just a bit
| |
| 18:51 | <vagrantc> it's kind of a coin toss, really
| |
| 18:52 | except when you start thinking about https boot
| |
| 18:52 | then ltsp tftp ... well ... :)
| |
| 18:52 | <alkisg> Well, ipxe will always be downloaded via tftp
| |
| 18:53 | The kernel/initrd is another thing
| |
| 18:53 | <vagrantc> yeah, those are all the things that spring to mind on the matter :)
| |
| 18:53 | <alkisg> man ltsp kernel ==> works, yey
| |
| 18:53 | <vagrantc> nice to get it "right" on the first go, but also sometimes just have to pick something and run with it :)
| |
| 18:54 | <alkisg> I think I switched from 'ltsp kernel' to 'ltsp-kernel' to sourcing, to executing... about 6 times so far :D
| |
| 18:54 | I think I got the best version now; I'm _almost_ sure :P
| |
| 18:54 | <vagrantc> heh
| |
| 18:56 | <alkisg> vagrantc: ubuntu lts releases sync from testing, afaik; IF ltsp19 is ready, when can it be uploaded to testing, wrt buster release schedule etc?
| |
| 18:58 | btw, another idea would be to drop `ltsp config service`, in favor of ltsp service... ltsp dnsmasq, ltsp isc-dhcp-server, ltsp nfs-server...
| |
| 18:58 | mads2 has joined IRC (mads2!~mads2@191.177.187.7) | |
| 18:58 | <alkisg> Then it's surely ltsp ipxe :D
| |
| 19:00 | And I think to simplify things, an `ltsp pnp` command may help, that autodetects what's missing and gets everything ready
| |
| 19:07 | Ah and we'll need an ltsp migrate command too, to help in cleaning up configuration files etc
| |
| 19:14 | spaced0ut has left IRC (spaced0ut!~spaced0ut@unaffiliated/spaced0ut, Remote host closed the connection) | |
| 19:27 | <vagrantc> alkisg: it depends on how long we want to keep ltsp5 around - do we want co-installable packages in the archive, etc.
| |
| 20:45 | adrianorg has joined IRC (adrianorg!~adrianorg@177.132.219.145) | |
| 20:46 | adrianor1 has left IRC (adrianor1!~adrianorg@179.187.27.248.dynamic.adsl.gvt.net.br, Ping timeout: 258 seconds) | |
| 20:49 | Faith has left IRC (Faith!~Paty_@unaffiliated/faith, Quit: Leaving) | |
| 20:57 | woernie has left IRC (woernie!~werner@p57A0E866.dip0.t-ipconnect.de, Remote host closed the connection) | |
| 22:33 | pppingme has left IRC (pppingme!~pppingme@unaffiliated/pppingme, Ping timeout: 258 seconds) | |
| 23:42 | pppingme has joined IRC (pppingme!~pppingme@unaffiliated/pppingme) | |