IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 1 September 2019 (all times are UTC)
| 03:03 | vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving) | |
| 06:32 | Vb has joined IRC (Vb!671f90f4@103.31.144.244) | |
| 06:33 | <Vb> Hi This is VB here from India, I was trying to understand if i can use Windows based applications on LSTP server on linux for windows 10 workstations
| |
| 06:33 | sip phones and microsoft excel are the applications we are looking to install
| |
| 06:33 | Please help
| |
| 06:37 | Vb has left IRC (Vb!671f90f4@103.31.144.244, Remote host closed the connection) | |
| 07:11 | <alkisg> vagrantc, I didn't put jetpipe to ltsp19 because it was python2, but also because I wanted to see if there's still need for it, now with only fat clients and cups and driverless printing
| |
| 07:11 | A last ltsp release sounds good, maybe even addressing that ldm security issue
| |
| 07:12 | If jetpipe in python3 doens't work (e.g. due to removed/wrong dependencies), it would be best if it was released with python2 though, we don't want to break printing in that last release
| |
| 07:30 | woernie has joined IRC (woernie!~werner@p5B296964.dip0.t-ipconnect.de) | |
| 09:07 | Klimm has joined IRC (Klimm!~Georg@p54897B9A.dip0.t-ipconnect.de) | |
| 09:08 | ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz) | |
| 09:08 | statler has left IRC (statler!~Georg@84.137.115.31, Ping timeout: 258 seconds) | |
| 09:47 | ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Remote host closed the connection) | |
| 09:49 | ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz) | |
| 10:36 | ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Remote host closed the connection) | |
| 10:36 | ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz) | |
| 13:06 | ZAJDAN has joined IRC (ZAJDAN!~zdenek@77.48.149.75) | |
| 13:07 | GodFather has joined IRC (GodFather!~rcc@d53-64-7-141.nap.wideopenwest.com) | |
| 14:33 | ZAJDAN has left IRC (ZAJDAN!~zdenek@77.48.149.75, Quit: Konversation terminated!) | |
| 14:33 | ZAJDAN has joined IRC (ZAJDAN!~zdenek@77.48.149.75) | |
| 14:44 | mmarconm has joined IRC (mmarconm!~mmarconm@unaffiliated/mmarconm) | |
| 18:13 | mmarconm has left IRC (mmarconm!~mmarconm@unaffiliated/mmarconm, Read error: Connection reset by peer) | |
| 19:09 | Klimm has left IRC (Klimm!~Georg@p54897B9A.dip0.t-ipconnect.de, Remote host closed the connection) | |
| 19:19 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
| 19:28 | <alkisg> vagrantc: re: jetpipe/python2, see irclogs...
| |
| 19:30 | <vagrantc> !irclog
| |
| 19:30 | <ltsp> I do not know about 'irclog', but I do know about these similar topics: 'irclogs'
| |
| 19:30 | <vagrantc> !irclogs
| |
| 19:30 | <ltsp> irclogs: This channel is logged, archives are available at http://irclogs.ltsp.org
| |
| 19:32 | <vagrantc> python2 is removed from debian, so it's either drop jetpipe or use python3
| |
| 19:32 | at least from my perspective
| |
| 19:32 | <alkisg> When you say last release, do you mean for buster, which supports python2, or for bullseye?
| |
| 19:32 | Afaik bullseye will get ltsp19, won't it?
| |
| 19:33 | <vagrantc> not really possible to do any significant changes for buster
| |
| 19:33 | it's released already
| |
| 19:33 | <alkisg> so the last release will e.g. break 18.04 or other existing users, and benefit who?
| |
| 19:33 | <vagrantc> at least, not within buster
| |
| 19:34 | if we want to support both concurrently, backports to buster, etc.
| |
| 19:34 | why will it break 18.04 ?
| |
| 19:34 | surely 18.04 has python3 ...
| |
| 19:34 | <alkisg> assuming jetpipe doesn't work correctly with python3
| |
| 19:35 | <vagrantc> well, the changes looked pretty minimal
| |
| 19:35 | might even work with either
| |
| 19:35 | <alkisg> are all the dependencies there?
| |
| 19:36 | <vagrantc> only one i might wonder is python3-serial
| |
| 19:36 | packages.ubuntu.com doesn't seem to be working; can you check?
| |
| 19:37 | it would also remove the need for python2 in ltsp at all; ldm was converted to python3 a while ago, so ltsp+ldm requires both
| |
| 19:37 | (although there's probably some other random stuff using python2)
| |
| 19:37 | the other option is just committing to dropping ltsp 5.x for bullseye
| |
| 19:38 | but ltsp will be removed from bullseye soon unless the python2 issue is fixed
| |
| 19:39 | so could revert the upstream change i made, and apply it to debian packaging only...
| |
| 19:39 | * alkisg is committed to only supporting ltsp19+ :) | |
| 19:39 | <vagrantc> sure, but we had talked about a transition period, too
| |
| 19:39 | <alkisg> But bullseye were to get ltsp19 in all cases
| |
| 19:39 | <vagrantc> e.g. the next ubuntu LTS release might have ltsp 5.x
| |
| 19:40 | <uumas> vagrantc: There's still about 2,5 years until bullseye is released while the next Ubuntu lts is in 8 months
| |
| 19:40 | <vagrantc> will ubuntu support python2 for the next LTS?
| |
| 19:40 | <alkisg> Possibly not
| |
| 19:41 | <vagrantc> well, python2 is officially unsupported upstream in just a few months
| |
| 19:41 | alkisg: so, then that's who would benefit from switching to python3...
| |
| 19:41 | <alkisg> If it works I don't have any issue with it
| |
| 19:42 | But if it's untested, it doesn't benefit anyone...
| |
| 19:42 | <vagrantc> https://pythonclock.org/ says 4 months
| |
| 19:42 | i don't have any way to test it
| |
| 19:42 | at least, not reliably
| |
| 19:43 | if there's no python2, it also doesn't benefit anyone to keep it :P
| |
| 19:43 | <alkisg> 18.04 still gets new releases from the ppa
| |
| 19:43 | So when anything breaks, I fix it
| |
| 19:44 | * vagrantc is curious how long LTS distros will continue to support python2 | |
| 19:45 | <vagrantc> surely ubuntu has a plan for this?
| |
| 19:45 | <uumas> vagrantc: Haven't heard an official word, but I'd say it should be dropped by 22, possibly 20
| |
| 19:46 | <alkisg> ubuntu 18.04 is supported for 5 years
| |
| 19:46 | they'll support python2 till then
| |
| 19:46 | <vagrantc> https://askubuntu.com/questions/1127122/python2-support-in-18-04-lts
| |
| 19:48 | * vagrantc would like to find something more official... | |
| 19:48 | <uumas> Oh yes, of course they'll have to support it for the support cycle of the last release which includes it, but the question is if that release is 18 or 20
| |
| 19:49 | <vagrantc> "have to support it" has sometimes meant dropping support for some packages.
| |
| 19:54 | <uumas> vagrantc: What's debian's stance on it? Buster is to be supported until 2024 and includes python2.
| |
| 19:54 | <vagrantc> good question
| |
| 19:55 | * alkisg can't believe they would drop 500 or so packages along with python2 | |
| 19:56 | <uumas> They can't drop the packages as that WOULD break lots of things, but the really can't really keep unsupprted packages in the official repos either. I.e.
| |
| 19:56 | They just have to support it
| |
| 19:56 | <alkisg> How so? Isn't that what they've always been doing?
| |
| 19:56 | I've seen lots and lots of unsupported packages; they just got removed in the next release
| |
| 19:57 | Ubuntu supports things in main, it doesn't care about universe, but python2 is in main
| |
| 19:58 | <vagrantc> anyways, we've seen this trainwreck coming from a long ways off ...
| |
| 19:58 | thankfully, not a huge impact for ltsp at this point
| |
| 19:58 | worst case is remove jetpipe
| |
| 19:59 | debian has dropped support for major known packages in the past in point releases
| |
| 20:00 | <uumas> But this is python we're talking about
| |
| 20:00 | It would break thousands of systems
| |
| 20:00 | <alkisg> What about the other python2 scripts in ltsp5? Are they automatically detected and will trigger a removal, even if they're not important for us?
| |
| 20:00 | For example... ltspfs/doc/examples/kde-desktop-icons:#!/usr/bin/python
| |
| 20:00 | ltsp/server/k12linux/scripts/chroot-creator:#!/usr/bin/python -tt
| |
| 20:00 | etc etc, lots of scripts that we might not even need, but they're there
| |
| 20:01 | translations/ldm/generate_locale_list.py:#!/usr/bin/python
| |
| 20:02 | <vagrantc> pretty sure generate_locale_list was made to work with either at some point
| |
| 20:02 | <alkisg> Will the shebang trigger the warning bot? It's not "python3" there...
| |
| 20:02 | <vagrantc> i didn't find any in ltsp itself that would be an issue from Debian's perspective ... i don't think anyone's usiong k12linux, for example
| |
| 20:03 | it might issue a warning, but i won't let it get removed over that
| |
| 20:03 | only thing that really matters is code that's actually used
| |
| 20:03 | * uumas sent a long message: < https://matrix.org/_matrix/media/v1/download/matrix.org/OeAwgxMHWMNVZiruXaXylFOF > | |
| 20:04 | <alkisg> (11:03:52 PM) ***uumas sent a long message: < https://matrix.org/_matrix/media/v1/download/matrix.org/OeAwgxMHWMNVZiruXaXylFOF >
| |
| 20:04 | That matrix has its shortcomings :D
| |
| 20:04 | <vagrantc> not up for clickbait :P
| |
| 20:04 | <alkisg> !ltsp-source
| |
| 20:04 | <ltsp> ltsp-source: https://github.com/ltsp/ltsp/
| |
| 20:04 | <alkisg> !ltsp5-source
| |
| 20:04 | <ltsp> ltsp5-source: https://code.launchpad.net/ltsp
| |
| 20:05 | <uumas> <alkisg "(11:03:52 PM) ***uumas sent a lo"> Oh, that's interesting
| |
| 20:06 | <alkisg> Wth, `git pull` tells me up to date, `git remote -v` says launchpad, yet I don't see commits after May...
| |
| 20:06 | OK wrong branch, on debian
| |
| 20:07 | Well, at least jetpipe runs (I can't test printers remotely)... that's a good sign
| |
| 20:08 | <vagrantc> thanks for testing that much :)
| |
| 20:08 | <alkisg> if o in ("-h", "--help"): # help text
| |
| 20:08 | sys.exit()
| |
| 20:08 | Hehe nice :D
| |
| 20:08 | ogra really hates man pages AND help messages
| |
| 20:09 | <vagrantc> alkisg: ah, for ldm i've patched to use bin/python3 in the shebangs, but didn't push upstream
| |
| 20:09 | so i haven't found any bin/python shebangs i'd be worried about
| |
| 20:09 | * alkisg really gets confused with the debian patches that belong upstream... | |
| 20:10 | <vagrantc> not all distros ship "python3"
| |
| 20:11 | in some python is python 3.x
| |
| 20:11 | <alkisg> Afaik pep says python3, so we don't care what those distros say
| |
| 20:11 | <vagrantc> oh
| |
| 20:11 | python enhancement proposal?
| |
| 20:11 | <alkisg> Let me verify...
| |
| 20:12 | <uumas> I guess python will be python3 on debian too once python2 is dropped
| |
| 20:12 | <vagrantc> not sure on that
| |
| 20:12 | which is why i used a distro-specific patch
| |
| 20:12 | <alkisg> https://www.python.org/dev/peps/pep-0394/ => " python3 should run some version of the Python 3 interpreter."
| |
| 20:12 | So python3 is guaranteed, python is not
| |
| 20:13 | If the python command is installed, it is expected to invoke either the same version of Python as the python3 command or as the python2 command.
| |
| 20:13 | <vagrantc> ok, let's push that upstream to ... LDM :P
| |
| 20:13 | <alkisg> They may even choose to not provide a python command at all. But python3 needs to be there.
| |
| 20:15 | <uumas> Apparently on Ubuntu (and I assume debian) `python` is just a symlink to `python2.7` (and on arch it's a symlink to `python3`)
| |
| 20:15 | I'd assume debian will just change that symlink to `python3`
| |
| 20:15 | <alkisg> Essentially scripts shouldn't be using "python" in shebangs anymore
| |
| 20:15 | <vagrantc> right
| |
| 20:15 | <alkisg> It's either python2 or python3
| |
| 20:16 | <uumas> FTFY: It's python3
| |
| 20:17 | <vagrantc> ftfy?
| |
| 20:17 | <uumas> *fixed that dor you
| |
| 20:17 | s/dor/for/
| |
| 20:18 | <vagrantc> alkisg: ok, well, that's clear at least :)
| |
| 20:18 | kjackal has joined IRC (kjackal!~quassel@2a02:587:3110:a100:a551:b0ce:a287:d7de) | |
| 20:18 | <alkisg> vagrantc: if you issue a new release, that means the PPA builds will start working again, right?
| |
| 20:19 | If I recall correctly, the last time you had some patches that were applied upstream and required that I maintained another debian dir, which I didn't find time to do...
| |
| 20:19 | <vagrantc> alkisg: not sure what you mean...
| |
| 20:19 | <alkisg> So I think that if you do another release now, builds will work again, which will be great
| |
| 20:20 | ltsp - 5.18.12+t201905250652~ubuntu18.04.1 ==> ok last build in May, not bad
| |
| 20:20 | <vagrantc> i don't see any patches in LDM that are applied upstream
| |
| 20:21 | <alkisg> I think it was in ltsp
| |
| 20:21 | <vagrantc> i see
| |
| 20:21 | yeah, i should be able to upstream all the patches
| |
| 20:21 | <alkisg> Anyway I mean I'm glad if you're planning to do a last release, I think it'll help
| |
| 20:22 | <vagrantc> well, can always release the python3 jetpipe and if nobody complains... then it either works or nobody's using it
| |
| 20:22 | <alkisg> I'd need to upload to ppa though, at least for ubuntu users to get it
| |
| 20:22 | Not sure how debian users get the new releases (since buster is released now)
| |
| 20:23 | so I'd need to test it first...
| |
| 20:23 | But sure, let's do that
| |
| 20:24 | uumas, where you the one that reported the ldm security issue?
| |
| 20:24 | <uumas> Yes
| |
| 20:24 | <alkisg> Did you check the code for a fix?
| |
| 20:24 | <vagrantc> i'm kind of dreading that issue
| |
| 20:25 | <uumas> alkisg: What code?
| |
| 20:25 | <alkisg> The ldm code, in case you could see where the problem was...
| |
| 20:25 | !ldm-source
| |
| 20:25 | <ltsp> ldm-source: at https://code.launchpad.net/~ltsp-upstream/ltsp/+git/ldm
| |
| 20:25 | <alkisg> ...something like that...
| |
| 20:27 | <uumas> Oh, no I didn't. I can have a look, but I have little experience with coding, so I'm probably not the best to spot that.
| |
| 20:28 | * alkisg is busy with ltsp19 and vagrantc doesn't even have any ltsp installations around, so ltsp5 can use all the help it can get... :/ | |
| 20:47 | woernie has left IRC (woernie!~werner@p5B296964.dip0.t-ipconnect.de, Remote host closed the connection) | |
| 20:50 | <alkisg> vagrantc: ah, I just remembered what your last proposal was. To upload ltsp5 under a new source name, maybe ltsp5? so that both the new and old ltsp can be available for a period of time
| |
| 20:51 | For 20.04 users, that would be the best option
| |
| 20:51 | <vagrantc> alkisg: yes
| |
| 20:51 | it may involve some interesting dances
| |
| 20:52 | <alkisg> I think that would attract more users to the new ltsp, and make it more stable for bullseye, where it will be the only option...
| |
| 20:53 | About the new ltsp, I think I'll fix the tmpfs issue that uumas reported with new kernels, and make ltsp.ipxe completely auto-generated, and release it as 19.09 beta
| |
| 20:54 | ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving) | |
| 20:54 | <alkisg> And then spend some time on getting "sch-scripts" working with the new ltsp, so that I can put it in a few school next week
| |
| 20:55 | ok pumpkin time... byez for now :)
| |
| 20:56 | * vagrantc waves | |
| 21:08 | kjackal has left IRC (kjackal!~quassel@2a02:587:3110:a100:a551:b0ce:a287:d7de, Ping timeout: 252 seconds) | |
| 23:30 | <vagrantc> uumas: i haven't yet figured out why, but the LDM_USERNAME variable is unset for fish, csh and tcsh...
| |
| 23:31 | the quick security fix is to exit when the username fails...
| |
| 23:31 | er, when the username is unset.
| |