sphalerite changed the topic of #nixos-dev to: NixOS Development (#nixos for questions) | NixOS 19.03 released! https://discourse.nixos.org/t/nixos-19-03-release/2652 | https://hydra.nixos.org/jobset/nixos/trunk-combined https://channels.nix.gsc.io/graph.html https://r13y.com | 19.03 RMs: samueldr,sphalerite | https://logs.nix.samueldr.com/nixos-dev
drakonis has joined #nixos-dev
<worldofpeace> Profpatsch: with jisti meet I think the call will just break down with lots of people
<worldofpeace> i don't particulally care what I have to use to have this, I just want to use a platform that is alligned to the needs of the hosts and has a low barrier to access for the participants
<qyliss> Having to use proprietary software would probably be enough to dissuade me from participating, fwiw
drakonis has quit [Quit: Leaving]
orivej has quit [Ping timeout: 258 seconds]
<worldofpeace> exactly, it can be a barrier to access for participants
<worldofpeace> so we have to use an open platform. it's understanable that this would dissuade lots of people in this environment
sir_guy_carleton has joined #nixos-dev
<Shados> infinisil: Main downside of rocket chat compared to mumble is that it doesn't actually have voice *channels*, just ad-hoc conference calls. They also have no plans to add that.
drakonis has joined #nixos-dev
drakonis_ has quit [Ping timeout: 258 seconds]
<Profpatsch> Mumble might actually be a quite good platform, bar the screen sharing.
drakonis_ has joined #nixos-dev
drakonis has quit [Ping timeout: 252 seconds]
<worldofpeace> i'm seeing "voice communication" and no mention of video. It has to have video.
<worldofpeace> to be clear, video participation is encouraged not required.
drakonis has joined #nixos-dev
harrow has quit [Quit: Leaving]
ivan has quit [Read error: Connection reset by peer]
harrow has joined #nixos-dev
ivan has joined #nixos-dev
drakonis has quit [Quit: WeeChat 2.4]
drakonis_ has quit [Ping timeout: 246 seconds]
drakonis_ has joined #nixos-dev
harrow has quit [Read error: Connection reset by peer]
ivan has quit [Read error: Connection reset by peer]
ivan has joined #nixos-dev
harrow has joined #nixos-dev
sir_guy_carleton has quit [Quit: WeeChat 2.4]
FRidh has joined #nixos-dev
<FRidh> hydra died yesterday https://hydra.nixos.org/machines
<srhb> Uh oh
callahad has quit [Ping timeout: 258 seconds]
callahad has joined #nixos-dev
vcunat has joined #nixos-dev
<vcunat> the /machines page hasn't worked for several days already but it built packages just fine, until about five hours ago
<vcunat> (time estimate by looking at https://hydra.nixos.org/steps )
<vcunat> Maybe the two problems aren't really related.
<niksnut> I'll have a look
<niksnut> hm, a lot of errors from the GitlabStatus plugin in the log
<niksnut> so it looks like hydra-queue-runner is spending a few hours at startup running notification plugins that shouldn't even be running
Orbstheorem has left #nixos-dev ["WeeChat 1.8"]
<globin> niksnut: do you have some log?
<globin> niksnut: ah sry, thanks for fixing that..
Synthetica has joined #nixos-dev
orivej has joined #nixos-dev
qyliss^work has quit [Quit: bye]
qyliss has quit [Quit: bye]
qyliss has joined #nixos-dev
qyliss^work has joined #nixos-dev
ixxie has joined #nixos-dev
<ekleog> qyliss: fwiw zoom is accessible without proprietary software (web client in chromium)
<ekleog> it's what I do for the exact same reason as you
<qyliss^work> proprietary javascript though, presumably?
<ekleog> qyliss^work: well, proprietary javascript is something you'll have on basically all webpages, while I totally agree that it's not ideal it's something that I think we can reasonably accept (eg. in France you cannot legally refuse proprietary javascript, as the tax website requires javascript to work and it's illegal [to the lowest level] to pay taxes by post nowadays)
<gchristensen> qyliss^work: is that a hypothetical?
<ekleog> domenkozar[m]: RFC on setup.nix having two links one after the other is weird
<gchristensen> domenkozar[m]: maybe indicate that the ZFS one is a link to Phoronix, which adds verisimilitude
<srhb> domenkozar[m]: It's not very obvious that "project source" is not part of the same link for "RFC on setup.nix [...]" -- maybe add parens. Otherwise LGTM
<qyliss^work> Just because proprietary JavaScript is difficult to avoid doesn't make it any better than other forms of proprietary software.
<qyliss^work> gchristensen: not sure I understand the q?
<gchristensen> is that a concern you hold, or is that a concern you hypothesise other people might hold?
<ekleog> qyliss^work: it does run in a sandbox, which makes it better than most other forms of proprietary software to me
<qyliss^work> It's a concern I hold.
<domenkozar[m]> srhb: ekleog gchristensen fixed, anything else? :)
<qyliss^work> weekly lgtm
<fadenb> worldofpeace: I used a tiny server wit jitsi meet with hundres of people without any issues. It scales very well
<domenkozar[m]> thanks!
<gchristensen> scale is not my concern :/
<domenkozar[m]> gchristensen: do you have an opinion on https://github.com/NixOS/nix/issues/2542#issuecomment-492577156
<gchristensen> egh
<gchristensen> LnL: ^ any opinions?
<gchristensen> I'm inclined to also allow /private/var/empty. I do think we should set it to something if it isn't that, though
<domenkozar[m]> it's really error prone to parse the output
<gchristensen> it is?
<domenkozar[m]> current implementation fails to parse the current value
<gchristensen> it does?
<worldofpeace> tbh one can spend lots of time looking for a platform, and become wanderlust. I'd like too be use something open really, but zoom is like nice idk :P. fadenb I've had different experiences with just the public one
<gchristensen> domenkozar[m]: I hate it :(
orivej has quit [Ping timeout: 245 seconds]
<domenkozar[m]> :/
ixxie has quit [Ping timeout: 258 seconds]
<niksnut> it seems just a matter of time until we have to drop multi-user support on macOS, since apple seems determined to make that sort of thing impossible
<gchristensen> I, very unhappily and reluctantly, agree
<agent13> /quit
<gchristensen> agreed
agent13 has quit [Remote host closed the connection]
<domenkozar[m]> I don't understand on basis of what? :)
<domenkozar[m]> I'd really like to fix this bug as just commenting out user creation works well.
<LnL> gchristensen: /var is a symlink to private/var so those are equivalent
<domenkozar[m]> ah
<LnL> gchristensen: but this is the generalised part, I don't think we want /private/var/empty on linux
<domenkozar[m]> then the problem is in parsing
<domenkozar[m]> $ sudo /usr/bin/dscl . -read /Users/nixbld1 NFSHomeDirectory<br />
<domenkozar[m]> NFSHomeDirectory:
<domenkozar[m]> /private/var/empty 1
<domenkozar[m]> this is parsed as one line
<domenkozar[m]> but I don't know what it was supposed to catch
<LnL> it's to avoid updating when the user already exists?
<domenkozar[m]> LnL: are you on mojave?
<LnL> yeah
<domenkozar[m]> do you get the same output?
<domenkozar[m]> maybe this never worked really.
<LnL> and I'm pretty sure my laptop was when I got it
orivej has joined #nixos-dev
<LnL> works fine
<gchristensen> it maybe doesn't/never has worked on Mojave, but it definitely does and did work
hedning_ has quit [Remote host closed the connection]
<domenkozar[m]> I should be more precise: I think the old functions of getting home directory never worked
<domenkozar[m]> but then changing /var/empty to /var/empty did work
<gchristensen> it definitely did
<gchristensen> although I notice my link ^ doesn't demonstrate that :)
<gchristensen> actually it does
<gchristensen> because if it didn't work, it would show a log message about updating the home directory
<domenkozar[m]> in that case the output changed
<gchristensen> yeah, dscl changes its output under ... circumstances ...
<domenkozar[m]> maybe it should check if string contains /var/empty..
<LnL> oh hold on, this is remote
<domenkozar[m]> remote in what way?
<LnL> ssh
<LnL> that might be a problem for the weird permission stuff with dscl
<domenkozar[m]> I'm so happy Microsoft will wipe out macs guys.
<domenkozar[m]> they seem to at least care :)
<domenkozar[m]> LnL: I'm still confused why we need to set home for a freshly created user though
<LnL> we don't have to anymore it seems, the default used to be /Users/foo I think
<domenkozar[m]> should we then just make adding user on mojave a noop?
rsa has joined #nixos-dev
<gchristensen> it tries to account for people who have setup nix build users improperly using the assorted quality of "nix daemon install!" scripts out there
<domenkozar[m]> seems like a good time to drop that? :)
<gchristensen> not sure
<gchristensen> maybe? :)
<samueldr> could ~user be used instead?
<samueldr> if the issue is with dscl
<gchristensen> what do you mean?
<samueldr> letting bash expand
<gchristensen> oh cool
<samueldr> echo ~nixbld1
<samueldr> not positive it will be able to if dscl isn't
<gchristensen> dscl can, it is just outputting in the silly format
<samueldr> right
<samueldr> (I was going off on the "this is remote" tangent)
vcunat has quit [Quit: Leaving.]
<domenkozar[m]> nice
Jackneill has quit [Read error: Connection reset by peer]
<domenkozar[m]> ah but that needs eval due to variables
drakonis has joined #nixos-dev
<domenkozar[m]> $ echo ~nixbld1
<domenkozar[m]> /private/var/empty 1
<domenkozar[m]> what's with that extra 1?
<samueldr> that's using the macOS 3.x series bash?
<samueldr> wondering if it matters and if there is a behaviour difference there
<gchristensen> well it sounds like each one has its own empty home dir
<gchristensen> ie, /private/var/empty\ 1"
_e has quit [Quit: WeeChat 2.4]
_e has joined #nixos-dev
drakonis has quit [Quit: WeeChat 2.4]
<matthewbauer> domenkozar: that's really weird. maybe somewhere they enforce that each user has a unique home dir?
<gchristensen> they don't even require each user have a unique ID
<matthewbauer> Weird! I thought we hit a bug in the installer related to id conflicts?
<matthewbauer> Also can someone confirm this happens: https://github.com/NixOS/nix/commit/b6eb8a2d7e2ea8b083fdac15f537679ffe633183#commitcomment-33548097 ? I thought I reported the issue somewhere but can't find it
orivej has quit [Ping timeout: 258 seconds]
Jackneill has joined #nixos-dev
drakonis has joined #nixos-dev
drakonis_ has quit [Ping timeout: 244 seconds]
drakonis has quit [Ping timeout: 258 seconds]
<{^_^}> LnL7/nix-darwin#140 (by teehemkay, 5 days ago, open): `nix search` needs an explicitly name `nixpkgs` entry in `NIX_PATH`
<LnL> I would rather change the behaviour of installabels tho, unless there's a good reason it's needed
<LnL> not needing -f '<nixpkgs>' would be even better, but it's probably too late to change that now :/
<ryantm> zimbatm: nixpkgs-update now has a proper license https://github.com/ryantm/nixpkgs-update/blob/master/LICENSE
<zimbatm> ryantm++
<{^_^}> ryantm's karma got increased to 6
cocreature has quit [Quit: A monad is just a monoid in the category of endofunctors, what's the problem?]
cocreature has joined #nixos-dev
orivej has joined #nixos-dev
pie___ has joined #nixos-dev
avn has quit [Ping timeout: 258 seconds]
avn has joined #nixos-dev
avn has quit [Quit: Lost terminal]
avn has joined #nixos-dev
<arianvp> Does packet support IPv6?
<gchristensen> absolutely yes
<arianvp> What subnet size do you get?
<gchristensen> by default each device gets 1, but I think you can get a good hunk with BGP
alp has quit [Remote host closed the connection]
dtz has joined #nixos-dev