andi- changed the topic of #nixos to: NixOS stable: 19.09 \o/ https://discourse.nixos.org/t/nixos-19-09-release/4306/2 || https://nixos.org || Latest NixOS: https://nixos.org/nixos/download.html || Latest Nix: https://nixos.org/nix/download.html || Logs: https://logs.nix.samueldr.com/nixos/ || use ,channels for a list of Nix* related channels || nixcon videos: https://tinyurl.com/nixcon2019 || Link to the output of nix-info
Pwnna has quit [Quit: Bye]
<{^_^}> [nixpkgs] @NotGate opened pull request #84143 → update libbass → https://git.io/JvFr5
ambroisie has joined #nixos
shibboleth has quit [Quit: shibboleth]
<cole-h> What do we do about duplicate PRs? ^ duplicates an older, yet-to-be-merged PR: https://github.com/NixOS/nixpkgs/pull/76692
<{^_^}> #76692 (by iimog, 13 weeks ago, open): libbass: 2.4.14 -> 2.4.15
oranginal has quit [Quit: Leaving]
mexisme has quit [Ping timeout: 246 seconds]
genevino has joined #nixos
thc202 has quit [Ping timeout: 256 seconds]
thetet has quit [Quit: Leaving.]
justanotheruser has joined #nixos
<simpson> I have a Nix expression which builds a link farm from a tree of source files; it recursively explores the tree for source files which it knows how to compile, and individually builds each file.
<simpson> I want to build a second farm, using the first farm as part of the executable code for the builder. (This is a bootstrapping step; some source files should be compiled after the initial compile.) But I also want the two farms to be merged. Is there a Nix idiom for this, or should I build the two farms independently and then merge them?
growpotkin has quit [Quit: ZNC 1.7.5 - https://znc.in]
growpotkin has joined #nixos
ambroisie has quit [Ping timeout: 240 seconds]
Soo_Slow has quit [Quit: Soo_Slow]
ixxie has quit [Ping timeout: 265 seconds]
<{^_^}> [nixpkgs] @enolan opened pull request #84144 → minecraft: 2.1.11314 -> 2.1.13509 → https://git.io/JvFoW
blumenkranz has joined #nixos
<lovesegfault> Mesa 20 is still not on unstable small :O
* lovesegfault cries in staging
<blumenkranz> Hello. I'm attempting to deploy to a libvirtd environment with NixOps. Everything is fine until it gets to the point where postgresql attempts to start, but I get a "timeout" error, and I really don't get what am I doing wrong.
<blumenkranz> Apparently there used to be an issue with 19.09 (the version I am using for the environment) where this happened, but it was fixed in master. Does anybody know if the current 19.09 tag is bugged?
<bqv[m]> lovesegfault: you made me check if in advanced again :/
<bqv[m]> It*
<lovesegfault> It hasn't
<lovesegfault> and there are 60k builds on the staging-next queue
<blumenkranz> Do old NixOS versions still get security updates? I'm considering downgrading.
<b42> blumenkranz: i've got 19.09 with postgresql running in libvirtd guest and haven't run into this - anything interesting in the logs?
<lovesegfault> 142k builds queued in hydra total
<blumenkranz> b42: That's the thing, there is no /var/log/postgresql
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JvFow
<blumenkranz> I am baffled. Will try with systemctl now, but the console hasn't spewed anything interesting
<infinisil> blumenkranz: Nope, old versions only get updates for ~1 month after a new version is released
<b42> blumenkranz: journalctl -b -u postgresql
<b42> ?
mexisme has joined #nixos
bebarker has joined #nixos
<b42> or maybe even "journalctl -b" to see all messages not just from postgres
<blumenkranz> b42: Wow. The DB does startup correctly, but then systemd just decides it has timeout'd and it proceeds to kill it
<b42> o_O
<blumenkranz> b42: I get "database system is ready to accept connections", then immediately after, systemd kills the service
bbarker has quit [Ping timeout: 256 seconds]
<blumenkranz> Thanks Lenart Poettering
<blumenkranz> postgresql.service: Start-post operation timed out. Stopping.
mexisme has quit [Ping timeout: 246 seconds]
johnw has quit [Quit: ZNC - http://znc.in]
<b42> blumenkranz: got any initialScript/ensureUsers/ensureDatabases? looks like that's where service times out
<blumenkranz> b42: Not really, I just have enabled the postgres service. But I also get the following line
<blumenkranz> n2ig6d12rk5p00bgim3lzn2cz36csd1k-unit-script-postgresql-post-start[3728]: /nix/store/n2ig6d12rk5p00bgim3lzn2cz36csd1k-unit-script-postgresql-post-start: line 5: kill: (3719) - No such process
<blumenkranz> It's after the DB has already shut down. I guess it has nothing to do
<blumenkranz> Just for reference, are you using commit d5291756487d70bc336e33512a9baf9fa1788faf, b42?
<{^_^}> [nixpkgs] @sikmir opened pull request #84145 → qmapshack: add gdal/routino tools to PATH → https://git.io/JvFoF
jlv52 has quit [Remote host closed the connection]
<b42> blumenkranz: i'm on d011e4749457af484adf2e90062c83a44ad072a4 - d529175 looks quite old
<b42> hmm, perhaps even the first 19.09 commit?
<lopsided98> blumenkranz: you shouldn't use the 19.09 tag; use the release-19.09 branch
<energizer> https://nixos.wiki/wiki/Printing says "Navigate to http://localhost:631/ in a web browser to configure printers." is it possible to configure the printer with nix instead?
<blumenkranz> b42: Yeah, I know for sure 19.09 used to have this same bug at some point. Might just be using an old version.
<blumenkranz> lopsided98: Ah, that might be what's wrong. Thanks for the tip!
<b42> you should be using the nixos-19.09 branch actually
<blumenkranz> Is there any place where I can read about how does nixpkgs channels versioning work, and how does support work for older versions?
<b42> it follows release-19.09 but must pass CI and there are guaranteed(tm) packages in binary cache
rogue_koder has quit [Ping timeout: 260 seconds]
<blumenkranz> Will 19.09 be updated in case of vulnerabilities?
tertl3 has joined #nixos
<blumenkranz> I mean, nixos-19.09
rogue_koder has joined #nixos
<blumenkranz> I'm kinda new to channels. Been using nixos-unstable all this time because I need muh bleeding edge, but now I want some stability for this server.
<ryantm> The Las Vegas Functional Programming Group meetup https://www.meetup.com/las-vegas-functional-programming/events/mfmwqrybcgbdb/ is going to be meeting via Jitsi in 10 minutes ttps://meet.jit.si/lvfpug I'm going to talk about Niv and whatever else people are interested in.
<ryantm> Fixed link: https://meet.jit.si/lvfpug
spacefrogg has quit [Quit: Gone.]
aw has quit [Quit: Quitting.]
jbrock has joined #nixos
aw has joined #nixos
<blumenkranz> Thanks b42. Definitely going with nixos-19.09
spacefrogg has joined #nixos
mexisme has joined #nixos
gustavderdrache has quit [Quit: Leaving.]
<{^_^}> [nixpkgs] @aanderse opened pull request #84146 → nixos/postgresql: add statements option to replace the ensureDatabases and ensureUsers options → https://git.io/JvFKt
<{^_^}> [nixpkgs] @marsam opened pull request #84148 → flow: 0.121.0 -> 0.122.0 → https://git.io/JvFKB
guest1234 has quit [Ping timeout: 240 seconds]
<{^_^}> [nixpkgs] @marsam opened pull request #84149 → abseil-cpp: 20191119 -> 20200225.1 → https://git.io/JvFKE
<{^_^}> [nixpkgs] @colemickens opened pull request #84150 → mirage: init → https://git.io/JvFKK
rogue_koder has quit [Remote host closed the connection]
rogue_koder has joined #nixos
h0m1 has quit [Ping timeout: 252 seconds]
h0m1 has joined #nixos
gxt has quit [Remote host closed the connection]
gxt has joined #nixos
growpotkin has quit [Ping timeout: 256 seconds]
<{^_^}> [nixpkgs] @JeffLabonte opened pull request #84151 → brave: 1.5.115 -> 1.5.122 → https://git.io/JvF6Y
Kyndig has joined #nixos
<{^_^}> [nixpkgs] @marsam opened pull request #84152 → lxd: 3.23 -> 4.0.0 → https://git.io/JvF6u
teto has quit [Ping timeout: 246 seconds]
Pwnna has joined #nixos
zebrag has joined #nixos
<{^_^}> [nixpkgs] @colemickens opened pull request #84153 → obs-v4l2sink: init → https://git.io/JvF6r
inkbottle has quit [Ping timeout: 256 seconds]
<Kyndig> Hello, is there a protocol for asking questions?
<simpson> Yes: Ask questions, get answers.
<Kyndig> :)
xkapastel has quit [Quit: Connection closed for inactivity]
bvdw has quit [Read error: Connection reset by peer]
<Kyndig> I'm running nixpkgs, single user with the store in my home dir on an old redhat server; created an overlay to make a custom <prefix>/etc/gitconfig. It worked and I see the config file in the store path, but git is trying to read /etc/gitconfig instead of the prefixed path (evidenced by an strace). Am I wrong to expect it to read the prefixed etc/?
bvdw has joined #nixos
<simpson> I'd expect it to look at /etc/ first. I'm not sure how one would point it at the store instead.
orivej has quit [Ping timeout: 260 seconds]
orivej_ has joined #nixos
<Kyndig> I thought that the build tool replaced all the paths with $out dir paths.
<Kyndig> of course ~/ would still be ~/
<Kyndig> so i would expect .gitconfig to work
<{^_^}> [nixpkgs] @eadwu opened pull request #84154 → clight: include location assertion → https://git.io/JvF6D
<{^_^}> [nixpkgs] @chkno opened pull request #84155 → darcs: Fix build (Use old GHC) → https://git.io/JvF69
<ottidmes> Kyndig: not paths like /etc, just libs and bins, AFAIK
<Kyndig> ahh
<Kyndig> so i should be running nixos, not nixpkgs
<Kyndig> well i am at home; unfortunately don't have a choice at work
<ottidmes> Kyndig: at work you could workaround it by making a wrapper setting $HOME to another path that contains your git config
<{^_^}> [nixpkgs] @marsam opened pull request #84156 → grpc: 1.27.1 -> 1.28.0 → https://git.io/JvF6F
<Kyndig> that works for intactive use
<Kyndig> but i'm trying to change the build behavior of all packages that use git to build
<{^_^}> [nixpkgs] @kylesferrazza opened pull request #84157 → change howoldis link to status page link → https://git.io/JvF6N
<Kyndig> namely the sslverify = false
<Kyndig> need it
<Kyndig> behind a proxy
<{^_^}> [nixpkgs] @eadwu opened pull request #84158 → plymouth: attempt to make flicker-free → https://git.io/JvF6A
<ottidmes> Kyndig: why could you not create a wrapper named git and give it prio on PATH, should have nothing to do with interactive use
<Kyndig> ahh; i see
<Kyndig> a wrapper that's a real store wrapper
<Kyndig> hmm; that should work
<Kyndig> dunno why i didn't think of it
<Kyndig> thanks
<ottidmes> No problem, good luck!
<Kyndig> i'm kinda slow
<Kyndig> new
<Kyndig> :)
coco has quit [Ping timeout: 258 seconds]
<ottidmes> Slow but steady works too :) And it takes some time to learn all the tricks needed to work comfortably with Nix(OS)
<blumenkranz> b42: Hey, I now get "error: cannot figure username" when attempting to deploy nixos-19.09. Apparently this is a known bug, but how did you manage to bypass it?
<{^_^}> [nixpkgs] @marsam opened pull request #84159 → skim: 0.8.0 -> 0.8.1 → https://git.io/JvFit
orivej_ has quit [Ping timeout: 265 seconds]
pbogdan has quit [Ping timeout: 256 seconds]
morgrimm has joined #nixos
orivej has joined #nixos
<morgrimm> Hey folks - would it be possible to write a derivation for a shell that simply builds as normal, but moves dotfiles into place? (maybe evaluating them with the current variables in scope?)
<morgrimm> i.e. build fish as normal, but post build, but after build move a set of dotfiles into `$out/etc/fish` ?
<{^_^}> [nixpkgs] @colemickens opened pull request #84160 → quaternion: update, libQuotient: init → https://git.io/JvFiY
cjpbirkbeck has quit [Quit: cjpbirkbeck]
slack1256 has joined #nixos
ottidmes has quit [Ping timeout: 260 seconds]
kandinski has joined #nixos
<DigitalKiwi> anyone seen ethernet wired in reverse?
iyzsong has joined #nixos
rardiol has quit [Ping timeout: 260 seconds]
<samueldr> hm?
<energizer> morgrimm: sounds like home-manager
<DigitalKiwi> so say you have a type B from the flat side it's left to right; orange white; orange; green white; blue; blue white; green; brown white; brown; these were wired like that but from right to left
rardiol has joined #nixos
<morgrimm> energizer: ... Yeah, you're right - home-manager totally works
<morgrimm> that's sweet
<DigitalKiwi> so instead of pin 1-1 2-2 3-3 4-4 5-5 6-6 7-7 8-8 it was 1-8 2-7 3-6 4-5 5-4 6-3 7-2 8-1
<DigitalKiwi> but only on one end the other end was in the right order (but in wall)
jlv has joined #nixos
orivej has quit [Read error: Connection reset by peer]
<davidak[m]> DigitalKiwi is it something like this https://en.wikipedia.org/wiki/Ethernet_crossover_cable ?
orivej has joined #nixos
<DigitalKiwi> no
<davidak[m]> sounds strange
<davidak[m]> maybe it's not for ethernet, but a proprietary protocol. only use the same socket?
alevy has joined #nixos
<jlv> Either I'm missing something or `services.xserver.inputClassSections` isn't working.
andi- has quit [Ping timeout: 256 seconds]
slack1256 has quit [Remote host closed the connection]
lovesegfault has quit [Quit: WeeChat 2.8]
rogue_koder has quit [Ping timeout: 265 seconds]
<DigitalKiwi> https://www.dropbox.com/s/9b87h9uq6mly07r/2020-04-02%2021.44.33.jpg?dl=0 the one on the right is right, the one on the left is how they were
andi- has joined #nixos
felixfoertsch23 has joined #nixos
felixfoertsch has quit [Ping timeout: 252 seconds]
felixfoertsch23 is now known as felixfoertsch
kaliumxyz has quit [Ping timeout: 265 seconds]
<{^_^}> [nixpkgs] @marsam closed pull request #84149 → abseil-cpp: 20191119 -> 20200225.1 → https://git.io/JvFKE
rogue_koder has joined #nixos
morgrimm has quit [Ping timeout: 240 seconds]
ArchFeh has joined #nixos
Maxdaman1us is now known as Maxdamantus
<DigitalKiwi> davidak[m]: did the picture help
<{^_^}> [nixos-homepage] @ryantm opened pull request #378 → add Discourse Team → https://git.io/JvFPw
oranginal has joined #nixos
<{^_^}> [nixos-homepage] @garbas merged pull request #378 → add Discourse Team → https://git.io/JvFPw
<{^_^}> [nixos-homepage] @garbas pushed 2 commits to governance: https://git.io/JvFPX
OrangeTetra has joined #nixos
<OrangeTetra> I've noticed some packages like KVM and vim-surround have other packages that look like the same thing... How does one tell the difference? (I'm mostly thinking from ad-hoc teriminal perspective)
rogue_koder has quit [Ping timeout: 260 seconds]
rogue_koder has joined #nixos
oranginal has quit [Ping timeout: 265 seconds]
knupfer has joined #nixos
<evils> OrangeTetra: what other packages?
<davidak[m]> DigitalKiwi i have never seen that. might be some special cable
<OrangeTetra> evils: KVM and qemu-KVM are the same, vim-surround and surround seem to be the same...
knupfer has quit [Remote host closed the connection]
<evils> > vimPlugins.surround.out
<{^_^}> "<derivation /nix/store/y8zdrm9z93k0vrcv0503djfvaaqk8lp7-vimplugin-vim-surround-2019-11-28.drv>"
<evils> > vimPlugins.vim-surround.out
knupfer has joined #nixos
<{^_^}> "<derivation /nix/store/y8zdrm9z93k0vrcv0503djfvaaqk8lp7-vimplugin-vim-surround-2019-11-28.drv>"
ArchFeh has quit [Quit: archfeh]
ArchFeh has joined #nixos
bbarker has joined #nixos
<OrangeTetra> So... They are the same?
alevy has quit [Quit: WeeChat 2.7.1]
<evils> yea, nix search shows them as the same in the parenthesis
<DigitalKiwi> davidak[m]: yeah...i'm not even sure what to what words to use to search for
rogue_koder has quit [Ping timeout: 256 seconds]
<OrangeTetra> kemu.out
<OrangeTetra> woops
<OrangeTetra> qemu.out
<evils> > kvm.out
<{^_^}> "<derivation /nix/store/rzfarqwww9pn64a88d3a6fj6bbm5gpfp-qemu-host-cpu-only-4.2.0.drv>"
<evils> > qemu.out
<{^_^}> "<derivation /nix/store/q5dlg9cvl40kw3wlrz9k9yx3i3dpwgr1-qemu-4.2.0.drv>"
<OrangeTetra> > qemu_kvm.out
<{^_^}> "<derivation /nix/store/rzfarqwww9pn64a88d3a6fj6bbm5gpfp-qemu-host-cpu-only-4.2.0.drv>"
captjakk has quit [Remote host closed the connection]
<evils> > kvm.pname
<{^_^}> "qemu-host-cpu-only"
bebarker has quit [Ping timeout: 265 seconds]
captjakk has joined #nixos
<{^_^}> [nixos-homepage] @ryantm opened pull request #379 → fixup Discourse team → https://git.io/JvFXv
<OrangeTetra> evils: Is there a way to get this information from my terminal?
<evils> nix repl
OrangeTetra has quit [Remote host closed the connection]
OrangeTetra has joined #nixos
captjakk has quit [Ping timeout: 265 seconds]
<evils> OrangeTetra: though nix search gives you the name attribute in the parenthesis, which is usually (pname-version)
rogue_koder has joined #nixos
hmpffff has joined #nixos
<OrangeTetra> evils: I only got the message starting with "though" did you send something before that? I look like you might have.
captjakk has joined #nixos
<evils> OrangeTetra: my reply was "nix repl", which is exactly what the bot uses for `> kvm.pname` etc
<OrangeTetra> Ah, okay thank you.
hmpffff_ has quit [Ping timeout: 246 seconds]
<evils> OrangeTetra: nix search --json will give you an explicit "pkgName" field, which is the pname attribute, and is usually the unique identifier for a package
<OrangeTetra> It outputs in a difficult to read format, though, so I probably won't use that.
<evils> OrangeTetra: pipe it to jq, like `nix search --json kvm | jq`
<{^_^}> [nixos-homepage] @garbas merged pull request #379 → fixup Discourse team → https://git.io/JvFXv
<{^_^}> [nixos-homepage] @garbas pushed 2 commits to governance: https://git.io/JvFXO
tertl3 has quit [Quit: Connection closed for inactivity]
<OrangeTetra> evils: ah, nice. Thank you.
<sasycto> I just noticed in my version of nix it seems like `du` is actually the binary for `sort`
Supersonic has quit [Disconnected by services]
Supersonic112 has joined #nixos
Supersonic112 is now known as Supersonic
<sasycto> apparently it's getting it from /run/current-system/sw/bin/du
<sasycto> I need to install whatever derivation has du then, clearly
<clever> sasycto: what do you get when you run `ls -l` on that path?
<sasycto> A lot of simlinks, and the du one links to coreutils-8.31 du, so I guess coreutils has a bug, lol
<sasycto> or the derivation is wrong
knupfer has quit [Remote host closed the connection]
<wrunt> I think busybox has a thing where all core commands (such as du and sort) are symlinks to a single binary, which behaves differently depending on arg[0]
<wrunt> maybe coreutils is the same
<wrunt> that would explain what you're seeing, sasycto
<pjt_014> clever: is there a platform option that covers just x86-64, i686 and aarch64?
<pjt_014> If it covers a few that are 'close enough' like gnu64 that's fine too
<clever> pjt_014: dont think there is one
knupfer has joined #nixos
knupfer has quit [Remote host closed the connection]
oida has quit [Remote host closed the connection]
oida has joined #nixos
captjakk has quit [Remote host closed the connection]
captjakk has joined #nixos
OrangeTetra has quit [Ping timeout: 250 seconds]
kaliumxyz has joined #nixos
captjakk has quit [Ping timeout: 256 seconds]
<samueldr> wrunt: coreutils has the same
<samueldr> single call binary
<sasycto> that's weird.
<samueldr> multi-call*
oranginal has joined #nixos
<oranginal> Is there a recommended method of setting a (say) GTK theme in NixOS?
<oranginal> or just... flipping into dark mode?
lovesegfault has joined #nixos
<{^_^}> [nixos-artwork] @samueldr merged pull request #49 → 20.03 Markhor art work → https://git.io/Jvdp8
<{^_^}> [nixos-artwork] @samueldr pushed 3 commits to master: https://git.io/JvFXj
waleee-cl has quit [Quit: Connection closed for inactivity]
captjakk has joined #nixos
jlv has quit [Remote host closed the connection]
<pjt_014> Anyone want to give a quick look-over for this?: https://github.com/NixOS/nixpkgs/pull/84078
<{^_^}> #84078 (by petersjt014, 21 hours ago, open): catcli: init at 0.5.13
<pjt_014> 1st package I've ever added :D
<pjt_014> also evils: If you're around, I've made a few changes to it.
<etu> pjt_014: I see 6 commits in the log. I only recognize two of them as needed. The rest should be squashed/rebased on master if there was a conflict :)
captjakk has quit [Ping timeout: 265 seconds]
ArchFeh has quit [Remote host closed the connection]
<lovesegfault> bricewge: Are you around?
fusion809 has joined #nixos
<pjt_014> It's more the actual package file/build process result and such that I'm worried about. I'd prefer not to wrestle with git for another 2 hours if possible.
<{^_^}> [nixpkgs] @lovesegfault opened pull request #84161 → ddcci: 0.3.2 -> 0.3.3 → https://git.io/JvF1n
isHavvyGhosting has quit [Ping timeout: 252 seconds]
sasycto has quit [Remote host closed the connection]
c0c0 has joined #nixos
knupfer has joined #nixos
isHavvyGhosting has joined #nixos
ddellacosta has quit [Ping timeout: 256 seconds]
endformationage has quit [Quit: WeeChat 2.6]
kaliumxyz has quit [Ping timeout: 256 seconds]
vk3wtf has quit [Quit: WeeChat 2.7.1]
vk3wtf has joined #nixos
palo1 has joined #nixos
palo has quit [Ping timeout: 265 seconds]
palo1 is now known as palo
kaliumxyz has joined #nixos
OrangeTetra has joined #nixos
<evils> pjt_014: i'd be happy to talk you through using `git rebase --interactive`
chagra_ has joined #nixos
oranginal has quit [Ping timeout: 265 seconds]
<pjt_014> I suppose
<pjt_014> hm
<pjt_014> well this is a thing
kqb has joined #nixos
chagra has quit [Ping timeout: 258 seconds]
rogue_koder has quit [Ping timeout: 260 seconds]
isHavvyGhosting has quit [Ping timeout: 252 seconds]
drakonis has quit [Quit: WeeChat 2.7.1]
isHavvyGhosting has joined #nixos
lord| has quit [Ping timeout: 256 seconds]
<ndarwincorn> :exit
<ndarwincorn> sorry, got my clients mixed up
ndarwincorn has quit [Quit: WeeChat 2.8]
rogue_koder has joined #nixos
OrangeTetra has quit [Quit: Leaving]
lord| has joined #nixos
kqb has quit [Ping timeout: 250 seconds]
oranginal has joined #nixos
opthomasprime has joined #nixos
knupfer has quit [Quit: knupfer]
knupfer has joined #nixos
knupfer has quit [Client Quit]
eoli3n__ has joined #nixos
knupfer has joined #nixos
lord| has quit [Ping timeout: 240 seconds]
xelxebar_ has joined #nixos
CcxWrk has quit [Quit: ZNC 1.7.4 - https://znc.in]
CcxWrk has joined #nixos
<srhb> lejonet: ceph 14.2.8 is now woring without the horrible patch to cherrypy loading. :-) https://github.com/NixOS/nixpkgs/pull/83749 -- I guess I'll start an Octopus PR soon after that's in.
<{^_^}> #83749 (by srhb, 4 days ago, open): ceph: 14.2.7 -> 14.2.8
<bricewge> lovesegfault: Yes, what's up?
<lovesegfault> bricewge: Oh, hi! I was pinging you because I touched a pkg you maintain: https://github.com/NixOS/nixpkgs/pull/84161
<{^_^}> #84161 (by lovesegfault, 57 minutes ago, open): ddcci: 0.3.2 -> 0.3.3
<lovesegfault> :)
lord| has joined #nixos
cole-h has quit [Ping timeout: 256 seconds]
<bricewge> Thanks for the heads up, i'll review your PR
<lovesegfault> bricewge: cheers!
lord| has quit [Read error: Connection reset by peer]
rogue_koder has quit [Ping timeout: 240 seconds]
plutes has quit [Ping timeout: 260 seconds]
rogue_koder has joined #nixos
knupfer has quit [Quit: knupfer]
knupfer has joined #nixos
knupfer has quit [Client Quit]
knupfer has joined #nixos
knupfer has quit [Client Quit]
rogue_koder has quit [Read error: Connection reset by peer]
knupfer has joined #nixos
<bdju> https://paste.debian.net/hidden/d5b60768/ getting some errors trying to install "stig".
<lejonet> srhb: Sweeet!
vk3wtf has quit [Quit: WeeChat 2.7.1]
knupfer has quit [Client Quit]
knupfer has joined #nixos
lord| has joined #nixos
kaliumxyz has quit [Ping timeout: 250 seconds]
rauno has joined #nixos
ATuin has joined #nixos
FRidh has joined #nixos
mallox has joined #nixos
<bricewge> lovesegfault: I think you can remove the two `--replace` lines you modifed, https://gitlab.com/ddcci-driver-linux/ddcci-driver-linux/-/commits/master/ddcci/Makefile fixed it.
linarcx has joined #nixos
<lovesegfault> bricewge: tried without them and it exploed
<lovesegfault> it tried to perform the build inside of the source contained in the nix store
<{^_^}> [nixpkgs] @jluttine opened pull request #84162 → xonotic: add desktop entries → https://git.io/JvFym
blumenkranz has quit [Remote host closed the connection]
opthomasprime has quit [Remote host closed the connection]
<bricewge> That's odd.
Darkmatter66_ has quit [Ping timeout: 265 seconds]
opthomasprime has joined #nixos
vk3wtf has joined #nixos
<bricewge> The issue with those two line it's not straight forward what is the intent since src should already be set $PWD by default.
<bricewge> lovesegfault: Could you try with `--replace 'src ?=' 'src :='` it should do the same job
<lovesegfault> bricewge: Will do in a second, waiting on a fetch
<{^_^}> [nixpkgs] @lovesegfault opened pull request #84163 → linuxPackages.nvidia_x11: 440.59 -> 440.64 → https://git.io/JvFyl
mexisme has quit [Ping timeout: 252 seconds]
mexisme has joined #nixos
<lovesegfault> bricewge: trying
<srhb> src shouldn't be pwd in unpackPhase, should it?
<srhb> I would expect it to still point to the (immutable) store path, not the build dir.
Darkmatter66 has joined #nixos
<lovesegfault> bricewge: it kaboomed
<srhb> Er, in prePatch.
<lovesegfault> ddcci-driver> make[4]: *** No rule to make target '/build/source/ddcci/ddcci.o', needed by '__build'. Stop.
roosemberth has joined #nixos
<{^_^}> [nixpkgs] @Izorkin opened pull request #84164 → mariadb: backport updates to 20.03 → https://git.io/JvFyu
<bricewge> Thank you lovesegfault
<lovesegfault> :)
zaeph1 has joined #nixos
opthomasprime has left #nixos [#nixos]
captjakk has joined #nixos
<bricewge> I approved it. But I should remove myself from maintainership of Nix package since I have now switch to Guix.
zaeph has quit [Ping timeout: 260 seconds]
<bricewge> Thank you for the PR lovesegfault!
<lovesegfault> bricewge: How do you like Guix?
<lovesegfault> I read some wonky stuff that they were moving to Hurd?
captjakk has quit [Ping timeout: 250 seconds]
<bricewge> Very much! The language is really easy to hack on the development of new features is quick.
oranginal has quit [Quit: Leaving]
eoli3n__ has quit [Ping timeout: 250 seconds]
<bricewge> Yes it will be supporting the Hurd kernel very soon.
<lovesegfault> Ah, so _also_, not _only_
<bricewge> I thought “[...] guix install linux-libre to run Linux in userspace.” gave it away
gulplant1 has joined #nixos
<bricewge> Yes not only :)
eoli3n__ has joined #nixos
<bricewge> I don't see who one could use Hurd as a daily driver
gulplante has quit [Ping timeout: 256 seconds]
mexisme has quit [Ping timeout: 246 seconds]
<srhb> lovesegfault: (wasn't the Hurd only news published on April 1st? :))
ArchFeh has joined #nixos
<lovesegfault> srhb: Maybe? I saw the headline scroll by when I opened my rss reader
<srhb> I think it was an April Fool's day joke.
<xfix> considering it was April 1, 2020, and Hurd isn't really ready for anything, probably
lsix has joined #nixos
thetet has joined #nixos
linarcx has quit [Quit: WeeChat 2.8]
<xfix> also, https://lists.gnu.org/archive/html/guix-devel/2020-03/msg00081.html says they managed to get GNU hello working on Hurd, which is cool, but hardly usable at this point
<{^_^}> [nixpkgs] @feijoas opened pull request #84165 → flat-remix-icon-theme: 20191122 -> 20200116 → https://git.io/JvFSf
kenran has joined #nixos
linarcx has joined #nixos
<xfix> likely supporting Hurd is serious, but not deprecating Linux
<bricewge> Yes, It was a good one!
<bricewge> xfix: Better journalist than Phoronix https://www.phoronix.com/scan.php?page=news_item&px=GNU-Guix-Going-Hurd
ArchFeh has quit [Quit: archfeh]
o1lo01ol1o has joined #nixos
lovesegfault has quit [Ping timeout: 252 seconds]
o1lo01ol1o has quit [Ping timeout: 256 seconds]
<{^_^}> [nixpkgs] @rnhmjoj merged pull request #81223 → papermc: init at 1.15.2r121 → https://git.io/Jvz6k
<{^_^}> [nixpkgs] @rnhmjoj pushed 2 commits to master: https://git.io/JvFSV
gulplant1 is now known as gulplante
<{^_^}> [nixpkgs] @danieldk opened pull request #84166 → softmaker-office: 974 -> 976 → https://git.io/JvFSX
mexisme has joined #nixos
<emery[m]> og
<{^_^}> [nixpkgs] @ehmry opened pull request #84167 → Quaternion → https://git.io/JvFS9
kiloreux_ has joined #nixos
kiloreux has joined #nixos
<kiloreux> I am trying to compile nethack-ftp on nix. But getting `error: use of undeclared identifier 'ARG_MAX` How can I get the value of ARG_MAX though nix?
zeenk has joined #nixos
mawis has joined #nixos
zupo has joined #nixos
mallox has quit [Quit: WeeChat 2.7.1]
mallox has joined #nixos
knupfer has quit [Quit: knupfer]
<{^_^}> [nixpkgs] @saschagrunert opened pull request #84168 → conmon: 2.0.14 -> 2.0.15 → https://git.io/JvFSj
kqb has joined #nixos
knupfer has joined #nixos
FRidh2 has joined #nixos
FRidh has quit [Ping timeout: 265 seconds]
<{^_^}> [nixpkgs] @ehmry closed pull request #84167 → Quaternion → https://git.io/JvFS9
hmpffff_ has joined #nixos
turion has joined #nixos
maddo has joined #nixos
<{^_^}> [nixpkgs] @domenkozar merged pull request #84157 → change howoldis link to status page link → https://git.io/JvF6N
<{^_^}> [nixpkgs] @domenkozar pushed 2 commits to master: https://git.io/JvF9m
hmpffff has quit [Ping timeout: 272 seconds]
<{^_^}> [nix] @edolstra merged pull request #3470 → libexpr: remove unused attrError → https://git.io/JvF3A
<{^_^}> [nix] @edolstra pushed 2 commits to master: https://git.io/JvF9n
smatting has joined #nixos
Athas has quit [Quit: ZNC - http://znc.sourceforge.net]
Athas has joined #nixos
Athas has quit [Client Quit]
<{^_^}> [nixpkgs] @filalex77 opened pull request #84169 → caddy2: 2.0.0-beta10 -> 2.0.0-rc.1 → https://git.io/JvF9u
Athas has joined #nixos
bvdw has quit [Read error: Connection reset by peer]
bvdw has joined #nixos
OJ[m] has joined #nixos
kenran has quit [Quit: leaving]
civodul has joined #nixos
turion has quit [Quit: Leaving.]
thetet has quit [Quit: Leaving.]
opthomasprime has joined #nixos
o1lo01ol1o has joined #nixos
Mateon1 has quit [Remote host closed the connection]
Mateon1 has joined #nixos
o1lo01ol1o has quit [Ping timeout: 264 seconds]
cr4y1 has joined #nixos
linarcx has quit [Quit: WeeChat 2.8]
eoli3n__ has quit [Ping timeout: 256 seconds]
eoli3n__ has joined #nixos
opthomasprime has quit [Quit: Leaving.]
opthomasprime has joined #nixos
mexisme has quit [Ping timeout: 272 seconds]
opthomasprime has left #nixos [#nixos]
zupo has quit [Read error: Connection reset by peer]
mexisme has joined #nixos
andymandias has joined #nixos
zupo has joined #nixos
fendor has joined #nixos
CptCaptain has joined #nixos
m0rphism has joined #nixos
captjakk has joined #nixos
fusion809 has quit [Quit: Leaving]
fusion809 has joined #nixos
<{^_^}> [nixpkgs] @Mic92 merged pull request #77619 → parallel-full: init SQL and CSV support → https://git.io/JvJOn
<{^_^}> [nixpkgs] @Mic92 pushed commit from @tomberek to master « parallel-full: init SQL and CSV support (#77619) »: https://git.io/JvFQL
captjakk has quit [Ping timeout: 250 seconds]
<{^_^}> [nixpkgs] @zowoq opened pull request #84170 → fzf: 0.21.0-1 -> 0.21.1 → https://git.io/JvFQO
mokasin_ has joined #nixos
<mokasin_> Hey, I found an apperantly official hardened profile in https://github.com/NixOS/nixpkgs/blob/master/nixos/modules/profiles/hardened.nix. How do I use it? Is there any documentation on this?
Naptra has joined #nixos
cfricke has joined #nixos
<mokasin_> Specifically, is it safe to enable that or do I risk to get locked out from my remote system?
<{^_^}> [nixpkgs] @ehmry opened pull request #84171 → Gomuks → https://git.io/JvFQW
zupo has quit [Ping timeout: 256 seconds]
zupo has joined #nixos
<Aleksejs> Hi, is it possible to see the result sites file that nix generated from .services.nginx.virtualHost entry?
<{^_^}> [nixpkgs] @filalex77 opened pull request #84172 → kondo: init at 0.3 → https://git.io/JvFQ6
<pjt_014> mokasin_: Importing it as described would probably be correct. If it modifies the kernel, also something like "boot.kernelPackages = pkgs.linuxPackages_hardended" in the config file unless it already does that.
<symphorien> Aleksejs: systemctl cat nginx will give you the path to the nginx.conf file
<mokasin_> Aleksejs, you can get the path to the config looking at systemctl status nginx and have a look there
CMCDragonkai1 has joined #nixos
<mokasin_> pjt_014, well, I guess only one way to find out :)
<Aleksejs> thanks :)
<pjt_014> yup. also if it interferes with your config you may need to use mkOverride if you want your settings to prevail
o1lo01ol1o has joined #nixos
<pjt_014> foo = pkgs.lib.mkOverride 0 bar I think
Darkmatter66_ has joined #nixos
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
zupo has joined #nixos
Darkmatter66 has quit [Ping timeout: 264 seconds]
<mokasin_> pjt_014, I am not sure I understand what you are saying. what does mkOverride 0 do?
mexisme has quit [Ping timeout: 246 seconds]
o1lo01ol1o has quit [Ping timeout: 264 seconds]
<{^_^}> [nixpkgs] @peterhoeg merged pull request #83799 → cantata: make qtmultimedia and vlc optional → https://git.io/Jv7di
<{^_^}> [nixpkgs] @peterhoeg pushed 2 commits to master: https://git.io/JvFQj
<mlatus[m]> <mokasin_ "pjt_014, I am not sure I underst"> incase the profile and your config both define option foo, the value with `mkOverride 0` will be picked
mexisme has joined #nixos
<mlatus[m]> mokasin_: i.e. `mkOverride 0` sets higher priority
<mokasin_> mlatus[m], ah, okay. thx
<{^_^}> [nixpkgs] @markuskowa merged pull request #84100 → fava: 1.13 -> 1.14 → https://git.io/JvFUH
<{^_^}> [nixpkgs] @markuskowa pushed 2 commits to master: https://git.io/JvF7J
zupo has quit [Ping timeout: 260 seconds]
Kyndig has quit [Ping timeout: 252 seconds]
leonardp has quit [Remote host closed the connection]
mawis has quit [Ping timeout: 240 seconds]
thc202 has joined #nixos
<mokasin_> pjt_014, it seems, my system did not explode. *thumbsup*
<pjt_014> wooooooo
sigmundv_ has joined #nixos
<pjt_014> If you did the kernel thing you'll need to reboot
<mokasin_> I feel the same way ;)
<pjt_014> at least I needed to when goung from 4.x to 5.x
<mokasin_> yes, the profile does it already and I've rebooted
<pjt_014> hm. that's the scary part done
emery[m] is now known as ehmry[m]
<pjt_014> welp gotta sleep
<mokasin_> hmm, there seems to be a linuxPackages_latest_hardened.kernel 5.4 kernel
<{^_^}> [nixpkgs] @bkchr opened pull request #84173 → Fix wg-quick after wireguard got upstreamed → https://git.io/JvF7C
<{^_^}> [nixpkgs] @primeos opened pull request #84174 → chromium: 80.0.3987.162 -> 80.0.3987.163 → https://git.io/JvF7W
<{^_^}> [nixpkgs] @filalex77 opened pull request #84175 → emplace: 0.2.14 -> 0.2.15 → https://git.io/JvF70
<{^_^}> [nixpkgs] @primeos pushed to release-20.03 « chromium: 80.0.3987.162 -> 80.0.3987.163 »: https://git.io/JvF7E
__monty__ has joined #nixos
kenran has joined #nixos
ehmry has left #nixos ["https://quassel-irc.org - Chat comfortably. Anywhere."]
Kyndig has joined #nixos
linarcx has joined #nixos
pjt_014 has quit [Ping timeout: 260 seconds]
<mokasin_> noob question: How do I get access to mkOverride in my configuration.nix? It is not in scope
<ikwildrpepper> mokasin_: add lib to the function arguments of the module you are in
<ikwildrpepper> {config, lib, ...}: { ... }
knupfer has quit [Remote host closed the connection]
knupfer has joined #nixos
<ikwildrpepper> then you can use lib.mkOverride
<mokasin_> ikwildrpepper, does the order matter?
<ikwildrpepper> mokasin_: no
<mokasin_> cool
<mokasin_> what arguments are passed to a module?
<kenran> If I have a Haskell `default.nix` and my `shell.nix` returns `(pkgs.haskellPackages.callPackage ./default.nix {}).env`, but I now want more "buildInputs", for example if I want to have `cabal-install` inside my nix-shell. How would I do that?
<mokasin_> config, lib, pkg, ... and?
<kenran> I tried returning: (the-above).env // { buildInputs = [ pkgs.haskellPackages.cabal-install ]; },
<kenran> but that didn't work out. I still don't have a `cabal` executable inside the shell
mallox has quit [Quit: WeeChat 2.7.1]
<{^_^}> [nixpkgs] @primeos merged pull request #84174 → chromium: 80.0.3987.162 -> 80.0.3987.163 → https://git.io/JvF7W
<{^_^}> [nixpkgs] @primeos pushed 2 commits to master: https://git.io/JvF7p
<kenran> ahhh, there might be an .overrideAttrs on the .env
orivej has quit [Ping timeout: 240 seconds]
o1lo01ol1o has joined #nixos
pbogdan has joined #nixos
ratsclub has joined #nixos
thetet has joined #nixos
mawis has joined #nixos
o1lo01ol1o has quit [Remote host closed the connection]
o1lo01ol1o has joined #nixos
mallox has joined #nixos
<{^_^}> [nixpkgs] @Ma27 merged pull request #84170 → fzf: 0.21.0-1 -> 0.21.1 → https://git.io/JvFQO
<mokasin_> Has somebody insight into why the hardened profile does not use linuxPackages_latest_hardened instead of an older kernel?
<{^_^}> [nixpkgs] @Ma27 pushed 2 commits to master: https://git.io/JvF5s
<tilpner> mokasin_: Probably the same reason NixOS doesn't use linuxPackages_latest by default, but linuxPackages, which is the current longterm branch
<mokasin_> tilpner, oh, for some reason I thought it was using a 5.x kernel. my bad. makes sense
<tilpner> mokasin_: Current longterm is 5.4
<mokasin_> tilpner, well, then linuxPackages_hardened is a 4.17 kernel
<tilpner> > linuxPackages_hardened.kernel.name
<{^_^}> "linux-5.4.30"
<mokasin_> wtf
<tilpner> Your nixpkgs is too old
<mokasin_> r u on unstable?
<tilpner> It's also on 20.03
<mokasin_> ah, okay. I'm still using 19.09. :)
hmpffff has joined #nixos
<mokasin_> Now the confusion is complete ;). And lifted.
<tilpner> Which makes sense, considering 20.03 has not been released yet
gxt has quit [Remote host closed the connection]
Darkmatter66_ has quit [Quit: ZNC 1.7.5 - https://znc.in]
hmpffff_ has quit [Ping timeout: 240 seconds]
Jackneill has joined #nixos
thibm has joined #nixos
gxt has joined #nixos
Darkmatter66 has joined #nixos
alci has joined #nixos
chloekek has joined #nixos
laduke has quit [Ping timeout: 272 seconds]
dermetfan has joined #nixos
dermetfan has quit [Client Quit]
<alci> Hi, could anyone tell me where I could get some help regarding a nixos.configuration failed rebuild other than using github issue?
<ratsclub> Which problem did you have?
<ratsclub> I would sugest you to try to get help here and perhaps document it on github
linarcx has quit [Read error: Connection reset by peer]
<alci> the github issue is #83767. Someone closed it saying that it has beens solved while I still can't build it...
<{^_^}> https://github.com/NixOS/nixpkgs/issues/83767 (by alciono, 3 days ago, closed): gucharmap not building during nixos-rebuild
ottidmes has joined #nixos
justanotheruser has quit [Ping timeout: 240 seconds]
<alci> Each time I had a similar isuue I just had to wait untill someone fixed it and rebuilt successfully. Not this time. What should I do next?
<tilpner> Well, the fix is on unstable, which channel are you using?
<alci> unstable
<tilpner> When did you last update?
<alci> the details of the issue is on github #83767
<{^_^}> https://github.com/NixOS/nixpkgs/issues/83767 (by alciono, 3 days ago, closed): gucharmap not building during nixos-rebuild
<tilpner> No, please don't make read that
<alci> I keep trying the rebuild at no avail...
<tilpner> There is no formatting at all, it's unreadable
<tilpner> Did you update the channel?
gulplante has quit [Ping timeout: 240 seconds]
<alci> I did not do anything else than rebuild. I will do un update and see. thanks
<tilpner> Your channel is a few hours too young
<tilpner> *too old
<tilpner> Your channel points to d96bd3394b734487d1c3bfbac0e8f17465e03afe, from Thu Mar 19 15:32:34 2020 +0100
<alci> going to try now...
<tilpner> And the fix was added Thu Mar 19 18:46:09 2020 +0000
<ratsclub> So close, yet so far
rauno has quit [Remote host closed the connection]
fusion809 has quit [Ping timeout: 252 seconds]
musicmatze[m] has joined #nixos
<mokasin_> Is the unstable channel considered stable enough for a home server setup? So a non-critical personal server?
<tilpner> alci: In the future, please put your paste in codeblocks. See https://guides.github.com/features/mastering-markdown/
<alci> I have been on unstable for over 5 years now without any major probs
<tilpner> mokasin_: Many people run their systems from unstable, and it works often
<tilpner> mokasin_: But it may require more intervention than if you were running the latest stable version
<bqv[m]> drakonis hey, seeing this atm (ignore the substitutor stuff), do you think guile 3 would fix this? https://ipfs.io/ipfs/QmQfbpbNYCYuPZQCXwJ7TUk8LhY6tnhjqdTvyD6eaqJfhX
<tilpner> mokasin_: Why do you want to use unstable?
laduke has joined #nixos
<ottidmes> mokasin_: you could also always opt for a middle ground, using stable as a basis but cherry pick packages to be used from unstable, where its important to use the very latest, I do that for all my systems
<alci> unstable offers more packages
<benny> I run unstable for 2 years and it's mostly fine, big updates like with any rolling distribution are an issue, especially things like kde5
<benny> ottidmes: that seems wise
linarcx has joined #nixos
c0c0 has quit [Ping timeout: 256 seconds]
* tilpner was going to suggest the exact same as ottidmes, depending on answer
<ratsclub> As you posted an IPFS link, lemme ask you guys something? How can I upgrade my ipfs version? I'm at 0.4.21 and I feel like I'm not connecting to as many nodes as I should
<chloekek> Is it possible to stop core.* files from appearing in my home directory when programs crash? systemd.coredump.enable = false; did not do the trick.
<tilpner> chloekek: Re-read the description of systemd.coredump.enable
<chloekek> Thanks. Is it possible to have them just never be generated at all?
<chloekek> So neither appear anywhere nor be processed by systemd.
<tilpner> What does your ulimit -c say?
<chloekek> unlimited
<tilpner> 0 would probably prevent it
<tilpner> But I haven't tested it
<tilpner> You can set that value globally from security.pam.loginLimits
<ottidmes> bqv[m]: do you know if there is more information about the outputs of flakes? And how do you deal with creating pkgs beforehand, while your NixOS config might set nixpkgs.config that would be needed to define pkgs?
<{^_^}> [nixpkgs] @zowoq opened pull request #84177 → maintainers/teams: add podman team → https://git.io/JvFdz
<bqv[m]> drakonis Oh boy I think its building
linarcx has quit [Quit: WeeChat 2.8]
<mokasin_> tilpner, sometimes there is a feature in a software, I'd like to have earlier. nextcloud is an example
<mokasin_> ottidmes, I a bit vary of that, because I am not familiar enough with nix to assess what a backdraft I have to expect
<bqv[m]> I think flakes can have whatever outputs they like, but don't quote me on that. Packagewise I'm not sure what you mean?
<bqv[m]> ottidmes: have you seen my repo? Might help
<ottidmes> bqv[m]: I thought as much with the outputs, seemed arbitrary
<mokasin_> ottidmes, I've tried it once without issues. But I haven't found out yet, how to use a module form unstable, only how to load a package from another channel
<ottidmes> bqv[m]: yep, that means you missed my messages yesterday, I commented on how I made a git clone of it and saw a potential issue, not sure if you meant to copy over your entire config to the /nix/store, or does it not contain anything private (not sure how git crypt e.g. works)
mexisme has quit [Ping timeout: 272 seconds]
zupo has joined #nixos
<bqv[m]> ottidmes: ah, yes i did mean to do that
<chloekek> tilpner: Thanks, I set these settings: https://glot.io/snippets/fm5p2jqakl Let’s pray they work.
<ottidmes> mokasin_: module from unstable is like this, you just specify the module you want with disabledModules (google it for examples), and then import the unstable module instead: imports = [ "${pkgs.unstable.path}/nixos/modules/..." ];
<bqv[m]> i resolved on the fact that since the entire config is required to build anyway (and is therefore in the store) then using that is fine
<ottidmes> bqv[m]: wait, so even if you have a local git repo, it will copy it over to /nix/store when using flakes?
fusion809 has joined #nixos
<bqv[m]> yes.
<bqv[m]> at least, it will copy everything tracked by git
<tilpner> mokasin_: Ahh, if you require the newer modules, then using unstable is reasonable
<ottidmes> bqv[m]: I have to disagree with that idea though, there is a difference between a git aware copy and a complete snapshot of a directory (the former does not copy gitignore and stuff, while the latter does)
<alci> <tilpner> thanks a lot. I didn't have any issue for a long time and forgot the necessary update step! Now it's building ok. New issue: tor-browser not building... investigating
<tilpner> mokasin_: Mixing modules between channels can be done, but good luck if there's some assumption that you're breaking by using module combinations that have never tested before
<tilpner> *never been tested before
gulplante has joined #nixos
<tilpner> alci: tor-browser-bundle-bin
<tilpner> alci: The other package was removed
linarcx has joined #nixos
<bqv[m]> ottidmes: it's not very secure, agreed, but i also understand why it was done, and i'm happy to use it as an excuse to link my config out
linarcx has quit [Read error: Connection reset by peer]
justanotheruser has joined #nixos
<{^_^}> [nixpkgs] @primeos pushed to master « chromium: Ignore unknown warning options »: https://git.io/JvFdx
<mokasin_> ottidmes, thanks. I have to save that snippet
<ottidmes> bqv[m]: but I am telling you, there is a complete difference in terms of security between what you are doing and what flakes is doing (I asumme, at least).
<mokasin_> tilpner, yes, exactly. I fear, I'll break my system and cannot return, because of some database migration, for example
Darkmatter66 has quit [Quit: ZNC 1.7.5 - https://znc.in]
<mokasin_> tilpner, ottidmes thx for the insights
xkapastel has joined #nixos
roosemberth has quit [Quit: WeeChat 2.8]
<bqv[m]> ottidmes: what? no, as i understand it nix is already importing the config into the store as part of the flake mechanism. e.g. i'm absolutely certain it does that for flake inputs, and i believe it does it for the flake in question too?
Darkmatter66 has joined #nixos
<alci> tilpner, all going well now. Many thanks.
linarcx has joined #nixos
<tilpner> mokasin_: That can happen, and it will be hard to help you if you have a system that's been patched together from many channels
<tilpner> mokasin_: FWIW, I've been using unstable matrix-synapse for a few months, and it's been working out okay-ish
<tilpner> mokasin_: No data loss, but the module began using postgres options that were on unstable only
linarcx has quit [Read error: Connection reset by peer]
<tilpner> mokasin_: So what do you do? Use the unstable postgres module
<ottidmes> bqv[m]: yes, but since the git imports need all files to be tracked, I am quite certain it will not just copy over any files in gitignores, like in my case, my secret files that are within the same repo, and doing ${../.} would just copy over the directory recursively, stupidly, including my secret files, while a git clone (which I assume flakes does for git repos), would not end up having them
<ottidmes> inside the store
linarcx has joined #nixos
jbrock has quit [Quit: jbrock]
<tilpner> mokasin_: My point is: it can work (does for me), but if it breaks you might hear "we told you so"
<tilpner> TheSirC[m]: How are you using overlays/default.nix?
zaeph1 is now known as zaeph
<TheSirC[m]> <tilpner "TheSirC: How are you using overl"> ```
<bqv[m]> huh, i seem to have grown a new gpg key...
<ottidmes> mokasin_: ^ yep, nixpkgs works together well within the same channel, if not, thats a bug, but outside of that you run the risk of breakage, due to the breaking of assumptions
<tilpner> TheSirC[m]: There were no words in your last message
<FRidh2> If you can't use Nix but need a Python interpreter, see https://github.com/FRidh/static-python for statically-linked builds. One major bug is left though
<TheSirC[m]> tilpner: It was a code snippet, I just import it
<tilpner> TheSirC[m]: No, there is no "just import it". Your method of importing may be the problem here
_rvl has quit [Quit: ZNC 1.7.4 - https://znc.in]
<ottidmes> TheSirC[m]: you probably do not have the format of (self: super: { ...stuff... }), maybe you have just { ...stuff... }, and that's not a valid overlay, even if you don't use the arguments
<TheSirC[m]> <tilpner "TheSirC: No, there is no "just i"> `imports = [ ../../overlays/default.nix];`
<simpson> FRidh2: Fun; reminds me of https://github.com/squeaky-pl/portable-pypy
<tilpner> TheSirC[m]: Okay, you treat it like a module, that's good
<TheSirC[m]> I know. That's why I included it in the first code snippet
alci has quit [Quit: Leaving]
<tilpner> I don't recall if value modules work, or if they need to be functions
<FRidh2> simpson: if we get pypy to work again then that can be the next...i don't have the patience for those builds though :p
<TheSirC[m]> ottidmes: Hum so you thin that it is one of the overlay that is miswritten
kenran has quit [Ping timeout: 256 seconds]
romildo has joined #nixos
<TheSirC[m]> * ottidmes: Hum so you think that it is one of the overlay that is miswritten
<TheSirC[m]> * ottidmes: Hum so you think that it is one of the overlay that is miswritten
<ottidmes> TheSirC[m]: lol, I see the message three times, but yeah, it could be, since that is what is being reported
<FRidh2> simpson: the note under the title implies the official builds are portable already nowadays?
_rvl has joined #nixos
<tilpner> TheSirC[m]: I think overlays/default.nix might end up in your overlays
<tilpner> TheSirC[m]: But it's a module, so fails the overlay check
<tilpner> TheSirC[m]: Check from nix repl, or filter it out explicitly
<TheSirC[m]> tilpner: Wouldn't that trigger infinite recursion
<romildo> Should backports from master to NixOS 20.03 go into which branch? release-20.03 or nixos-20.03?
<tilpner> TheSirC[m]: No, I don't think so
<TheSirC[m]> <tilpner "TheSirC: Check from nix repl, or"> I tried the replk without luck
<tilpner> romildo: The one with the newer commit
<simpson> FRidh2: The official builds include portable-pypy-style builds, yes. They do not use Nix or other systematic tools, but have an ad-hoc script.
<tilpner> romildo: Wrong, sorry. I think it's release-20.03
<nh2> romildo: release-20.03. nixos-20.03 is what gets auto-advanced by CI when release-20.03 builds successfully
<ottidmes> TheSirC[m]: no, just the error you got, cause it fails the moment it encounters it again, and its not a recursive module imports either
<TheSirC[m]> <ottidmes "TheSirC: lol, I see the message "> ottidmes: So I am interested in you view because oddly, the numbe of error message correspond to the number of files directly in the folder 🤔
c0c0 has joined #nixos
<TheSirC[m]> ottidmes: I will fix the `default.nix` to try to avoid loading itself
kenran has joined #nixos
<ottidmes> TheSirC[m]: That seems to be the problem indeed, as mentioned by tilpner, if that does not work for you, I would definitely try and debug it by first creating a list of files (not imported yet) and see if you agree with the files being generated
<TheSirC[m]> Would there be any reason why my argument `(map ..)` would not work in the repl ?
<ottidmes> TheSirC[m]: could be you are using some lib/builtins that are not available, try this: nix repl '<nixpkgs/lib>'
<bqv[m]> ottidmes: oh, christ i get you now
<bqv[m]> i wonder if lib.cleanSource would fix that
<bqv[m]> ottidmes: hang on, no
<TheSirC[m]> ottidmes: Could you see my code snippet in my first message ? I didn't realize you guys were on Freenode
<bqv[m]> even that is demonstrably not true
<bqv[m]> like i said, nix copies the flake's git'structure first
<bqv[m]> so e.g. there are files in my directory here that aren't copied into the store
<bqv[m]> and aren't even copied by that ${../.}
<bqv[m]> because they aren't part of the flake
<TheSirC[m]> ottidmes: Here it is in a pastebin : https://paste.rs/jtF
<bqv[m]> this is full restricted mode
<tilpner> ,matrixbridge TheSirC[m]
<{^_^}> TheSirC[m]: The IRC<->Matrix bridge sends multi-line messages as links: `* Someone sent a long message: < https://matrix.org/_matrix/media/v1/download/matrix.org/... >`. This means that highlights don't work, people have to open your link, and it gets less visibility than multiple single-line messages
<ottidmes> TheSirC[m]: not sure how it works, rather an IRC beginner myself, but I do see matrix pastes being reported
<bqv[m]> ahaha oh wow really?
<bqv[m]> i didn't know the bridge did that
<bqv[m]> that's somewhat awful
<TheSirC[m]> <bqv[m] "that's somewhat awful"> Indeed...
<ottidmes> bqv[m]: what are you talking about flakes for, ${../.} is flake unrelated, right? or is it made aware of what files flakes have access to and only copies over those?
<TheSirC[m]> ottidmes: IRC n00b too but I am on the Matrix side, you can find the snippet in the link above if you could not see it the first time
<bqv[m]> ottidmes: so, in the context of flake evaluation, the current working directory is NOT the directory you evaluate nix from
<bqv[m]> it's a git skeleton copied into the store
<ottidmes> TheSirC[m]: I could see it just fine, like tilpner mentioned, I saw that "someone sent a long message" message
<ratsclub> is there a matrix channel for nixos?
<bqv[m]> hence, it doesn't matter if you re-copy it into the store
<{^_^}> [nix] @edolstra pushed to flakes « getFlake: In pure mode, check that the argument is an immutable flakeref »: https://git.io/JvFbe
<TheSirC[m]> ottidmes: Oh so it renders it as a "spoiler" of some sort ?
<ottidmes> bqv[m]: Ah, like so! Well, at least it is good to be sure about it :) it would be rather bad if you go through all the trouble to properly setup secret files only to find them world readable due to a nicety
<bqv[m]> lol
<TheSirC[m]> ratsclub: Yes ! I think I am in it and you're too but through a bridge
<tilpner> There is another unbridged channel
<tilpner> #nix:matrix.org
<romildo> Trying to checkout release-20.03 gives the error: error: pathspec 'release-20.03' did not match any file(s) known to git. What else should I do checkout this branch in order to cherry pick a commit from master?
mawis has quit [Ping timeout: 256 seconds]
<bqv[m]> tilpner: it's absolutely dead in comparison, though
<bqv[m]> because all nix nerds seem to prefer irc
<tilpner> romildo: git fetch origin release-20.03
<tilpner> (Or just blanket git fetch)
vykook has quit [Ping timeout: 265 seconds]
<mokasin_> tilpner, using unstable on one package is like a virus spreading. In the end, using unstable altogether sounds more reasonable? At least, it was tested together to some extent.
<tilpner> mokasin_: I don't agree with that first sentence
<romildo> tilpner, that gives: fatal: couldn't find remote ref release-20.03
<tilpner> romildo: Oh, okay. Maybe just git fetch
<ottidmes> bqv[m]: what do I gain by going to the matrix? I heard people complain about it being slower (but maybe that is due to the forwarding here)
<tilpner> ottidmes: There's little point in using Matrix as an IRC client right now, if you already have a bouncer set up via other means
<{^_^}> [nixpkgs] @gnxlxnxx opened pull request #84178 → swaylock-effects: init at v1.6-0 → https://git.io/JvFbt
<mokasin_> tilpner, well, if upgrading matrix-synapse requires an updated postgresql, which in turn breaks another package... but maybe this is not so relevant in praxis for a small installation, i dont know
<bqv[m]> ottidmes: what tilpner said, but you can use both
<tilpner> mokasin_: That's modules, and in that context I agree. You mentioned packages though, and in that context I disagree
<ottidmes> tilpner: its on my TODO somewhere :P I just try to keep my weechat open as long as possible :P
<romildo> tilpner, that does not work either: fatal: 'release-20.03' does not appear to be a git repository
<bqv[m]> and matrix as a protocol is modern and nice, and not just in the newfangled "slack has emoji!!" sense
<mokasin_> ottidmes, matrix is actually quite fast if not used on matrix.org, which is just overloaded. But it is indeed a WIP
<tilpner> romildo: What was the full command you ran?
<romildo> tilpner, git fetch release-20.03
<bqv[m]> ottidmes: fyi i use weechat-matrix
<bqv[m]> i'm actually waiting for nixos-unstable-small to advance past my weechat-matrix update rn
<tilpner> romildo: Just git fetch
<bqv[m]> it's not advanced for 23 hours...
<mokasin_> tilpner, gotcha. I'm still need to get used to nix terminology and mix it often up.
<simpson> bqv[m]: Just build on your own, if you're impatient. There's no need to wait for the official channel, if you know that your desired packages will build.
vykook has joined #nixos
hmpffff has quit [Remote host closed the connection]
noudle has joined #nixos
hmpffff has joined #nixos
<bqv[m]> simpson: i can, but perks of a declarative config, i'd quite like to not have to temporarily hack my configuration just to use it
<bqv[m]> it's not quite as simple as "just build it locally"
<romildo> tilpner, still does not work.
<bqv[m]> i have to change my config, build that, run that, and then rebase it every time i make any other change so i can easily revert it later
<TheSirC[m]> <ottidmes "qy: what do I gain by going to t"> ottidmes: Not slower in my experience I think that you win by having a single point that can talk to multiple protocol
<TheSirC[m]> At least that what hooked me 🤷‍♂️
<bqv[m]> TheSirC: the irc bridge is slow, in my experience, but that's just because freenode is ridiculous. and either way you can still use matrix AND irc
<romildo> tilpner, git branch -a | grep release-20.03 does not list anything.
ratsclub has quit [Remote host closed the connection]
<TheSirC[m]> <bqv[m] "TheSirC: the irc bridge is slow,"> I must admit that I never did experiments to check
<bqv[m]> to be honest, i might actually just cut my losses and point my flake at nixpkgs/master
<mokasin_> Is nixos.org/channels down? I get a 404
<bqv[m]> no more waiting, ever
<tilpner> romildo: What does git remote get-url origin say?
<{^_^}> [nixpkgs] @kolloch opened pull request #84179 → buildRustCrate: Allow version specific dependency renaming + parameter doc → https://git.io/JvFbz
<tilpner> romildo: You may have to specify the proper remote, git fetch origin or your equivalent
<romildo> tilcreator, git@github.com:romildo/nixpkgs.git
<bqv[m]> TheSirC: it will become patently obvious if you ever try and have a rapid conversation, or even just try and use an irc bot from here
<tilpner> romildo: That's the wrong one then. How did you call nixos/nixpkgs?
<mokasin_> never mind, typo
<simpson> bqv[m]: Lucky 10000: Something like $(NIX_PATH=nixpkgs=... nixos-rebuild) will work to temporarily use a local checkout. You can use this to cheaply switch between configured channels and a local nixpkgs checkout.
vykook has quit [Ping timeout: 264 seconds]
<bqv[m]> simpson: won't work for flakes.
<mokasin_> bqv[m], typical error message ;)
<tilpner> bqv[m]: Works here
<romildo> tilpner, I did no understand your question 'How did you call nixos/nixpkgs?'
<simpson> bqv[m]: To the extent that it won't, that is WAI, I gather. But sure, overlays and flakes complicate the process somewhat. I'm just saying that one does not need to be beholden to the channels.
<tilpner> romildo: You must have added upstream nixpkgs (github.com/nixos/nixpkgs) as a remote at some point
<tilpner> romildo: How did you call that remote?
linarcx has quit [Quit: WeeChat 2.8]
<tilpner> romildo: Pass that name to git fetch
lsix1 has joined #nixos
<{^_^}> [nixpkgs] @rnhmjoj merged pull request #84173 → Fix wg-quick after wireguard got upstreamed → https://git.io/JvF7C
<{^_^}> [nixpkgs] @rnhmjoj pushed 2 commits to master: https://git.io/JvFbw
<tilpner> That should also explain why all the previous commands didn't work, your origin was your own fork
<bqv[m]> simpson: as far as the flake mechanism allows, i actually *do* have to be beholden to channels unless i want to modify my config to not use them, i.e. switch to nixpkgs/master
<bqv[m]> but to be honest i think i will just do that
vykook has joined #nixos
<romildo> tilpner, git remote -v gives:
<romildo> upstreamhttps://github.com/NixOS/nixpkgs.git (push)
<romildo> upstreamhttps://github.com/NixOS/nixpkgs.git (fetch)
<romildo> origingit@github.com:romildo/nixpkgs.git (push)
<romildo> origingit@github.com:romildo/nixpkgs.git (fetch)
<tilpner> git fetch upstream
<TheSirC[m]> OK, let's say I have a list with the path of nix files `theList = [ /A/a.nix /B/b.nix ./c.nix ]` why wouldn't `(map (import) theList)` work ? for reference it sends back `cannot coerce a list to a string`
<TheSirC[m]> It is humiliating but at this point it defies my understanding
<TheSirC[m]> So let's just improve my understanding
<tilpner> That should work
boegel has quit [Ping timeout: 240 seconds]
<tilpner> Are you just executing that, or is there more code you're not showing?
melling has joined #nixos
<tilpner> FYI All your parentheses are unnecessary
<TheSirC[m]> <tilpner "FYI All your parentheses are unn"> Thatnks I did not know that
<TheSirC[m]> <tilpner "Are you just executing that, or "> I am executing that in the repl
<TheSirC[m]> I can snippet the whole code
<tilpner> TheSirC[m]: I would prefer if you created a paste of your repl history, so I can see what you tried and the errors
<ottidmes> TheSirC[m]: mapping import over a list of files should work just fine, it might be due to the file contents though
<TheSirC[m]> tilpner: Your wishes are my commands
<tilpner> TheSirC[m]: Use map builtins.pathExists theList to check if all paths are valid and exist
<romildo> tilpner, thanks
<TheSirC[m]> <tilpner "TheSirC: Use map builtins.pathEx"> All true
<TheSirC[m]> Here is a path of the file I am loading : https://paste.rs/kbp
vykook has quit [Ping timeout: 256 seconds]
<tilpner> TheSirC[m]: Then the mistake is inside one of those files
<tilpner> TheSirC[m]: Try importing each in turn, to figure out which it is
<{^_^}> [nixpkgs] @marsam merged pull request #84159 → skim: 0.8.0 -> 0.8.1 → https://git.io/JvFit
<{^_^}> [nixpkgs] @marsam pushed 2 commits to master: https://git.io/JvFNd
kleisli_ has joined #nixos
<{^_^}> [nixpkgs] @doronbehar opened pull request #84180 → qwt6: 6.1.2 -> 6.1.4 → https://git.io/JvFAJ
<TheSirC[m]> tilpner: And here is the log of my repl (sorry for the dealy, I had some issue yanking from tmux ) : https://paste.rs/Ixc
vykook has joined #nixos
<{^_^}> [nixpkgs] @marsam merged pull request #84168 → conmon: 2.0.14 -> 2.0.15 → https://git.io/JvFSj
andymandias has quit [Quit: ZNC 1.7.4 - https://znc.in]
<{^_^}> [nixpkgs] @marsam pushed 2 commits to master: https://git.io/JvFAk
<ottidmes> bqv[m]: BTW what I meant with pkgs created before config in your flake, is that what if you have say nixpkgs.config.oraclejdk.accept_license = true; be dependent on NixOS config, then it is not known beforehand what is nixpkgs.config, but I guess you would just have to create pkgs twice, one specific for the config, and one specific for the package outputs
kleisli has quit [Ping timeout: 256 seconds]
<TheSirC[m]> tilpner: They all import well and good with the repl
linarcx has joined #nixos
andymandias has joined #nixos
<tilpner> TheSirC[m]: You probably meant: fnA = map import (validFiles ./.);
linarcx has quit [Read error: Connection reset by peer]
<tilpner> TheSirC[m]: Right now, you're creating a list that contains the map function
linarcx has joined #nixos
<bqv[m]> ottidmes: ah, yes, i'm not sure on that, that's a situation i haven't considered before. i haven't currently got any packages complex enough to need to worry about that
<TheSirC[m]> <tilpner "TheSirC: Right now, you're creat"> Isn't taht what `nixpkgs.overlays` expects ?
<tilpner> TheSirC[m]: When I said your parentheses are unnecessary, that didn't apply to putting the result in a list. Then you need some of them
<simpson> Yes. Parens don't matter for function application, but they *do* matter for list elements.
<tilpner> TheSirC[m]: Then it would be: fnA = [ (map import (validFiles ./.)) ];
<{^_^}> [nixpkgs] @edolstra pushed to release-20.03 « nix-daemon.nix: Add option nix.registry »: https://git.io/JvFAR
<TheSirC[m]> simpson: :mind_blown:
<TheSirC[m]> tilpner: I am trying that right away
<TheSirC[m]> tilpner: We agree that you mean `nixpkgs.overlays = (map import (validFiles ./.))];` right ?
<{^_^}> [nixpkgs] @schmittlauch opened pull request #84181 → [backport] haskell shellFor: Fix hoogle → https://git.io/JvFAd
<tilpner> TheSirC[m]: That depends on information I don't have, so I need to guess
melling has quit [Quit: WeeChat 2.7.1]
<tilpner> TheSirC[m]: Assuming those files are overlays, yes, that's correct
<tilpner> TheSirC[m]: If you really want a nested list, see the second example
<tilpner> (But you probably don't)
melling has joined #nixos
<tilpner> TheSirC[m]: Oh come on, you pasted a mix of both my examples
<tilpner> TheSirC[m]: You want: nixpkgs.overlays = map import (validFiles ./.);
<tilpner> No ]
boegel has joined #nixos
<TheSirC[m]> tilpner: I just noticed that..sorry...
mawis has joined #nixos
<TheSirC[m]> And of course...it works...
<tilpner> TheSirC[m]: map creates a list, and the value of nixpkgs.overlays should be a list
<tilpner> TheSirC[m]: If you do [ (map foo bar) ], then you end up with [ [ <something> <somethingelse> ] ]
<TheSirC[m]> Thank you a lot for your help. tilpner: simpson ottidmes
<TheSirC[m]> Coming from a stranger on the internet that might seem like nothing but thanking for helping others...
<tilpner> Which is a list, but not a list of overlays
<tilpner> It would be a list of lists of overlays
<TheSirC[m]> Which makes total sense from a type stand point
<{^_^}> [nixpkgs] @matthiasbeyer opened pull request #84182 → Update boost → https://git.io/JvFAA
<TheSirC[m]> I do not know why but to me `map` consumes the list when it actullay just apply the function to the lists elements
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<Unode> Hi all, does anyone have an example of how to add custom lines to the system's xorg.conf ?
<simpson> TheSirC[m]: No worries. Everybody is a neophyte at some point; everybody has to learn and to ask questions. Also, that list/paren precedence is *very* quirky, and it trips up even experienced folks.
<{^_^}> [nixpkgs] @doronbehar opened pull request #84183 → Update uhd → https://git.io/JvFAx
<Unode> Trying to add trackpoint settings as part of evdev but can't figure out which options to use
<{^_^}> [nixpkgs] @Mic92 merged pull request #84119 → todoist-electron: 1.19 -> 1.20 → https://git.io/JvFZV
<{^_^}> [nixpkgs] @Mic92 pushed 2 commits to master: https://git.io/JvFAp
<bqv[m]> i do kinda wish nix had haskell's `$`
<simpson> Unode: services.xserver.extraConfig I think.
<TheSirC[m]> simpson: You're right but not everyone takes its time to help neophytes
<bqv[m]> the precedence rules are inconvenient as it stands, coming from haskell
<Unode> simpson: thanks
<TheSirC[m]> So I prefer to thank the ones who do
pbogdan has quit [Remote host closed the connection]
zupo has joined #nixos
lsix1 has quit [Ping timeout: 272 seconds]
<ottidmes> bqv[m]: question about your private files, if paths are relative to the copied repo to /nix/store, won't my current way of directly using say fileContents ./some.pass no longer work? How do you deal with that
zupo has quit [Client Quit]
<bqv[m]> ottidmes: see for reference ::/users/root.nix which imports from my secrets directory
<bqv[m]> you can import a file, but it has to be tracked by git, and if it is, you can just navigate relative to the current nix file
fling has quit [Ping timeout: 264 seconds]
<bqv[m]> unless i'm misunderstanding how fileContents work, is it just the same as builtins.readFile?
<bqv[m]> (or import?)
boegel has quit [Ping timeout: 240 seconds]
_e has joined #nixos
<{^_^}> [nixpkgs] @romildo opened pull request #84184 → [20.03] freeoffice: 973 -> 976 → https://git.io/JvFxR
pbogdan has joined #nixos
bebarker has joined #nixos
<{^_^}> [nix] @edolstra pushed to flakes « Make --override-input sticky »: https://git.io/JvFxu
rauno has joined #nixos
<bqv[m]> dammit
<ottidmes> bqv[m]: its just builtins.readFile with a removeSuffix "\n"
bbarker has quit [Ping timeout: 260 seconds]
<bqv[m]> https://github.com/edolstra/dwarffs/commit/5efa1289915b2e69771b0a2e237a7e1d25509b8a#diff-d4595c85b426c374ef7509eff440229c this commit makes my flake not build because the nixFlakes in nixpkgs doesn't support a version 5 flake.lock
boegel has joined #nixos
<{^_^}> [nixpkgs] @cpcloud opened pull request #84185 → scmpuff: init at 0.3.0 → https://git.io/JvFxa
<bqv[m]> ottidmes: ah, then you're fine. it counts as a store path in scope if it's tracked by git, so as long as you use relative paths it will work as before
<bqv[m]> if you want to avoid it being in git unencrypted then nrxdp's method of scoping it all to a secrets directory and using git-crypt is probably easiest
bbarker has joined #nixos
linarcx has quit [Quit: WeeChat 2.8]
bebarker has quit [Ping timeout: 264 seconds]
NieDzejkob has joined #nixos
<NieDzejkob> Huh, I just noticed that the homepage URL for patchelf (https://nixos.org/patchelf.html) 404's. Where do I report this?
<{^_^}> [nixpkgs] @cdepillabout merged pull request #84181 → [backport] haskell shellFor: Fix hoogle → https://git.io/JvFAd
<{^_^}> [nixpkgs] @cdepillabout pushed 2 commits to release-20.03: https://git.io/JvFxj
<gchristensen> github.com/nixos/nixos-homepage please NieDzejkob :)
linarcx has joined #nixos
zupo has joined #nixos
thetet has quit [Quit: Leaving.]
Ariakenom has joined #nixos
thetet has joined #nixos
thetet has quit [Client Quit]
boegel has quit [Ping timeout: 264 seconds]
romildo has quit [Quit: Leaving]
waleee-cl has joined #nixos
teto has joined #nixos
<{^_^}> [nixpkgs] @romildo merged pull request #84165 → flat-remix-icon-theme: 20191122 -> 20200116 → https://git.io/JvFSf
<{^_^}> [nixpkgs] @romildo pushed 2 commits to master: https://git.io/JvFpg
<NieDzejkob> gchristensen: huh, looks like it's intentional: https://github.com/NixOS/nixos-homepage/commit/179733c176c1f625bb18d4eba38480b888471a7a
<ottidmes> bqv[m]: but of course its not tracked by git, why would I put pass files in my git repo, so my current approach will break, guess I will have to look at workarounds (I am looking into flakes more than actually planning to convert already, but at least now I already know and documented some things to look out for)
eoli3n__ has quit [Ping timeout: 240 seconds]
oida has quit [Remote host closed the connection]
<bqv[m]> ah okay, yeah
<{^_^}> [nixpkgs] @matthiasbeyer opened pull request #84186 → pulseeffects: 4.7.1 -> 4.7.2 → https://git.io/JvFpM
<bqv[m]> if i had realised flakes put the source configuration in the store, i might have developed a better solution (assuming one exists) before switching fully
<bqv[m]> but it's too late now
<exarkun> is there a nice pattern for constructing a hash key for CI for the dependencies of a derivation? I want a key that changes when any of the buildInput-type inputs to a derivation change but not when, eg, the src of the derivation changes
<exarkun> all I can think of is writing a function to walk all the build input attributes and extract their output path and then hash all of that
<{^_^}> [nixpkgs] @marsam merged pull request #81622 → gotop: 3.0.0 -> 3.4.5 → https://git.io/JvaOc
<{^_^}> [nixpkgs] @marsam pushed 2 commits to master: https://git.io/JvFpQ
fendor has quit [Read error: Connection reset by peer]
ambroisie has joined #nixos
domogled has joined #nixos
<{^_^}> [nixpkgs] @edolstra pushed to master « Don't pin 'nixpkgs' in the system registry by default »: https://git.io/JvFhJ
<{^_^}> [nixpkgs] @edolstra pushed to release-20.03 « Don't pin 'nixpkgs' in the system registry by default »: https://git.io/JvFhU
linarcx has quit [Read error: Connection reset by peer]
linarcx has joined #nixos
plp_ has quit [Quit: No Ping reply in 180 seconds.]
<exarkun> That will miss out on some dependencies though, those just referenced in build phases and not included in the explicit input lists
plp_ has joined #nixos
<exarkun> How do I get those?
<ottidmes> exarkun: you probably want to use nix-store to gather the info
captjakk has joined #nixos
<gchristensen> NieDzejkob: can you open an issue?
<exarkun> is there a nix-store command that will work on the expression and not the store path? I have to compute the cache key before the build for it to be useful.
<simpson> If only the intensional store RFC were ready.
<bqv[m]> > stdenv.lib.licenses.gpl3Plus
<{^_^}> { fullName = "GNU General Public License v3.0 or later"; shortName = "gpl3Plus"; spdxId = "GPL-3.0-or-later"; url = <CODE>; }
mokasin_ has quit [Quit: Leaving]
ZaraChimera has quit [Quit: Going offline, see ya! (www.adiirc.com)]
<bqv[m]> cole-h: ty for review, do you think i should use fetchgit over fetchurl? i don't see the benefit of using fetchgit, but happy to switch
<bqv[m]> no other instances of notabug in nixpkgs apparently, on a quick ripgrep
captjakk has quit [Ping timeout: 260 seconds]
xkapastel has quit [Quit: Connection closed for inactivity]
linarcx has quit [Quit: WeeChat 2.8]
<kenran> How can I find out which nix package contains the command line tool `which`? For some reason it's not in my pure nix-shell and I want it back!
<peelz> kenran: it's just "which" lol
<bqv[m]> or anyone else who wants to comment, since cole is probably asleep. https://github.com/NixOS/nixpkgs/pull/84004#pullrequestreview-386837689 for reference
<bqv[m]> huh, thought it was findutils
<peelz> it might be in findutils but it definitely has its own package too
<peelz> > which
<{^_^}> "<derivation /nix/store/jdrkz8i5dgvvw1jxj50mrrc5x34ykdai-which-2.21.drv>"
<exarkun> there is a nix tool that will tell you where to find a binary
<exarkun> but I forget its name right now
<peelz> nix-locate?
<peelz> * nix-index
<kenran> peelz: wow lol, thanks! I actually did `nix search which` before but must have overlooked it -_-
<exarkun> ah yea nix-index, that's the one
<bqv[m]> i just use ,locate here if it's not installed, readlink -f if it is
jgarte[m] has joined #nixos
<bqv[m]> anyone got a service for nix-index? i might add it to my syste
mokasin has joined #nixos
matthiaskrgr has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net - currently broken?]
nikivi has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
<peelz> bqv[m]: for automatic re-indexing or what?
<peelz> I just have it added to my packages
<bqv[m]> peelz: yeah
<typetetris> Can someone give me a few pointers how to package a qt app? I tried following the manual, but had no success. The package in question x2godesktopsharing doesn't have a makefile and one needs to generate one with `qmake` but the result contains paths to `lrelease` which are simply invalid on my machine. (They point to bin in qmake store path, but lrelease resides in qttools store path).
<bqv[m]> similar to services.locat
<peelz> bqv[m]: hmm, let me know if you find/make one
<mokasin> I've upgrade to unstable from 19.09 and I am running into permission issues with ACME. I get errors like [acme-start] Error while loading the certificate for domain example.com
<mokasin> open certificates/example.com.crt: permission
<mokasin> Interestingly /var/lib/acme/.lego and subfolders have permission for specific users
<mokasin> So basically two users, user1 and user2 use acme, and .lego belongs to user1 and is quite restrictive. It looks like a bug or an upgrade issue from simp-le
<{^_^}> [nixpkgs] @romildo merged pull request #84166 → softmaker-office: 974 -> 976 → https://git.io/JvFSX
<{^_^}> [nixpkgs] @romildo pushed 3 commits to master: https://git.io/JvFhD
gustavderdrache has joined #nixos
gxt has quit [Ping timeout: 240 seconds]
gxt has joined #nixos
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
FRidh2 has quit [Ping timeout: 240 seconds]
FRidh2 has joined #nixos
fusion809 has quit [Read error: Connection reset by peer]
fusion809 has joined #nixos
<bqv[m]> i'll probably build a module out of those two
<peelz> oh nice
<LnL> doesn't that exist in nixos also?
cruxeternus has joined #nixos
linarcx has joined #nixos
ambroisie has quit [Ping timeout: 256 seconds]
zupo has joined #nixos
Jackneill has quit [Ping timeout: 250 seconds]
<{^_^}> [nixpkgs] @Ma27 merged pull request #84178 → swaylock-effects: init at v1.6-0 → https://git.io/JvFbt
<{^_^}> [nixpkgs] @Ma27 pushed 3 commits to master: https://git.io/JvFjv
ambroisie has joined #nixos
oida has joined #nixos
<bqv[m]> heyo, how does mkAlias work
<bqv[m]> LnL: if it does, i didn't find it
thetet has joined #nixos
hmpffff has quit [Quit: nchrrrr…]
iyzsong has quit [Quit: ZNC 1.7.1 - https://znc.in]
hmpffff has joined #nixos
opthomasprime has joined #nixos
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<bqv[m]> oh, mkAlias doesn't even a thing
mawis has quit [Ping timeout: 252 seconds]
virus_dave has joined #nixos
jbrock has joined #nixos
linarcx has quit [Quit: WeeChat 2.8]
KeiraT has quit [Ping timeout: 240 seconds]
linarcx has joined #nixos
opthomasprime has quit [Remote host closed the connection]
matthiaskrgr has joined #nixos
nikivi has joined #nixos
Jackneill has joined #nixos
<{^_^}> [nixpkgs] @bjornfor merged pull request #84180 → qwt6: 6.1.2 -> 6.1.4 → https://git.io/JvFAJ
<{^_^}> [nixpkgs] @bjornfor pushed commit from @doronbehar to master « qwt6: 6.1.2 -> 6.1.4 »: https://git.io/JvFj0
zupo has joined #nixos
<{^_^}> [nixpkgs] @Infinisil merged pull request #84074 → nixos/treewide: Fix incorrectly rendered examples → https://git.io/Jvdbp
<{^_^}> [nixpkgs] @Infinisil pushed 2 commits to master: https://git.io/JvFjX
cosimone has joined #nixos
<{^_^}> [nixpkgs] @romildo opened pull request #84187 → matcha: 2020-03-29 -> 2020-04-03 → https://git.io/JvFjM
<ottidmes> anyone experience with docker on NixOS, any idea why my docker service starts fine, but when its stopped, its stuck, I have to kill it or in the case of poweroff, have it be terminated after a timeout (the reason I mostly want it fixed, dont want to wait 2 minutes for a reboot)
<ottidmes> when I run the docker command to stop it myself, it works fine
gxt has quit [Ping timeout: 240 seconds]
<bqv[m]> isn't there a docker2nix thing somewhere
<ottidmes> there is an option to generate services, which is what I am using
<bqv[m]> oh right
<bqv[m]> maintained by LnL, actually
<bqv[m]> interesting
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ambroisie has quit [Ping timeout: 264 seconds]
bvdw has quit [Read error: Connection reset by peer]
bvdw has joined #nixos
mallox has quit [Quit: WeeChat 2.7.1]
mallox has joined #nixos
Ariakenom has quit [Read error: Connection reset by peer]
Ariakenom_ has joined #nixos
<{^_^}> [nixpkgs] @flokli merged pull request #84164 → [20.03] mariadb: backport updates → https://git.io/JvFyu
<{^_^}> [nixpkgs] @matthiasbeyer opened pull request #84189 → Add peertube → https://git.io/JvbeJ
<{^_^}> [nixpkgs] @flokli pushed 8 commits to release-20.03: https://git.io/JvbeU
emilsp has quit [Ping timeout: 619 seconds]
gxt has joined #nixos
ambroisie has joined #nixos
KeiraT has joined #nixos
<{^_^}> [nixpkgs] @geistesk opened pull request #84190 → platformio: 4.1.0 -> 4.3.1 → https://git.io/JvbeE
xkapastel has joined #nixos
<mokasin> 20.03
<mokasin> oops, ignore that
mokasin has quit [Quit: Leaving]
<{^_^}> [nixpkgs] @terlar opened pull request #84191 → Unmark nix-diff as broken → https://git.io/JvbeF
peelz has quit [Quit: Leaving]
Ariakenom has joined #nixos
<{^_^}> [nixpkgs] @eadwu closed pull request #81454 → nvidia_x11: 440.59 -> 440.64 → https://git.io/Jv2L7
<{^_^}> [nixpkgs] @marsam merged pull request #84175 → emplace: 0.2.14 -> 0.2.15 → https://git.io/JvF70
<{^_^}> [nixpkgs] @marsam pushed 2 commits to master: https://git.io/Jvbvs
Ariakenom_ has quit [Ping timeout: 260 seconds]
peelz has joined #nixos
<{^_^}> [nixpkgs] @marsam merged pull request #84169 → caddy2: 2.0.0-beta10 -> 2.0.0-rc.1 → https://git.io/JvF9u
<{^_^}> [nixpkgs] @marsam pushed 2 commits to master: https://git.io/JvbvZ
<{^_^}> [nixpkgs] @mmahut pushed to revert-84025-auto-update/tessera « Revert "tessera: 0.10.2 -> 0.10.4" »: https://git.io/Jvbvl
<{^_^}> [nixpkgs] @primeos opened pull request #84192 → python3Packages.cryptography: 2.8 -> 2.9 → https://git.io/Jvbv4
<{^_^}> [nixpkgs] @marsam merged pull request #84172 → kondo: init at 0.3 → https://git.io/JvFQ6
<{^_^}> [nixpkgs] @marsam pushed 2 commits to master: https://git.io/JvbvR
<{^_^}> [nixpkgs] @mmahut opened pull request #84193 → Revert "tessera: 0.10.2 -> 0.10.4" → https://git.io/JvbvE
ATuin has quit [Quit: WeeChat 2.7.1]
<{^_^}> [nixpkgs] @marsam merged pull request #84152 → lxd: 3.23 -> 4.0.0 → https://git.io/JvF6u
<{^_^}> [nixpkgs] @marsam pushed 2 commits to master: https://git.io/Jvbvu
captjakk has joined #nixos
Darkmatter66_ has joined #nixos
<evils> should i mark a package that fails because 2 of its dependencies aren't available on darwin, as broken on darwin?
Darkmatter66 has quit [Ping timeout: 256 seconds]
<evils> these dependencies are suppose do work on darwin
<evils> s/suppose do/supposed to/
<bqv[m]> i will ask once again for anyone who's awake, should i prefer fetchgit to fetchurl? https://github.com/NixOS/nixpkgs/pull/84004#pullrequestreview-386837689
plutes has joined #nixos
plutes has quit [Remote host closed the connection]
plutes has joined #nixos
leonardp has joined #nixos
linarcx has quit [Quit: WeeChat 2.8]
<{^_^}> [nixpkgs] @Ma27 merged pull request #84186 → pulseeffects: 4.7.1 -> 4.7.2 → https://git.io/JvFpM
leonardp has left #nixos [#nixos]
<{^_^}> [nixpkgs] @Ma27 pushed 4 commits to master: https://git.io/Jvbf8
<{^_^}> [nixpkgs] @Ma27 merged pull request #84182 → Update boost → https://git.io/JvFAA
plutes has joined #nixos
plutes has quit [Changing host]
<peelz> bqv[m]: checking out a repo is usually more expensive than a tarball... so I'd say fetchurl > fetchgit but idk
<{^_^}> [nixpkgs] @cdepillabout merged pull request #84191 → Unmark nix-diff as broken → https://git.io/JvbeF
<{^_^}> [nixpkgs] @cdepillabout pushed 2 commits to haskell-updates: https://git.io/Jvbfg
<peelz> bqv[m]: I guess fetchgit looks cleaner than interpolating a bunch of variables... I can't see any other possible benefits
linarcx has joined #nixos
<evils> easier to get a specific rev, easier to change the source?
Ariakenom_ has joined #nixos
Ariakenom0 has joined #nixos
rogue_koder has joined #nixos
Ariakenom has quit [Ping timeout: 258 seconds]
Ariakenom_ has quit [Ping timeout: 260 seconds]
<bqv[m]> fair enough.
<bqv[m]> fetchgit it is
<bqv[m]> sorry for late response, at this point i wasn't even expecting a reply lol
<peelz> lol no worries
ArdaXi has quit [Ping timeout: 256 seconds]
ddellacosta has joined #nixos
nobrak` has joined #nixos
ArdaXi has joined #nixos
cfricke has quit [Quit: WeeChat 2.7.1]
nobrak has quit [Ping timeout: 256 seconds]
<ottidmes> bqv[m]: you do realize there are people from all over the world here, to me its Friday afternoon :P
<bqv[m]> me to
koleesch has joined #nixos
<ottidmes> bqv[m]: not sure about fetchgit, I believe it has the same problems as builtins.fetchGit in that it includes the history, which does not have to be much of a problem, but 100M vs 1.8G it is for nixpkgs, so I prefer fetchurl/fetchFromGitHub for most things
<bqv[m]> that's what i thought
<bqv[m]> but cole sounds like he prefers fetchgit
<bqv[m]> and in this case the host is notabug.org
<bqv[m]> so there's no fetchFrom* yet
gxt has quit [Ping timeout: 240 seconds]
<ottidmes> bqv[m]: seems like fetchgit should be safe to use if configured properly, just finding the naming confusing, https://github.com/NixOS/nixpkgs/blob/master/pkgs/build-support/fetchgit/default.nix deepClone means leaving out the .git? Intuitively I would think deep clone means including the .git (as in include history)
<b42> --deepClone Clone the entire repository / --no-deepClone Make a shallow clone of just the required ref
shibboleth has joined #nixos
Ariakenom has joined #nixos
<ottidmes> Yeah, I was confused about the leave, remove would have been cleared to me
Ariakenom_ has joined #nixos
<ottidmes> or keep
<ottidmes> I had in my mind, leave = let go off
<bqv[m]> i see
<bqv[m]> ok
virus_dave has quit [Quit: virus_dave]
<ottidmes> bqv[m]: so adding deepClone = false; should make it better than fetchurl
Ariakenom0 has quit [Ping timeout: 265 seconds]
cosimone has quit [Quit: Terminated!]
Ariakenom has quit [Ping timeout: 256 seconds]
rogue_koder has quit [Remote host closed the connection]
rogue_koder has joined #nixos
zupo has joined #nixos
noudle has quit []
zeenk has quit [Ping timeout: 272 seconds]
Ariakenom0 has joined #nixos
cybrian has quit [Remote host closed the connection]
rauno has quit [Remote host closed the connection]
justanotheruser has quit [Ping timeout: 272 seconds]
Ariakenom_ has quit [Ping timeout: 260 seconds]
cybrian has joined #nixos
kenran has quit [Ping timeout: 240 seconds]
linarcx has quit [Ping timeout: 260 seconds]
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/fa0bd3f902a (from 19 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
ottidmes has quit [Ping timeout: 256 seconds]
plutes has quit [Read error: Connection reset by peer]
cybrian has quit [Ping timeout: 240 seconds]
plutes has joined #nixos
linuus has quit [Ping timeout: 256 seconds]
wrunt has quit [Ping timeout: 256 seconds]
wrunt has joined #nixos
silver_hook has quit [Ping timeout: 256 seconds]
linuus has joined #nixos
emilsp has joined #nixos
silver_hook has joined #nixos
jakobrs has joined #nixos
<jonge[m]> hey there. i would like to create a nixos module that, if my option `foo.enable = true` is set, overrides what `system.build.installBootloader` does (because i need to use a patched install-grub.pl file). so i am not overriding a package but a module attribute. is that possible somehow?
erasmas has joined #nixos
<{^_^}> [nixpkgs] @lheckemann pushed to structured-attrs « stdenv/linux: fix iconv support in libunistring »: https://git.io/JvbUa
justanotheruser has joined #nixos
<jakobrs> jonge[m]: Have you tried just setting it via `config` in your module?
knupfer has quit [Quit: knupfer]
knupfer has joined #nixos
<jakobrs> config.system.build.installBootloader = <value>;
knupfer has quit [Client Quit]
knupfer has joined #nixos
gulplante has quit [Ping timeout: 264 seconds]
<mlatus[m]> jonge: should be something like `system.build = lib.mkAfter { installBootLoader = <value>; }
gulplante has joined #nixos
<jonge[m]> mlatus: awesome, i will try that in a few minutes. one question: why not `system.build.installBootLoader = lib. mkAfter ...;` ?
Ariakenom0 has quit [Quit: WeeChat 2.7]
pluplog has joined #nixos
domogled has quit [Ping timeout: 256 seconds]
<mlatus[m]> jonge: because of the typedef of the `system.build` is `types.attrs` rather than some submodule, so the merging is a little odd (I am no nixos module system expert, this is what I remember from my personal experiment, may be not correct, you could use nixos-option to test if your config is applied)
<jonge[m]> mlatus: this works. thank you for your help!
<mlatus[m]> jonge: cheers!
<Freneticks> Is there a way to see update pending (like apt --upgradable in debian), and check package that have security issue like audit cve ?
<tilpner> Freneticks: See vulnix
Maxdamantus has quit [Ping timeout: 240 seconds]
rogue_koder has quit [Read error: No route to host]
Darkmatter66_ has quit [Ping timeout: 256 seconds]
rogue_koder_ has joined #nixos
Maxdamantus has joined #nixos
CMCDragonkai1 has quit [Ping timeout: 256 seconds]
<Freneticks> tilpner: it doesn't seem to have a : Show only packages that have already been fixed
<Freneticks> i'm not a nix maintainer
Darkmatter66 has joined #nixos
<aanderse> so i know nothing about mac. i have been given access to a mac and tried to install nix as described by manual
<aanderse> turns out / is mounted read only
<aanderse> anyone know anything about mac and can help me out?
<mlatus[m]> jonge: can confirm, value of `types.attrs` are just merged using //, if you are still interested
<tilpner> Freneticks: Ahh, you want notifications about vulnerabilites you could fix by updating. I don't know a tool for that
<Freneticks> yep
<tilpner> #nixos-security might know one (or it might ignore you for a day)
linarcx has joined #nixos
<jonge[m]> mlatus: thank you!
<Freneticks> this kind of tool is kind of like the base of a os
<Freneticks> Also is there a way to know pending package before nixos switch ?
<tilpner> You can see what would be installed
<Freneticks> how ?
<tilpner> But what would be upgraded? That's not well defined
ixxie has joined #nixos
<tilpner> Let's say you have three versions of A installed: 1, 2, and 3. And now you update to versions 2, 4, and downgrade the other to 2. How do you display that?
<tilpner> Because packages can exist in multiple instantiations (and with independent closures), it's not as simple as "check what version A had before, compare to what it has now"
rogue_koder_ has quit [Ping timeout: 258 seconds]
rogue_koder_ has joined #nixos
<Freneticks> if i do nix-channel --update and then nixos switch (It install new things but it's very very clear what
alina has quit [Quit: leaving]
glitters` has joined #nixos
<tilpner> Freneticks: No, it doesn't say what changes. It tells you what it needs to fetch/build/realise to change, but not what packages will be replaced with which versions
<Freneticks> this is crazy
mallox has quit [Quit: WeeChat 2.7.1]
<tilpner> Freneticks: E.g. if you rollback, and then update again, it doesn't need to do anything, because that system is already built. So it doesn't print anything
markus1189 has quit [Ping timeout: 256 seconds]
aveltras has joined #nixos
glittershark has quit [Ping timeout: 256 seconds]
magnetophon has quit [Read error: Connection reset by peer]
<tilpner> Freneticks: If you want, you can compare what's in the build-time closures of each system. You can then look up CVEs for each version, and compare those
<Freneticks> yeah no i'm just playing maybe nixos is not for me
<Freneticks> I want total control over package
<tilpner> You have that
<Freneticks> I mean in a security way
<tilpner> You can easily script that, and perhaps it's already been done
<tilpner> (Easily only when you have an API to map (package name, version) -> list of CVEs)
<Freneticks> I declare my state in *.nix and then nixos switch, but then if i update the channel, new version of package are definied so i do nixos switch and many package is replaced but I can't know what and why, how do you call that control ?
<{^_^}> [nixpkgs] @primeos merged pull request #84107 → [19.09] chromium: 80.0.3987.149 -> 80.0.3987.163 (backport) → https://git.io/JvFqo
<{^_^}> [nixpkgs] @primeos pushed 5 commits to release-19.09: https://git.io/Jvbkq
ekleog_ has quit [Quit: WeeChat 2.7.1]
ekleog has joined #nixos
rogue_koder_ has quit [Ping timeout: 256 seconds]
<tilpner> Freneticks: I can edit, pin, and rollback those channels however I want. I control when things update, and can filter unfree packages by license, and insecure packages by name
philr has quit [Ping timeout: 240 seconds]
<{^_^}> [nixpkgs] @flokli merged pull request #84008 → notmuch: extract optional dependencies → https://git.io/Jvdau
<{^_^}> [nixpkgs] @flokli pushed 2 commits to master: https://git.io/JvbkG
<ekleog> ma27[m]: By chance, are you using a self-hosted nextcloud nixos module? (trying the maintainers one by one, but you're the last one ^^') When I'm trying to use the nextcloud-client nixos package or the Talk android app with my instance, I get an endlessly-looping “please wait” on the “connect” button -- do you also see that, or is it just me?
<tilpner> Freneticks: What you mean by lack of control is probably more a lack of insight, and then I might agree
knupfer has quit [Ping timeout: 240 seconds]
<Freneticks> probably that's why I ask
bennofs is now known as ALLES_bennofs
<ma27[m]> ekleog: yes I do
<ma27[m]> and no, nextcloud-client works for me
<ekleog> uhhh :( did you need to configure something special for it, aside from what's in the docs?
<ma27[m]> do you have $BROWSER set?
lovesegfault has joined #nixos
<Freneticks> tilpner: when i say update channel and I don't mean major upgrade like 18 > 19 but only refresh the channel
<ma27[m]> well it opens the login dialog in a browser
<ekleog> I don't, but it does open a firefox page, which also goes to an endlessly-looping “Grant Access” after I click it
rogue_koder_ has joined #nixos
jakobrs has quit [Ping timeout: 240 seconds]
chloekek has quit [Ping timeout: 256 seconds]
<ma27[m]> ohh I see....
<ma27[m]> but no, never experienced this
<tilpner> Freneticks: Yes, I assumed so
<ma27[m]> I'm currently on 2.6.4
<ekleog> oooh just saw Content Security Policy: The page’s settings blocked the loading of a resource at http://cloud.gaspard.io/login/v2/grant (“form-action”).
<ekleog> (I'm also on 2.6.4git for nextcloud-client; the version on 20.03 basically)
<tilpner> Freneticks: I realise it must seem strange to someone familiar with e.g. apt, but I've never cared for the full list of version changes
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<ekleog> oooh found it, there's no `s` to the `http`
<tilpner> Freneticks: I often check what version a package has on different versions of nixpkgs, but the full list of every change?
knupfer has joined #nixos
<ekleog> now, why is it not having an `s`…
<tilpner> Freneticks: That's nothing I ever thought to query. Perhaps there is an easy way that I just never noticed
<yorick> timokau[m]: so I have tensorflow without mkldnn, why is it spending 70% of its time in mkldnn::impl::cpu::(anonymous namespace)::kernel_mxn<float, false, false>
<Freneticks> tilpner: okay will search, I just want to know what my system do.
<{^_^}> [nixpkgs] @edolstra opened pull request #84194 → kscreen: Fix #82141 → https://git.io/JvbkP
<Freneticks> If he upgrade a package, he need to print me the version, and maybe the cve fixed
<ekleog> ooh, looks like services.nextcloud.config.overwriteProtocol is probably it
<ekleog> weird that it didn't do it before
<tilpner> Freneticks: Well, that's the conceptual difference from apt to nix. It doesn't upgrade a package in place, it leaves the old package just as it was.
<ekleog> but then, it was already there on 19.09, so it's probably in the release notes of 19.03 or 19.09 that it should have been, so too late to fix I guess
<tilpner> Freneticks: And instead it builds your system entirely fresh (reusing what was already there) every time, it doesn't ever look at the old package
rogue_koder_ has quit [Ping timeout: 256 seconds]
rogue_koder has joined #nixos
<Freneticks> tilpner: yeah i know that
Blessjah has quit [Remote host closed the connection]
<{^_^}> Channel nixos-20.03-small advanced to https://github.com/NixOS/nixpkgs/commit/cf0855c899c (from 4 hours ago, history: https://channels.nix.gsc.io/nixos-20.03-small)
<ekleog> ma27[m]: I can confirm it worked! Thank you for your help :D
magnetophon has joined #nixos
markus1189 has joined #nixos
MinceR is now known as MinceR`
MinceR` is now known as MinceR
<ekleog> bachp globin fpletz: As I poked you before on my issue, it was due to services.nextcloud.config.overwriteProtocol ; and is now solved :)
hmpffff has quit [Quit: nchrrrr…]
koleesch_ has joined #nixos
koleesch__ has joined #nixos
rogue_koder has quit [Ping timeout: 240 seconds]
elibrokeit is now known as ztrawhcse
systo has quit [Quit: .]
ztrawhcse is now known as elibrokeit
systo has joined #nixos
koleesch has quit [Ping timeout: 272 seconds]
koleesch_ has quit [Ping timeout: 256 seconds]
elibrokeit is now known as ztrawhcse
ztrawhcse is now known as guys
linarcx has quit [Quit: WeeChat 2.8]
fiddlerwoaroof_ is now known as fiddlerwoaroof
orivej has joined #nixos
jco has joined #nixos
<ornxka> i try to build lxd 4.0 and i get error about selinux cant find glibc symbol
<ornxka> /nix/store/ajrrkivdfvp8dp4vdg5hp1h5hblmanc9-binutils-2.31.1/bin/ld: /nix/store/c4n3vccbswvapz92419v2y8yzyhgs6r1-libselinux-2.9/lib/libselinux.so.1: undefined reference to `gettid@GLIBC_2.30'
<jco> Hello! I messed up with my nix-path: getting warning: Nix search path entry 'path/to/nixpkgs' does not exist, ignoring
guys is now known as elibrokeit
<jco> how to recover from this situation?
<ornxka> is $NIX_PATH set to anything
cap_sensitive has joined #nixos
<jco> ornxka: it is
<jco> should remove it?
<ornxka> maybe either try unsetting it or setting it to something that makes more sense
<ornxka> i do not actually know what the problem is or how to fix it but thats where i would probably start
<cap_sensitive> Hi. I'm trying to add NUR to my pkgs: https://github.com/nix-community/NUR. Could someone explain why is 'inherit pkgs' needed??
mexisme has joined #nixos
<{^_^}> [nixpkgs] @drewrisinger opened pull request #84197 → pythonPackages.pint: 0.9 -> 0.11 → https://git.io/JvbIE
jrddunbr has quit [*.net *.split]
sauyon has quit [*.net *.split]
sauyon has joined #nixos
jrddunbr has joined #nixos
Shell is now known as VShell
VShell is now known as Shell
<{^_^}> [nixpkgs] @ryantm merged pull request #84144 → minecraft: 2.1.11314 -> 2.1.13509 → https://git.io/JvFoW
<{^_^}> [nixpkgs] @ryantm pushed 2 commits to master: https://git.io/JvbIo
rogue_koder has joined #nixos
<ornxka> i changed libselinux to unstable and now its rebuilding absolutely freaking everything
mexisme has quit [Ping timeout: 256 seconds]
<ornxka> do a lot of packages use that or something?
<melling> cap_sensitive: it passes through your reference from the outer scope, i think it is to allow you to specify a specific nixpkgs for nur to use
cole-h has joined #nixos
linarcx has joined #nixos
M-Gregoire495 has quit [Quit: The Lounge - https://thelounge.chat]
M-Gregoire495 has joined #nixos
<cap_sensitive> melling: Ha! I see! So that inherit pkgs is passed to the nur! Thanks!
<melling> cap_sensitive: yes, it's like shorthand for doing { pkgs = pkgs }. you might want to pass through a different nixpkgs from the default, for instance with your own overrides etc
<jco> ornxka: thx for the advice, gonna rollback :)
jco has quit [Quit: WeeChat 2.7]
drakonis has joined #nixos
emacsomancer has quit [Read error: Connection reset by peer]
emacsomancer has joined #nixos
markus1189 has quit [Ping timeout: 250 seconds]
markus1189 has joined #nixos
<bsima> when I do `builtins.match ".*(a.).*" "abacadae"` I only get back `["ae"]` but I want to return all of the matches, does anyone know how I can do this?
melling has quit [Quit: WeeChat 2.7.1]
shibboleth has quit [Quit: shibboleth]
<bsima> hm, i think i'll just use lib.strings to split or duplicate the string i'm working on. not ideal for performance but i can't find a way to do multiple matches with nix regex
<cole-h> If you want performance, you're using the wrong language :P
hammernix has joined #nixos
<hammernix> How long does it generally take for a merged PR to make it into unstable channel? e.g. https://github.com/NixOS/nixpkgs/pull/83878
<{^_^}> #83878 (by marsam, 3 days ago, merged): ruby: 2.5.7 -> 2.5.8, 2.6.5 -> 2.6.6, 2.7.0 -> 2.7.1
<cole-h> Whenever unstable finishes evaluating/building successfully.
cap_sensitive has quit [Quit: WeeChat 2.7]
<{^_^}> [nixpkgs] @mhennings opened pull request #84198 → gitAndTools.stgit: Fix broken package. → https://git.io/JvbLl
lecethim1n has quit [Quit: leaving]
<hammernix> i see, so there's a build problem currently, seemingly gumming up new unstable channel releases?
gulplante has quit [Quit: WeeChat 2.7.1]
markus1189 has quit [Ping timeout: 256 seconds]
kqb has quit [Ping timeout: 250 seconds]
<ryantm> bsima: Unfortunately, it doesn't work like that, so you'll have to make something more complex. Maybe you could use lib.splitString
glitters` has quit [Remote host closed the connection]
zupo has joined #nixos
<bsima> ryantm: yeah i'm doing `lib.splitString "\n"` and then doing the match, since i'm matching on whole lines anyway
smatting has quit [Ping timeout: 265 seconds]
<{^_^}> [nixpkgs] @kalbasit merged pull request #84055 → prow: 2019-08-14 -> 2020-04-01 → https://git.io/Jvd94
<{^_^}> [nixpkgs] @kalbasit pushed to master « prow: 2019-08-14 -> 2020-04-01 (#84055) »: https://git.io/JvbLX
<gchristensen> anyone happen to know a version of the kernel which vbox works with? :)
<etu> 5.4 and older
<gchristensen> aye.
<gchristensen> bummer :/
<etu> yeah
<etu> I wanna go to 5.5, that's how I know :D
<gchristensen> hehe
<gchristensen> can you test a PR for me? :)
<etu> Sure, in a bit :)
markus1189 has joined #nixos
<{^_^}> nix-community/nixops-vbox#3 (by grahamc, 1 minute ago, open): Poetry2nixify
mawis has joined #nixos
chloekek has joined #nixos
Blessjah has joined #nixos
jakobrs has joined #nixos
<jakobrs> How do I use the Hydra interface to figure out when a specific package broke?
thetet has quit [Quit: Leaving.]
growpotkin has joined #nixos
fendor has joined #nixos
captjakk has quit [Remote host closed the connection]
rogue_koder has quit [Remote host closed the connection]
captjakk has joined #nixos
rogue_koder has joined #nixos
<{^_^}> [nixpkgs] @mmilata opened pull request #84199 → libmicrodns: 0.1.0 -> 0.1.2 → https://git.io/Jvbtt
<gchristensen> out of the box this mac boots NixOS but doesn't activate the network using the thunderbolt adapter on ens9
<gchristensen> any tips on making this happen automatically? after `ifconfig ens9 up` and `dhcpd -n ens9` it works fine
<scoates> hi; this is me. I was bugging gchristensen elsewhere (-:
<scoates> long time nix user, first time nixos caller
captjakk has quit [Remote host closed the connection]
captjakk has joined #nixos
<gchristensen> hehe
<{^_^}> [nixpkgs] @cole-h opened pull request #84200 → zoxide: 0.3.0 -> 0.3.1 → https://git.io/Jvbt4
<gchristensen> and contributing an ofborg builder :P
<scoates> …theoretically (-;
<jjwatt[m]> scoates: something like `networking.interfaces.ens9.useDHCP = true;`? Did you try that yet?
<scoates> yep
<jjwatt[m]> guess that doesn't automatically bring it up, eh
<scoates> (still nothing in `ifconfig` just `ifconfig -a`, after reboot)
<scoates> maybe useful information: I have lo and wlp3s0, but I'd like to not use the wireless device
jumper149 has joined #nixos
mawis has quit [Quit: WeeChat 2.7.1]
shibboleth has joined #nixos
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/e1a880fca93 (from 6 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<bqv[m]> Apologies jgarte and ottomides for not noticing pms
cybrian has joined #nixos
janneke has joined #nixos
mallox has joined #nixos
jbrock has quit [Quit: jbrock]
jbrock has joined #nixos
jbrock has quit [Client Quit]
jbrock has joined #nixos
zupo has joined #nixos
<gchristensen> hey wait libvirtd is really nice
<gchristensen> using nixops with the libvirtd backend for the first time feels like being transported to the future
cybrian has quit [Ping timeout: 265 seconds]
<mallox> hey, does anyone know how to roll out my dotfiles via git/stow in my configuration.nix? (I know about home-manager, but I want mutable dotfiles)
<jakobrs> Wait you can use nixops with libvirtd? I'd always assumed that it only supported virtualbox for "virtual" deployments
jbox has joined #nixos
ixxie has quit [Ping timeout: 264 seconds]
<gchristensen> okay the magic is gone, "OSError: [Errno 28] No space left on device" and I don't know why lol
<cole-h> Quickest about-face in the west
ixxie has joined #nixos
<scoates> ok, I think my problem is related to systemd; I have files for `wlp3s0` but not `ens9`. Is there a way to generate these post-install?
jbox has left #nixos ["WeeChat 2.8"]
jbox has joined #nixos
<scoates> Worst kind of screen shot but I don’t have any tooling on that box yet. https://usercontent.irccloud-cdn.com/file/JBg7R3ku/1585936870.JPG
<lordcirth> What do I put in configuration.nix to make lib.mkForce available in it?
<jakobrs> lordcirth: Add lib to the arguments list
<cole-h> If it has `{ pkgs, ... }:` at the top, you could add lib there.
<jakobrs> like { lib, pkgs, ... }:
<lordcirth> ok thanks
<cole-h> (or just `{ ... }`)
markus1189 has quit [Ping timeout: 256 seconds]
<lordcirth> I thought ... allowed any other arguments that were passed, but just having ... didn't work
<cole-h> `...` accepts other arguments but doesn't make them available unless you bind them
<lordcirth> Ah, I see! Thanks
<cole-h> e.g. `{ ... }@args:` makes `args.pkgs` and `args.lib` possible
<lordcirth> So it's more of a "silently ignore"
<gchristensen> scoates: w.a.g. here but what if you did networking.interfaces.ens9 = {}; ?
<scoates> I already that .useDHCP = true, but I can try the empty object yeah. Need a few mins.
<cole-h> Right. "Allow its presence, but don't make its presence known" kinda thing
<{^_^}> [nix] @edolstra pushed to flakes « Publish a tarball containing the crates we depend on »: https://git.io/JvbqW
domogled has joined #nixos
markus1189 has joined #nixos
jakobrs has quit [Quit: WeeChat 2.7.1]
<__monty__> Hmm, is will #83689 affect nixpkgs-unstable as well?
<{^_^}> https://github.com/NixOS/nixpkgs/pull/83689 (by flokli, 5 days ago, merged): go_1_14: 1.14 -> 1.14.1
<{^_^}> [nix] @edolstra pushed to flakes « Fix job name »: https://git.io/JvbqV
pbogdan has quit [Ping timeout: 264 seconds]
<flokli> __monty__: this will eventually trickle down to this channel too
<flokli> what's the question?
fendor has quit [Quit: Leaving]
<flokli> currently nixpkgs-unstable seems 6 days old: status.nixos.org
<scoates> gchristensen: same-same
<gchristensen> hrm.
<flokli> okay, fixed since two hours. should update soon :-)
<gchristensen> scoates: I don't know :/
<__monty__> flokli: I'm just waiting the fix. Some software I use suffers from the problem.
<scoates> how does one add an ethernet device after a system is already running? is that a known process?
pbogdan has joined #nixos
<colemickens> "build problem". hm. descriptive. :P
<gchristensen> scoates: it has always "just worked" for me, which is why I'm so stumped
<{^_^}> [nix] @edolstra pushed 2 commits to master: https://git.io/Jvbqi
<flokli> __monty__: you might be able to reboot to get a new kernel, if that's an option ;-)
<lordcirth> Is there any way to name or bookmark certain system generations?
<scoates> perhaps shame incoming for me.
<lordcirth> Other than just making a symlink to them somewhere, I guess
<gchristensen> uh oh :)
<__monty__> flokli: nix-darwin
<flokli> umm… so there's more in 1.14.1 than the fix for https://github.com/golang/go/issues/37807 ?
<{^_^}> golang/go#37807 (by gopherbot, 3 weeks ago, closed): runtime: mlock of signal stack failed: 12 [1.14 backport]
pluplog has quit [Read error: Connection reset by peer]
<flokli> scoates: not sure, but this sounds like your ethernet interface appears to late for the shellscripted networking stack to pick it up
<__monty__> flokli: #82606 was my original issue. Which points to https://github.com/golang/go/issues/37459
<{^_^}> https://github.com/NixOS/nixpkgs/issues/82606 (by toonn, 2 weeks ago, open): direnv 2.21.2 on macOS (10.13.6) is broken
<{^_^}> golang/go#37459 (by carlca, 5 weeks ago, closed): SIGILL: illegal instruction on any go tool under macOS
<flokli> scoates: you might want to try with networkd, which does all that properly also when devices appear later
<flokli> __monty__: ah, sweet - another issue
pluplog has joined #nixos
<flokli> well, if you're fancy, you can just checkout nixpkgs to a commit after that merge, and point -I nixpkgs=/path/to/checkout to switch to it
<flokli> if you want to migitate this for now
<scoates> uh-oh indeed
<__monty__> Would that involve lots of building though?
<scoates> rebooting for shame (-;
<scoates> remember that part about first-time nixos caller? (-;
<scoates> (I didn't `nixos-rebuild`… derp) I have a different problem now, but it does look like it created those systemd units
<scoates> )
<{^_^}> [nixpkgs] @mmilata opened pull request #84202 → [20.03] nixos/sympa: fix outgoing emails, update package version → https://git.io/JvbqN
lsix1 has joined #nixos
jb55 has quit [Quit: jb55]
markus1189 has quit [Ping timeout: 260 seconds]
thetet has joined #nixos
o1lo01ol1o has quit [Remote host closed the connection]
<flokli> __monty__: there's only one way to find out ;-)
o1lo01ol1o has joined #nixos
jb55 has joined #nixos
<monokrome> hmmm... Is there not a way to create an offline mirror of nixos packages?
<monokrome> Like, since they all use git and such
o1lo01ol1o has quit [Remote host closed the connection]
zupo has quit [Ping timeout: 260 seconds]
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JvbmT
<bqv[m]> Goddamn it. Literally MINUTES before I push to my pr changing fetchurl to fetchgit, jon has let us know that we should prefer fetchurl. I have the worst timing
<__monty__> flokli: I'm not very brave and I have the patience of a saint ; )
<{^_^}> [nixpkgs] @zimbatm opened pull request #84203 → nim: 1.0.6 -> 1.2.0 → https://git.io/Jvbmk
<flokli> __monty__: I don't know how much you rebuild, as I don't know how many packages you pull in ;-)
<__monty__> monokrome: What do you mean? If you clone the nixpkgs repo you can build everything.
o1lo01ol1o has joined #nixos
jb55 has quit [Client Quit]
<monokrome> __monty__: like, every package?
<tilpner> Not really, nixpkgs alone isn't that useful
<tilpner> You need access to all the sources, or a cache that had
<__monty__> monokrome: Copying all the build products from the cache wouldn't be hard per se either.
hammernix has quit [Remote host closed the connection]
<__monty__> tilpner: Ah, true.
<monokrome> but is there a way to build a completely offline mirror of nixpkgs? or no?
<monokrome> like if I want to point the nixos installer at a local mirror
<tilpner> monokrome: Depends. What do you want to with it?
<bqv[m]> monokrome: how much space do you have :p
<tilpner> You can easily copy the cache for a single revision
<monokrome> 40TB
<__monty__> You could clone the repo and evaluate the src attribute for everything I guess?
<monokrome> hmm
zupo has joined #nixos
<simpson> monokrome: To anticipate several possible use-cases: How do you want to receive security updates?
<monokrome> I was considering that but hoping there was a better way
<tilpner> There is, but answer the question
<scoates> now I'm seeing `failed to mount /rpool`… pretty sure I'm doing something *different* dumb.
<monokrome> simpson: by updating my offline mirror...
<scoates> gchristensen: ^ network problem seems to be solved at least.
<gchristensen> nice
thibm has quit [Quit: WeeChat 2.6]
<gchristensen> scoates: let's see your config :)
<exarkun> I have a `nix-build` command that fails because "The pkg-conf packages does not contain the package: libchallenge_bypass_ristretto (pkg-conf dependency).". How can I get at pkg-conf to poke around and see what's going on? The build command is this, fwiw:
<exarkun> nix-build --keep-failed -A 'components.exes.PaymentServer-exe' --arg callPackage '(import <nixpkgs> { }).callPackage' ./nixos/pkgs/zkapissuer.nix
kenran has joined #nixos
<simpson> monokrome: Okay. Due to how Nix works, are you okay with rebuilding your entire offline cache every time there's a security update? I'm not sure how disconnected "offline" is; presumably it's still hooked up to some sort of LAN.
<monokrome> wouldn't it only rebuild things that changed?
<monokrome> when I `nixos-rebuild`, my whole system doesn't rebuil
<tilpner> Yes, but with nixpkgs a lot of things can change for little reason
<monokrome> yeah that's fine
<monokrome> expected, even
<lordcirth> monokrome, yes, but "changed" includes "anything in it's closure changed in the slightest"
<tilpner> Update a core package? All dependent packages have changed
<simpson> And you currently benefit from the *online* caches, which are pretty sizeable and have a really good hit rate.
<monokrome> yes I know
<monokrome> we can assume there's a security issue in musl
<scoates> heh hard to share without networking (-; need a min
<monokrome> time to rebuild eveerything
o1lo01ol1o has quit [Ping timeout: 260 seconds]
<monokrome> that's fine
<gchristensen> monokrome: nix absolutely suppors offline mirrors
<simpson> So that's why we're not just curious about what you're planning to build, but hoping to glean essential details.
<monokrome> gchristensen: Is there a process for creating one?
<tilpner> monokrome: See PM
<nh2> can anybody see the chat in peti's live stream at https://www.twitch.tv/peti343/ ? It doesn't load for me
<nh2> always says "Connecting to Chat"
<monokrome> thanks tilpner!
<gchristensen> I don't mean to be dismissive, but identify the store paths you want, and mirror them. without knowing more about your use case and what exactly you want to mirror, I don't have more info. for example, do you want to mirror every prebuilt thing for a given evaluation?
<simpson> The main surprise is that you're asking for *everything*, including stuff that is in an expensive long tail. So, like, who's gonna build that, are you really gonna hold all of it, etc. That's all. Not trying to pry *too* much at what is almost certainly a plan for a walled corporate LAN of some sort.
<timokau[m]> yorick: I don't know. I should really remove myself from the tensorflow maintainers, I'm not really maintaining it :/
<lordcirth> monokrome, do you want it totally offline, or just minimize WAN load?
<lordcirth> If the latter, just slap squid in front
<monokrome> totally offline
<scoates> (I am also a zfs noob)
snicket_ has joined #nixos
<gchristensen> scoates: fileSystems."/rpool" = <- probably don't need this bit
* scoates nods
<snicket_> ,channels
<{^_^}> Largest Nix-related channels: #nixos, #nixos-dev, #nix-darwin, #nixos-chat, #nixos-aarch64, #nixos-wiki, #nixos-borg, #nixos-security, #nix-core
<cransom> 'slapping squid' isn't really a thing anymore with https servers. you have to hijack connections now.
<scoates> it specifically says to not edit. can I edit? (-:
<gchristensen> yeah :)
* scoates living on the edge
<gchristensen> it is just that if you run nixos-generate-config again, it will fearlessly replace that file
jb55 has joined #nixos
<tilpner> cransom: You don't *have* to use https for this
<lordcirth> cransom, well, I think you can set your mirror on the client to be your squid server, and have squid rewrite the upstream URL back?
<lordcirth> Haven't done it that way myself, though
<snicket_> is it normal that when booting raspberry pi 3 from 'nixos-sd-image-19.09.2358.0ee9ceff38f-aarch64-linux.img' there is no default /etc/nixos/configuration.nix file anymore?
<scoates> highdpi console is cool (-:
<scoates> gchristensen: looks good now
<gchristensen> woot!
zaeph1 has joined #nixos
lorimer has quit [Ping timeout: 252 seconds]
<{^_^}> [nixpkgs] @peti merged pull request #84039 → haskellPackages: mass unmark non-broken packages → https://git.io/JvdDJ
<{^_^}> [nixpkgs] @peti pushed 2 commits to haskell-updates: https://git.io/Jvbm8
zaeph has quit [Ping timeout: 264 seconds]
markus1189 has joined #nixos
<{^_^}> [nixpkgs] @joachifm merged pull request #73763 → Improvements to the NixOS Hardened Profile → https://git.io/JeKSh
<{^_^}> [nixpkgs] @joachifm pushed 3 commits to master: https://git.io/Jvbm0
lorimer has joined #nixos
<{^_^}> [nixpkgs] @danielfullmer opened pull request #84204 → [WIP] nixos/systemd-boot: boot counting and automatic fallback → https://git.io/Jvbmg
selfsymmetric-mu has joined #nixos
<selfsymmetric-mu> Is there an issue filed for this vulnerability? https://www.us-cert.gov/ncas/current-activity/2020/04/01/google-releases-security-updates-chrome
gxt has joined #nixos
<selfsymmetric-mu> My version of Chromium is at `80.0.3987.149`, when it ought to be `.162`.
markus1189 has quit [Remote host closed the connection]
linarcx has quit [Read error: Connection reset by peer]
<{^_^}> #78450 (by primeos, 9 weeks ago, open): Chromium and Google Chrome need new maintainers
<cole-h> But channels also appear to be blocked
<cole-h> (or were)
<bqv[m]> monokrome:
<bqv[m]> Thats about as hard as having an offline gentoo builder
<selfsymmetric-mu> cole-h++
<{^_^}> cole-h's karma got increased to 11
erictapen has joined #nixos
erictapen has quit [Client Quit]
erictapen has joined #nixos
<scoates> gchristensen: ssh key please?
<gchristensen> github.com/grahamc.keys -- can't do more right now, 20.03 go/no-go starting
relaxmax has joined #nixos
<scoates> 👍
kenran has quit [Ping timeout: 256 seconds]
aveltras has quit [Quit: Connection closed for inactivity]
<{^_^}> [nixpkgs] @peti pushed 2 commits to haskell-updates: https://git.io/Jvbm5
markus1189 has joined #nixos
FRidh2 is now known as FRidh
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/Jvbmd
<jared-w> gchristensen: how tf did you get that url lol
<tilpner> jared-w: Substitute your user name in there
<cole-h> Yep, all GitHub keys are public
linarcx has joined #nixos
<cole-h> Can also see https://github.com/cole-h.gpg for your public GPG key
gustavderdrache has quit [Quit: Leaving.]
hmpffff has joined #nixos
koleesch_ has joined #nixos
snicket_ has quit [Remote host closed the connection]
<jared-w> oooh, nice. TIL
hmpffff has quit [Read error: Connection reset by peer]
<{^_^}> [nixpkgs] @dywedir opened pull request #84205 → hcxtools: 5.3.0 -> 6.0.0 → https://git.io/JvbmA
gustavderdrache has joined #nixos
hmpffff has joined #nixos
kfound has joined #nixos
<jared-w> I'm conflicted. I want linux 5.6.2 so that iwlwifi will finally work with 5ghz wifi again. We upgraded to gigabit at my house and I'm stuck on a slow 2.4ghz connection :(
<jared-w> On the other hand, I should responsibly wait until the nvidia_x11 PR gets merged...
<jared-w> RIP, forgot to turn f@h on overnight. Noooo
koleesch__ has quit [Ping timeout: 264 seconds]
<ornxka> i just have a big git repo with master/release in it and make changes/merge PRs in it as necessary
<ornxka> waiting for things takes too long
thetet has quit [Quit: Leaving.]
thetet has joined #nixos
<jared-w> It's more a self control thing so I don't nerdsnipe myself for 6 hours by accident instead of working
<atemu12[m]> <jared-w "I'm conflicted. I want linux 5.6"> Couldn't you just use 5.4?
thetet has quit [Client Quit]
<jared-w> atemu12[m]: my wifi card is too new. Support needs 5.5
<ornxka> <jared-w> It's more a self control thing so I don't nerdsnipe myself for 6 hours by accident instead of working
<jared-w> Miraculously, 2.4 wifi works pretty well; more or less flawlessly. And most 5ghz wifi networks seem to work great too. It's just 5ghz above a certain speed that makes the card fall over and die
<ornxka> ah....
<ornxka> you are smarter than me
<jared-w> ornxka: only because I've learned through years of stupidity ;)
<{^_^}> [nixpkgs] @peti pushed 552 commits to haskell-updates: https://git.io/JvbYm
o1lo01ol1o has joined #nixos
markus1189 has quit [Ping timeout: 256 seconds]
<ornxka> i am literally doing nixos stuff when i should be working at this very moment
<ornxka> i have much to learn i guess
domogled has quit [Remote host closed the connection]
<atemu12[m]> jared-w: So what about linuxPackages_5_5 then?
domogled has joined #nixos
<jared-w> heh, same. I tend to tinker around until I get my system into a stable state and then I freeze it and force myself to not do anything to it for like a month or two. Then I tinker around and lose a day or two of max productivity and repeat the cycle. Seems to work pretty well
fuzzy-id has joined #nixos
<ornxka> that sounds like me
<jared-w> atemu12[m]: I'm on 5.4 right now, my bad; my wifi card is the AX200 wifi 6 card from intel. I could probably do 5.5 but it's more broken for wifi than 5.4 is; 5.4 won't crash loop the wifi card fast enough to freeze my computer if I connect to my 5ghz wifi network. 5.5 will
<ornxka> i try not to update when i can avoid it but "chrome security update" sounds pretty important
<{^_^}> [nixpkgs] @shazow opened pull request #84206 → steam: Update generated runtime dependencies → https://git.io/JvbYC
<jared-w> ornxka: I just use a firefox overlay that's non-deterministic, so that marches forward regardless of nixpkgs. Pretty nice
<fuzzy-id> i use "fetchFromGitHub" to hook a package of mine into the system. as sha256 attribute i use the one given by `nix-prefetch-url https://github.com/OWNER/REPO/archive/REV.tar.gz'. but that results in the error "hash mismatch in fixed-output derivation …".
<{^_^}> [nixpkgs] @peti closed pull request #84155 → darcs: Fix build (Use old GHC) → https://git.io/JvF69
o1lo01ol1o has quit [Ping timeout: 265 seconds]
<fuzzy-id> so, what's the correct url to pass to nix-prefetch-url?
<tilpner> fuzzy-id: Try with --unpack
<fuzzy-id> yeah, that's it. thanks!
<armin> just recognized mandelbulber2 is not packaged for nix :(
<armin> considering how much i LOVE fractals and that i have quite some complex setup with remote rendering and such, that's probable something i like to change.
chloekek has quit [Ping timeout: 240 seconds]
chagra_ has quit [Ping timeout: 265 seconds]
snicket has joined #nixos
jbox has quit [Ping timeout: 252 seconds]
<tilpner> > mandelbulber.src.urls # armin
hammernix has joined #nixos
jbox has joined #nixos
jbox is now known as Guest62117
<armin> tilpner: i'm new to nixOS though - maintaining a couple of packages in various distributions, but fearing nixOS packages a bit still...
<armin> tilpner: a co-worker (he's also hiding in the depth of this channel) got me into this rabbit-hole and now there doesn't seem to be a way back out
<armin> damn you, eric.
<cole-h> I would argue NixOS packaging is easier than any other distribution. Make it work (the hardest part), submit a PR upstream, wait for it to get merged, done.
<armin> cole-h: come on, every packager of every distribution would say that about their packaging mechanism ;)
<lordcirth> It's easy, unless the official build process is "run this easy custom script that downloads all the deps"
<lordcirth> (looking at you, CKAN)
<armin> :D
<tilpner> armin: That line was meant to tell you it's already packaged
<cole-h> Except for Debian, RedHat, etc, where, for official inclusion, you need to get a sponsor or something
<FRidh> cole-h: packaging != maintaining
<armin> tilpner: oh!
<armin> tilpner: so why didn't i find it even with nix search -u?
<tilpner> armin: Old cache, or old nixpkgs
Guest62117 is now known as atlas_
atlas_ has quit [Quit: WeeChat 2.8]
<tilpner> Or wrong search term
<{^_^}> [nixpkgs] @peti pushed 2 commits to haskell-updates: https://git.io/JvbYj
<tilpner> Perhaps you included "2" in the search term
<armin> tilpner: ok thanks a lot for your kind help. :)
<armin> i tried both!
<tilpner> It shows up in the search here
<tilpner> Does nix search hello work for you?
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JvbOv
<armin> yes that works :D
<tilpner> Odd
<armin> it's very likely layer8 error, i'm new to all of this and still try to find my way around
<tilpner> Ahh, old nixpkgs
<tilpner> It seems to be more recent
<armin> oh okay :D
<tilpner> Not your fault then
<armin> again, thx for telling me so!
<armin> i'm super addicted to fractals btw, if you like some 4k wallpapers, maybe this is for you: fhttps://unix.porn/fractals/
<{^_^}> [nix] @edolstra pushed to flakes « Fix tests.githubFlakes evaluation »: https://git.io/JvbOT
<armin> oh, erroneous paste, https://unix.porn/fractals/
<armin> no way, the default.nix file is THAT simple for it?
<armin> :O
<{^_^}> [nixpkgs] @peti pushed commit from @chkno to haskell-updates « darcs: Fix build (Use old GHC) »: https://git.io/JvbO3
hammernix has quit [Remote host closed the connection]
<{^_^}> [nixpkgs] @edolstra closed pull request #83881 → Revert "nixos: default `environment.homeBinInPath` to false" → https://git.io/Jv5P7
<tilpner> armin: Ooh, pretty. I'll have to buy a new display to use them properly though
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JvbOZ
<armin> tilpner: it was the only reason i spent 400 bucks on the cheapest 4k 32" tft with usb-c i could find
<lordcirth> armin, nice wallpapers!
<armin> thanks <3
<armin> the site is hosted on a raspberry pi behind a layer 2 vpn, sorry for the slow transfer rates
<tilpner> Now you can administer your fractal rendering client farm with NixOS
<armin> i just had a look at nixops today
<armin> bit mind blowing, but hey
markus1189 has joined #nixos
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « all-cabal-hashes: update to Hackage at 2020-04-03T06:25:45Z »: https://git.io/JvbO4
<{^_^}> [nixpkgs] @ehmry merged pull request #84203 → nim: 1.0.6 -> 1.2.0 → https://git.io/Jvbmk
<{^_^}> [nixpkgs] @ehmry pushed commit from @zimbatm to master « nim: 1.0.6 -> 1.2.0 »: https://git.io/JvbOz
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « git-annex: update sha256 hash for new version 8.20200330 »: https://git.io/JvbOa
<{^_^}> [nixpkgs] @Ma27 merged pull request #83628 → [20.03] mongodb: update → https://git.io/JvQa3
<{^_^}> [nixpkgs] @Ma27 pushed 12 commits to release-20.03: https://git.io/JvbOP
<{^_^}> [nixpkgs] @prusnak opened pull request #84207 → openttd: 1.9.3 -> 1.10.0 → https://git.io/JvbO1
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « hackage2nix: drop obsolete version override »: https://git.io/JvbOy
pluplog has quit [Remote host closed the connection]
<ehmry[m]> ++ stdenv.lib.optional stdenv.targetPlatform.isGenode ./genode.patch;
<ehmry[m]> woops
zaeph1 is now known as zaeph
domogled has quit [Ping timeout: 265 seconds]
<{^_^}> [nixpkgs] @peti pushed 30 commits to haskell-updates: https://git.io/JvbOQ
<drakonis> genode huh
<drakonis> that's nice
blumenkranz has joined #nixos
<drakonis> now that is very cool
<ehmry[m]> all these matrix clients are aweful, I was trying to copy paste
kfound has quit [Remote host closed the connection]
<armin> oh, matrix, don't get me started
<{^_^}> [nixpkgs] @FRidh pushed 430 commits to staging-next: https://git.io/JvbOF
<{^_^}> [nixpkgs] @FRidh pushed 437 commits to staging: https://git.io/JvbOA
<{^_^}> [nixpkgs] @FRidh pushed 46 commits to staging-20.03: https://git.io/JvbOj
<ehmry[m]> uh yea, this genode stuff would go faster if rebuilding llvm didn't take so long
<{^_^}> [nixpkgs] @peti pushed 19 commits to haskell-updates: https://git.io/Jvb3f
<{^_^}> [nixpkgs] @Ma27 pushed to master « grocy: 2.6.1 -> 2.6.2 »: https://git.io/Jvb3J
<{^_^}> [nixpkgs] @mmahut opened pull request #84208 → openttd: 1.9.3 -> 1.10.0 → https://git.io/Jvb3k
<yorick> > error: the expression selected by the selection path 'asdf.config.system.build.toplevel' should be a set but is a boolean
<{^_^}> error: syntax error, unexpected $undefined, expecting ')', at (string):293:54
<yorick> it works the second time :|
* yorick disables gc
<{^_^}> [nixpkgs] @mmahut closed pull request #84208 → openttd: 1.9.3 -> 1.10.0 → https://git.io/Jvb3k
<yorick> this is on the nix stable
<{^_^}> [nixpkgs] @peti pushed 18 commits to haskell-updates: https://git.io/Jvb3O
xkapastel has quit [Quit: Connection closed for inactivity]
<{^_^}> [nixpkgs] @peti merged pull request #83963 → Update Haskell package set to LTS 15.6 (plus other fixes) → https://git.io/JvdO5
<{^_^}> [nixpkgs] @peti pushed 20 commits to master: https://git.io/Jvb3c
<{^_^}> [nixpkgs] @Ma27 pushed to release-20.03 « grocy: 2.6.1 -> 2.6.2 »: https://git.io/Jvb3B
fendor has joined #nixos
leotaku has quit [Ping timeout: 240 seconds]
<{^_^}> [nixpkgs] @FRidh merged pull request #74942 → itstool: fix double-shebang issue on macOS → https://git.io/JeDLW
<{^_^}> [nixpkgs] @FRidh pushed 2 commits to staging: https://git.io/Jvb3E
iceypoi has joined #nixos
<{^_^}> [nixpkgs] @jonringer merged pull request #83655 → lutris: custom wine installations require libkrb5. Backport #83645 → https://git.io/JvQ9V
<{^_^}> [nixpkgs] @jonringer pushed commit from @RohanHart to release-20.03 « lutris: custom wine installations require libkrb5 »: https://git.io/Jvb3u
ddellaco1 has joined #nixos
iceypoi has quit [Client Quit]
<{^_^}> [nixpkgs] @FRidh merged pull request #83973 → graphite2: 1.3.13 -> 1.3.14 → https://git.io/JvdGC
<{^_^}> [nixpkgs] @FRidh pushed commit from @r-ryantm to staging « graphite2: 1.3.13 -> 1.3.14 »: https://git.io/Jvb3a
leotaku has joined #nixos
captjakk has quit [Remote host closed the connection]
<scoates> is the go/no-go over?
iceypoi has joined #nixos
ddellacosta has quit [Ping timeout: 260 seconds]
<cole-h> Yep
<{^_^}> [nixpkgs] @FRidh merged pull request #83877 → maestral: 0.6.1 -> 0.6.3 → https://git.io/Jv5rx
<{^_^}> [nixpkgs] @FRidh pushed 2 commits to staging: https://git.io/Jvb3r
captjakk has joined #nixos
<{^_^}> [nixpkgs] @peti pushed 3 commits to haskell-updates: https://git.io/Jvb3K
<cransom> have we went?
<gchristensen> no-go
<{^_^}> [nixpkgs] @FRidh merged pull request #82885 → help2man: 1.47.12 -> 1.47.13 → https://git.io/Jv131
<{^_^}> [nixpkgs] @FRidh pushed commit from @r-ryantm to staging « help2man: 1.47.12 -> 1.47.13 »: https://git.io/Jvb36
<cransom> yeah, i agree on that dhcp thing. that bit me.
<{^_^}> [nixpkgs] @FRidh merged pull request #83139 → automake: 1.16.1 -> 1.16.2 → https://git.io/Jvb3i
<{^_^}> [nixpkgs] @FRidh pushed commit from @lsix to staging « automake: 1.16.1 -> 1.16.2 »: https://git.io/Jvb3P
palo has quit [Quit: WeeChat 2.7.1]
captjakk has quit [Ping timeout: 265 seconds]
ddellacosta has joined #nixos
ddellaco1 has quit [Ping timeout: 265 seconds]
ottidmes has joined #nixos
<{^_^}> [nixpkgs] @FRidh closed pull request #74102 → alsaLib: 1.1.9 -> 1.2.1.1 → https://git.io/JePGm
justanotheruser has quit [Ping timeout: 258 seconds]
cybrian has joined #nixos
captjakk has joined #nixos
<cole-h> Anybody feeling bored and want to review https://github.com/NixOS/nixpkgs/pull/84200? It has double 00 at the end, so you know it's a good one.
<{^_^}> #84200 (by cole-h, 2 hours ago, open): zoxide: 0.3.0 -> 0.3.1
FRidh has quit [Quit: Konversation terminated!]
palo has joined #nixos
<worldofpeace> cole-h: changelogs?
chloekek has joined #nixos
cybrian has quit [Ping timeout: 265 seconds]
<worldofpeace> it's helpful to reviewers to post any changelogs/release notes
<worldofpeace> so it's known what has changed
<cole-h> Roger, I did that last time but forgot for this one.
morgrimm has joined #nixos
<gchristensen> FYI: looks like there is some phishing going for github users, saying some private file was modified withe curlx. don't click it :P
<morgrimm> Does anyone have any really good examples of config repos with home-manager? I'm trying to wrap my head around how it needs to be configured (if it needs everything in nix configs, or if it can symlink existing dotfiles, or what)
<worldofpeace> cole-h: so from looking at those logs, as a qa person, I'd verify "Version output displays `git` revision information."
o1lo01ol1o has joined #nixos
mexisme has joined #nixos
fuzzy-id has quit [Ping timeout: 260 seconds]
<{^_^}> [nixpkgs] @prusnak opened pull request #84209 → freeciv: 2.6.0 -> 2.6.2 → https://git.io/Jvb3j
<cole-h> It relies on git, so it doesn't -- displays as `zoxide v0.3.1-unknown`
<{^_^}> [nixpkgs] @prusnak opened pull request #84210 → armagetronad: 0.2.8.3.4 -> 0.2.8.3.5 → https://git.io/Jvbse
<worldofpeace> it seems also you made the commit, so I trust it much more 😃
<cole-h> :P
<worldofpeace> will merge when ci passes
<cole-h> There's no easy way to get the actual commit info stuff to display without keeping .git
<worldofpeace> you could keep it by fetching it different. but I don't really think people should do that
<cole-h> Could go quick-and-dirty and just use the commit hash (after `leaveDotGit = true;`), but it would still differ from `git describe --tags`'s output.
<cole-h> Right, that's why I didn't :P
<cole-h> My local overlay displays the hash, but that's because I actually have the source on-disk (and `lib.commitIdFromGitRepo` it)
iceypoi has quit [Quit: WeeChat 2.3]
iceypoi has joined #nixos
mexisme has quit [Ping timeout: 260 seconds]
domogled has joined #nixos
<cole-h> worldofpeace++ Thanks for keeping me honest -- been trying to get better about having commit messages that explain the change (even if it's just a link to what's changed)
<{^_^}> worldofpeace's karma got increased to 87
<gchristensen> <3 cole-h worldofpeace++
<{^_^}> cole-h's karma got increased to 12, worldofpeace's karma got increased to 88
iceypoi has quit [Client Quit]
<gchristensen> mog: are you planning on doing some 20.03 artwork? :)
<cole-h> I think they already did
<gchristensen> oh!
<gchristensen> oh!
<worldofpeace> cole-h: as a merger, we can integrate so much faster if we have that information, and be more sure of any side-effect or if it's backportable.
<worldofpeace> gchristensen: yep, asked for that in advance
<gchristensen> isn't that a beaut'
<cole-h> 👍 That's why I want to get better at it -- make your (collective) jobs easier
iceypoi has joined #nixos
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/ed3ec0be54e (from 5 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
<worldofpeace> though I don't know where or how these images come to be https://github.com/NixOS/nixos-homepage/blob/master/logo/nixos-logo-19.09-loris-lores.png
<cole-h> gchristensen: Got a sec to see why https://github.com/NixOS/nixpkgs/pull/84183 is purple-flagged?
<{^_^}> #84183 (by doronbehar, 8 hours ago, open): uhd: 3.14.0.0 -> 3.15.0.0
<gchristensen> cole-h: I can't look back 9h, let's see if it does it again :)
<cole-h> :P
<worldofpeace> but it looks smaller/compressed
<cole-h> Fingers crossed
<cole-h> worldofpeace: maybe just a `convert -resize sizexsize img`?
<cole-h> (convert coming from imagemagick, of course0
fuzzy-id has joined #nixos
<worldofpeace> cole-h: probably. maybe it's documented somewhere what size
xkapastel has joined #nixos
<bqv[m]> oh hey, unstable small advanced
<cole-h> Or maybe it was a %, since that image is 100px x 93px (according to FF)
<samueldr> worldofpeace: literally just resizing the image from what mog shared :)
<monokrome> 19.09 is out? :o
<samueldr> monokrome: since last year
<samueldr> :3
<cole-h> 19.09 been out since 09/2019 :P
<monokrome> oh lol
<bqv[m]> lmao
<monokrome> thought it was the new one
<monokrome> oops
<monokrome> anyone recommend VPS that work well w/ NixOS btw?
<samueldr> worldofpeace: IIRC it's 100px wide, whatever in height this gives
<worldofpeace> samueldr: cool, will add that to my notes
<cole-h> Heard good things about Hetzner
<cole-h> But they're European-based IIRC
<monokrome> thnx =^.^=
<monokrome> that's ok w me
<monokrome> oh there's a whole wiki on it!
<samueldr> worldofpeace: confirmed, all 100px wide
<samueldr> monokrome: let me add that this is not an exhaustive list, but hosts the community had success with :)
<ottidmes> monokrome: Although I moved to some cheap black friday deal VPS in germany for mine, I have had a good experience at TransIP before, although you do have to bootstrap NixOS it worked perfectly
<samueldr> so there may be other hosts where it's trivial
<samueldr> monokrome: I personally use vultr to run {`-`} (the logging bot) on
<samueldr> the nixos.wiki coincidentally also uses a vultr vps
<worldofpeace> samueldr: thanks so much. it's hard to figure all these little pieces.
cosimone has joined #nixos
<cole-h> worldofpeace++ With you being so exhaustive (meant in a good way), hopefully this will be clearer for whoever you pass the RM torch to :)
<{^_^}> worldofpeace's karma got increased to 89
nobrak` has quit [Remote host closed the connection]
fuzzy-id has quit [Remote host closed the connection]
fuzzy-id has joined #nixos
nobrak has joined #nixos
<ottidmes> I am now extra careful with any host allowing custom ISOs, my current one claimed to support it, but it turned out you could request an ISO to be added to the global list, and I do not want my custom build NixOS installerd to be added to the global list... so I still had to bootstrap
<{^_^}> [nixpkgs] @worldofpeace merged pull request #84200 → zoxide: 0.3.0 -> 0.3.1 → https://git.io/Jvbt4
<{^_^}> [nixpkgs] @worldofpeace pushed 2 commits to master: https://git.io/Jvbs6
<worldofpeace> cole-h: I have lots of notes whenever I have to do something. So, if my brain permits me to write docbook, or I can find someone, it could be added to the manual.
<cole-h> You could also wait for that one RFC to allow us to switch to something not-XML-based ;)
<adisbladis> I've had great success using kexec on many hosting providers
cransom has quit [Quit: WeeChat 2.6]
fusion809 has quit [Remote host closed the connection]
pbogdan has quit [Remote host closed the connection]
fuzzy-id has quit [Ping timeout: 265 seconds]
cransom has joined #nixos
<ottidmes> yep, kexec worked for me on all hosts I had, so I don't worry about running NixOS anymore on any VPS (the KVM ones at least)
<bqv[m]> i think i used something i bastardised from nixos-infect
pbogdan has joined #nixos
kqb has joined #nixos
<{^_^}> [nixpkgs] @LnL7 merged pull request #80890 → darwin: fix NIX_LDFLAGS usages → https://git.io/Jv014
<{^_^}> [nixpkgs] @LnL7 pushed 3 commits to release-20.03: https://git.io/JvbGv
<bqv[m]> to be honest, if you're sufficiently flexible with your cli skills, you can always boot into a ramfs system and install whatever the McHeck you want. doubly easy if it has a small disk image around that you can just dd
<bqv[m]> that, or do some partition gymnastics
kleisli has joined #nixos
kleisli_ has quit [Ping timeout: 264 seconds]
<{^_^}> [nixpkgs] @andir opened pull request #84211 → firefox: 74.0 -> 74.0.1 → https://git.io/JvbGT
knupfer has quit [Quit: knupfer]
knupfer1 has joined #nixos
morgrimm has quit [Ping timeout: 240 seconds]
<bqv[m]> cole-h: so, shall i revert it to fetchurl?
kqb has quit [Quit: Leaving]
knupfer1 is now known as knupfer
<{^_^}> [nixpkgs] @rnhmjoj pushed 2 commits to release-20.03: https://git.io/JvbGc
<colemickens> Hi. I have `nix`, same version on two machines. But it seems that `--experimental-features` is only recognized on one? Am I doing something obviously wrong?
<bqv[m]> check your nix.conf
gxt has quit [Remote host closed the connection]
gxt has joined #nixos
lovesegfault has quit [Quit: WeeChat 2.8]
<colemickens> I don't see anything in user or system nix.conf that looks at all related
<colemickens> or any real differences at all
<ornxka> how long does chromium typically take to compile
<bqv[m]> traditionally, longer than you're willing to wait
<ornxka> i have 8 cores/16 threads
hmpffff has quit [Read error: Connection reset by peer]
<ornxka> i see
<gchristensen> any where from 20 minutes to 48 hours, depending on cores, memory, cpu speed, and storage type
<ornxka> i would like to see the machine that can compile chromium in 20 minutes
<bqv[m]> me too
<colemickens> ah, system-features = nixos-test, maybe thats it
<gchristensen> I was able to do it on one of these: https://www.packet.com/cloud/servers/c2-medium-epyc/ with a bit of tweaking
hmpffff has joined #nixos
<ornxka> this is not a slow machine and it has been several hours, even with my store on a pcie ssd
<ornxka> ahh
<colemickens> does the nixpkgs derivation still cut the cpu core in half :P
<{^_^}> [nixpkgs] @prusnak opened pull request #84212 → scummvm: add broken-sword-25 and dreamweb games → https://git.io/JvbGD
<gchristensen> I didn't write about it here https://grahamc.com/blog/an-epyc-nixos-build-farm because I couldn't make it work out of the box nicely
<peelz> Is there a difference between `{ a, ... }@args` and `args@{ a, ... }` ?
<gchristensen> the first one is preferred and the second one makes me go "wat"
<peelz> lol
<colemickens> bqv: it turns out I was looking at my two machines that don't require it. on the machine that does, its running nix unstable.
<peelz> gchristensen: is this syntax inspired by another language? not too familiar with functional programming
<ornxka> the last compiler warning said it was on file eighteen thousand out of about thirty-seven thousand, and that was about 45 minutes ago, so hopefully it will finish within the hour
<gchristensen> peelz: I dunno, actually!
<bqv[m]> wait what, really?
<bqv[m]> gchristensen: the second is the order in haskell
<bqv[m]> why does it confuse you
<bqv[m]> Cole Mickens: heh
<gchristensen> bqv[m]: well, I don't know haskell
<bqv[m]> there's your problem :p
<peelz> personally I prefer `{}@args`
<gchristensen> {}@args is the prevailing style in nixpkgs
<peelz> ah gotcha
<bqv[m]> as a haskeller, {}@args seems extremely inpleasant, but i guess i'll switch to using that then
<{^_^}> [nixpkgs] @teto opened pull request #84213 → vimPlugins.nvim-markdown-preview:init at2019-01-29 → https://git.io/JvbGb
<colemickens> guess I have no choice but to use nixUnstable now of course
<peelz> although `args@{}` is the syntax show in https://nixos.org/nixos/nix-pills/functions-and-imports.html
gxt has quit [Remote host closed the connection]
<cole-h> bqv[m]: Up to you. If it's still a tar.gz, I'd go a step further and `fetchTarball` it... idk what jonringer would have to say about that
gxt has joined #nixos
gustavderdrache has quit [Quit: Leaving.]
<niksnut> gchristensen: if you think that's weird: in C, x[5] and 5[x] are the same
<gchristensen> lol
<gchristensen> what a nightmare
<niksnut> it's only logical
<cole-h> Wait what
<gchristensen> haha, oh now I see it :)
<bqv[m]> oh yeah, that old trick
<peelz> [] is an operator so you can swap the operands lol
domogled has quit [Ping timeout: 260 seconds]
<bqv[m]> almost as unhelpful as the "goes to" operator
<peelz> don't remind me lol
<niksnut> a[b] is just sugar for *(a + b) and addition is commutative
<cole-h> I understand now
<cole-h> Still whacky to look at
gxt has quit [Remote host closed the connection]
<cole-h> I'd probably freeze up if I ever read that in some piece code
<gchristensen> it turns out C *is* how the computer works
gxt has joined #nixos
shibboleth has quit [Quit: shibboleth]
<ornxka> niksnut: technically the latter will give you compiler warnings due to mismatched types (i think? i dont normally do things the latter way)
<ornxka> but types in c are just suggestions most of the time anyway
lsix has quit [Ping timeout: 246 seconds]
civodul has quit [Quit: ERC (IRC client for Emacs 26.3)]
<gchristensen> :P
<cole-h> Steve!
<ornxka> i think of it more like "the computer is how C works" rather than the other way around
<bqv[m]> someone tell me the good news, do fetchurl and fetchTarball have the same hashes for the same url?
<ornxka> in C there is undefined behavior etc but there is not really any such thing in real life, the holes gets defined by the computer itself
smatting has joined #nixos
<peelz> bqv[m]: no, fetchTarball unpacks it first
<ornxka> so in the end everything is defined, just not necessarily in an intentional (or known at all) way
<bqv[m]> peelz: suffering
<peelz> bqv[m]: you can get the new hash with `nix-prefetch-url --unpack`
<bqv[m]> i'm sure i can, it's just a pain to when editing derivations
<peelz> true
<cole-h> Just `sha256 = stdenv.lib.fakeSha256;` and try to build :)
<bqv[m]> stdenv.lib.fakeSha256? christ i've been typing random numbers of zeroes all this time
<cole-h> I still do ;)
<gchristensen> I just ,tofu in here and then copy/paste each time
domogled has joined #nixos
<cole-h> ,tofu
<{^_^}> To get a sha256 hash of a new source, you can use the Trust On First Use model: use probably-wrong hash (for example: 0000000000000000000000000000000000000000000000000000) then replace it with the correct hash Nix expected. See: tofu-vim
<cole-h> You seriously copy-paste those 52 zeroes? haha
<gchristensen> yep.
gxt has quit [Remote host closed the connection]
<peelz> tofu-vim?
<bqv[m]> oh, fetchTarball is a builtin. cole-h: why fetchTarball over fetchurl?
<gchristensen> ,tofu-vim
<{^_^}> <esc>52i0<esc>
<peelz> lmao
<bqv[m]> isn't fetchTarball the one that doesn't require a hash
<bqv[m]> ahahah
<cole-h> ¯\_(ツ)_/¯
gxt has joined #nixos
<cole-h> Just what I'd do. It's your package, so do what you feel right
<bqv[m]> i just find a hash already in the file and ver0
<bqv[m]> hm, ok
<{^_^}> [nixpkgs] @edolstra opened pull request #84214 → Update nixUnstable + nixFlakes, remove nix1 → https://git.io/JvbZa
<cole-h> (I would still use fetchgit/etc but jonringer's point is valid)
<bqv[m]> ooh, i've been waiting for that ^
gxt has quit [Remote host closed the connection]
gxt has joined #nixos
<{^_^}> [nixpkgs] @bhipple merged pull request #84124 → pythonPackages.osqp: remove mkl dependency → https://git.io/JvFCT
<{^_^}> [nixpkgs] @bhipple pushed 2 commits to master: https://git.io/JvbZK
<{^_^}> [nixpkgs] @bhipple merged pull request #84120 → pythonPackages.cvxpy: add explicit numpy/scipy dependencies → https://git.io/JvFZK
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/9242acfbbc6 (from 2 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
<{^_^}> [nixpkgs] @bhipple pushed 2 commits to master: https://git.io/JvbZX
c0c0 has quit [Quit: WeeChat 2.6]
c0c0 has joined #nixos
gxt has quit [Remote host closed the connection]
gxt has joined #nixos
lsix has joined #nixos
lsix has quit [Client Quit]
smatting has quit [Ping timeout: 250 seconds]
knupfer has quit [Ping timeout: 260 seconds]
<peelz> bqv[m]: I would use fetchgit too. It just looks look cleaner than an interpolated URL. Personally, I default to git unless the repo is massive.
<bqv[m]> don't do this to me guys
<bqv[m]> i literally just pushed changing it back
<peelz> LOL I'm sorry, just giving my opinion. I'm sure it's fine either way
lsix1 has quit [Ping timeout: 260 seconds]
<{^_^}> [nixpkgs] @bhipple merged pull request #84197 → pythonPackages.pint: 0.9 -> 0.11 → https://git.io/JvbIE
<{^_^}> [nixpkgs] @bhipple pushed 3 commits to master: https://git.io/JvbZ9
ottidmes has quit [Quit: WeeChat 2.7.1]
thetet has joined #nixos
<{^_^}> [nixpkgs] @manveru closed pull request #81442 → bundler: 1.17.3 -> 2.1.4 → https://git.io/Jv2fC
maddo has quit [Quit: See ya]
<{^_^}> [nixpkgs] @manveru reopened pull request #81442 → bundler: 1.17.3 -> 2.1.4 → https://git.io/Jv2fC
<{^_^}> [nixpkgs] @Infinisil merged pull request #84075 → minecraft-server: build against jre_headless → https://git.io/JvdNC
<{^_^}> [nixpkgs] @Infinisil pushed 2 commits to master: https://git.io/JvbnL
ottidmes has joined #nixos
<ottidmes> What is going on, if I run any application through rofi desktop launcher I get the error: No GSettings schemas are installed on the system
<ottidmes> When I run the application directly from the command line, it works without issue
<ottidmes> Seems like an environment variable issue, but not sure what is the cause
<ottidmes> I am running latest nixos-20.03 with a few latest nixos-unstable packages (not used when testing this)
chloekek has quit [Ping timeout: 260 seconds]
<ottidmes> I am using i3 as my window manager, and thought maybe something changed causing the breakage, but even when I enable xfce with no window manager and desktop, it still fails with the same error, so it does not seem like I am missing any fundamental desktop manager related
<{^_^}> [nixpkgs] @worldofpeace opened pull request #84215 → Pantheon update 2020-04-03 → https://git.io/JvbnB
<peelz> ottidmes: was it working previously?
<ottidmes> peelz: yes
<ottidmes> I know vlc worked through rofi a few weeks ago when I was already on 20.03
<ottidmes> and now it crashes
<peelz> Uhh I have no idea. I'm still on NixOS 19.09
mallox has quit [Quit: WeeChat 2.7.1]
<cole-h> Ping worldofpeace: I think you did something with gsettings stuff recently? Maybe you can help ottidmes ^
<worldofpeace> I am one of the resident experts on gsettings
<ottidmes> It is in the manual: https://nixos.org/nixpkgs/manual/#ssec-gnome-common-issues but surely vlc is already fixed?
<energizer> ottidmes: do you have dconf installed?
<ottidmes> energizer: I do
<peelz> ottidmes: https://termbin.com/t3c6 you have both of these lines in your config?
<worldofpeace> ottidmes: Can you try gnome-calculator for me? I know this is wrapped
<ottidmes> worldofpeace: will do, all the apps that crash are wrapped, but guess thats to be expected, your point is about it being fixed, right?
mexisme has joined #nixos
<worldofpeace> ottidmes: ohh. I can try it by myself by doing rofi -show drun right?
<ottidmes> worldofpeace: yes,
<ottidmes> worldofpeace: I tried just now, gnome-calculator works, although it ran with run/rather than drun, but then again trying run with vlc through rofi fails too
lovesegfault has joined #nixos
<worldofpeace> ottidmes: It all works for me. Are you using gnomeplatform for qt gtk styling?
<worldofpeace> note, I've ran rofi like `XDG_DATA_DIRS=/run/current-system/sw/share ./result/bin/rofi -show drun`. That is because my environment will be polluted by being launched with a program that is wrapped
<ottidmes> worldofpeace: only gnome-calculator works for me. I was running no desktop manager before, just i3 with some fixes (which I called desktop lite) that has worked for me for ages, I enabled xfce to be sure, so should those styling things work for xfce too?
<ottidmes> worldofpeace: XDG_DATA_DIRS=/run/current-system/sw/share rofi -show drun fails for me too for vlc
<ottidmes> worldofpeace: Ah, maybe that's it, I have switched to gnomeplatform!
relaxmax has quit [Ping timeout: 265 seconds]
<worldofpeace> ottidmes: I was about to tell you. vlc is a pure qt app (I think)
<ottidmes> worldofpeace: yep, must be that, I am reverting: { qt5 = { enable = true; platformTheme = "gnome"; style = cfg.qt5Theme; }; } to my old config, hopefully that fixes things
<worldofpeace> ottidmes: the one where vlc crashes is a bug with qgnomeplatform. I've fixed that with a C++ patch here on master https://github.com/NixOS/nixpkgs/pull/83245/files#diff-fd9a2ea9a555ed9e0080a899d3c9c2ee
<ottidmes> worldofpeace: many apps seem to, mumble just crashed, pavucontrol-qt does too
<{^_^}> [nixpkgs] @drewrisinger opened pull request #84216 → [20.03] python3Packages.pint: 0.9 -> 0.11 → https://git.io/Jvbnp
<worldofpeace> ottidmes: vlc is just an example. it's any app that uses qt and then uses the qgnomeplatform theme. It depends on gsettings-desktop-schemas, but you can't wrap .so
<worldofpeace> *platformtheme
ixxie has quit [Ping timeout: 265 seconds]
<worldofpeace> should be in nixos-unstable in a few days
jumper149 has quit [Quit: WeeChat 2.7.1]
smatting has joined #nixos
ixxie has joined #nixos
<ottidmes> worldofpeace: cool, thanks a lot for the help, I have had too many changes recently so it was not easy to pinpoint the cause
<ottidmes> worldofpeace++
<{^_^}> worldofpeace's karma got increased to 90
<ottidmes> Its working again :D
Naptra has quit [Remote host closed the connection]
<worldofpeace> ottidmes: you're welcome. You were lucky in that there's already a committed fix too
<{^_^}> Channel nixos-20.03-small advanced to https://github.com/NixOS/nixpkgs/commit/449c060d125 (from 2 hours ago, history: https://channels.nix.gsc.io/nixos-20.03-small)
<{^_^}> [nixpkgs] @ggreif opened pull request #84217 → ghcHEAD: bump to 8.11.20200403 → https://git.io/JvbcU
<{^_^}> [nixos-homepage] @garbas pushed to redirect-old-urls « redirect old urls which are found by analytics »: https://git.io/JvbcT
<{^_^}> [nixos-homepage] @garbas opened pull request #380 → redirect old urls which are found by analytics → https://git.io/Jvbck
maiz has quit [Ping timeout: 272 seconds]
captn3m0 has quit [Read error: Connection reset by peer]
captn3m0 has joined #nixos
<ornxka> considering lobbying my congressman to reduce chromium build times, others welcome to join
<gchristensen> lol
<gchristensen> hopefully it'll work better than that bill to set pi to 3.2
<ornxka> haha
<MichaelRaskin> Well, amend ADA to penalise websites not accessible via a web browser with less than a million lines of code…
lostnet has joined #nixos
<cole-h> Ruh-roh. New purple label. gchristensen: https://github.com/NixOS/nixpkgs/pull/84216
<{^_^}> #84216 (by drewrisinger, 14 minutes ago, open): [20.03] python3Packages.pint: 0.9 -> 0.11
<gchristensen> great
selfsymmetric-mu has quit [Ping timeout: 260 seconds]
snicket has quit [Ping timeout: 240 seconds]
<gchristensen> I'm wishing I had log aggregation :)
nek0 has quit [Quit: ZNC 1.7.5 - https://znc.in]
<{^_^}> [nixpkgs] @worldofpeace merged pull request #83880 → clj-kondo: 2019.12.14 -> 2020.03.20 → https://git.io/Jv5iR
<{^_^}> [nixpkgs] @worldofpeace pushed 2 commits to master: https://git.io/Jvbc2
<cole-h> gchristensen: Is it OK to be listing the entire paths from the builders? Looking at a log, idk if "sensitive" information could ever be leaked as a result of my "add paths to error" change
<gchristensen> yep thats fine!
ixxie has quit [Ping timeout: 265 seconds]
<cole-h> OK, great./
nek0 has joined #nixos
__monty__ has quit [Quit: leaving]
nek0 has quit [Remote host closed the connection]
aranea has quit [Quit: aranea]
<bqv[m]> > (lib.fix lib.mergeAttrsWithFunc) { "a" = {}; } { b = "c"; }
<bqv[m]> why does this fail
<{^_^}> { a = <CODE>; b = <CODE>; }
<cole-h> > :p (lib.fix lib.mergeAttrsWithFunc) { "a" = {}; } { b = "c"; }
<{^_^}> value is a string while a set was expected, at /var/lib/nixbot/nixpkgs/master/repo/lib/deprecated.nix:183:28
<bqv[m]> oh, what
<bqv[m]> it doesn't
<bqv[m]> but it does in repl
<bqv[m]> ah
<bqv[m]> yeah, why does that fail, is there something about mergeAttrs i'm not seeing
nek0 has joined #nixos
thetet has left #nixos [#nixos]
<bqv[m]> it's almost like it's trying to form a type for the toplevel attrsets. that or it's actually comparing {} to "c" for whatever reason
erasmas has quit [Quit: leaving]
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<gchristensen> cole-h: github just can't keeep their stuff together.
<gchristensen> Error("missing field `created_at`", line:3, column: 1),
<cole-h> .....
<gchristensen> cole-h: can you extend this PR to include created_at? https://github.com/softprops/hubcaps/pull/249 and updated_at too I suppose while we're at it
<cole-h> Still feeling the after-effects of yesterday, huh
<{^_^}> softprops/hubcaps#249 (by grahamc, 1 day ago, open): Statuses.target_url: optional
work_ has joined #nixos
<gchristensen> and then PR that to grahamc/hubcaps' master
<gchristensen> and then updat eofborg's hubcaps :P
<gchristensen> in the cargo.toml
<{^_^}> Channel nixpkgs-19.09-darwin advanced to https://github.com/NixOS/nixpkgs/commit/0ee9ceff38f (from 29 hours ago, history: https://channels.nix.gsc.io/nixpkgs-19.09-darwin)
<cole-h> meaning created_at/updated_at -> Option<..>, right?
<gchristensen> yea
<cole-h> 👍
<gchristensen> thanks!
o1lo01ol1o has quit [Remote host closed the connection]
o1lo01ol1o has joined #nixos
fendor has quit [Quit: Leaving]
linarcx has quit [Ping timeout: 256 seconds]
m0rphism has quit [Ping timeout: 256 seconds]
alp has quit [Quit: ZNC 1.7.2 - https://znc.in]
o1lo01ol1o has quit [Ping timeout: 260 seconds]
<bqv[m]> ok nevermind, found something that works. "lib.fix lib.mergeAttrsWithFunc" -> "lib.fold lib.recursiveUpdate {}"
hmpffff has quit [Quit: nchrrrr…]
<bqv[m]> uhh those have slightly different signatures. ignore the fold part
<bqv[m]> recursiveUpdate did what i wanted
hmpffff has joined #nixos
hmpffff has quit [Client Quit]
hmpffff has joined #nixos
hmpffff has quit [Client Quit]
<{^_^}> [nixpkgs] @worldofpeace pushed to staging-next « gst_all_1.gst-plugins-bad: fix build with neon 0.31 »: https://git.io/JvbCW
gustavderdrache has joined #nixos
hmpffff has joined #nixos
hmpffff has quit [Client Quit]
hmpffff has joined #nixos
hmpffff has quit [Client Quit]
gentauro has quit [Read error: Connection reset by peer]
gentauro has joined #nixos
<{^_^}> [nixpkgs] @equirosa opened pull request #84218 → tut: init at 0.0.2 → https://git.io/JvbC4
cosimone has quit [Ping timeout: 260 seconds]
vykook has quit [Ping timeout: 265 seconds]
hmpffff has joined #nixos
hmpffff has quit [Client Quit]
<{^_^}> [nixpkgs] @kylesferrazza opened pull request #84219 → Allow jdiskreport to build on Darwin → https://git.io/JvbCR
hmpffff has joined #nixos
hmpffff has quit [Client Quit]
alp has joined #nixos
captn3m0 has quit [Ping timeout: 240 seconds]
captn3m0 has joined #nixos
o1lo01ol1o has joined #nixos
o1lo01ol1o has quit [Ping timeout: 265 seconds]
lostnet has quit [Remote host closed the connection]
noonien has joined #nixos
<{^_^}> [nixpkgs] @bhipple merged pull request #84216 → [20.03] python3Packages.pint: 0.9 -> 0.11 → https://git.io/Jvbnp
<{^_^}> [nixpkgs] @bhipple pushed 3 commits to release-20.03: https://git.io/JvbC6
<{^_^}> [nixpkgs] @bhipple merged pull request #84184 → [20.03] freeoffice: 973 -> 976 → https://git.io/JvFxR
<{^_^}> [nixpkgs] @bhipple pushed 3 commits to release-20.03: https://git.io/JvbCi
ZaraChimera has joined #nixos
<{^_^}> [nixpkgs] @worldofpeace pushed 3 commits to staging-next: https://git.io/JvbCQ
kiloreux has quit [Remote host closed the connection]
kiloreux_ has quit [Remote host closed the connection]
<immae> Hello, I’m a bit surprised by how much the apache configuration (nixos/modules/services/web-servers/apache-httpd/default.nix) has become invasive in latest channel, like forcing in every virtual host the DocumentRoot directory to be "Require all granted" for instance (which seems like a non-wanted enforced option when we want non-public vhosts), did I miss some option?
<{^_^}> [nixpkgs] @svalaskevicius opened pull request #84220 → hack: invoke wrapper hook manually → https://git.io/JvbCd
mbrgm_ has joined #nixos
<hpfr[m]> latest nixos unstable fetches a nonexistant path from repo.steampowered.com
mbrgm has quit [Ping timeout: 265 seconds]
mbrgm_ is now known as mbrgm
vykook has joined #nixos
<gchristensen> immae: maybe open an issue and ping @aanderse?
<immae> gchristensen: I just found the PR and poked him there :)
mexisme has quit [Ping timeout: 260 seconds]
<immae> do you think it’s better to open an issue directly?
mexisme_ has joined #nixos
InvisibleRasta has joined #nixos
justanotheruser has joined #nixos
<gchristensen> not sure :)
<immae> I’ll wait a few days and poke via an issue in absence of response, then
xelxebar has quit [Ping timeout: 240 seconds]
<{^_^}> [nixpkgs] @flokli merged pull request #83601 → [20.03] buildRustCrate fixes backports → https://git.io/JvQnY
<{^_^}> [nixpkgs] @flokli pushed 12 commits to release-20.03: https://git.io/JvbWL
vykook has quit [Ping timeout: 265 seconds]
sigmundv_ has quit [Ping timeout: 240 seconds]
<immae> Ah actually it’s not recent, I just never noticed it before, and it’s overriden by any manually entered <Directory> directive coming after
vykook has joined #nixos
xelxebar has joined #nixos
philr has joined #nixos
<bqv[m]> switching to master has resulted in far more compilations than i expected...
<cole-h> Don't need to switch to master if you're always running master
<cole-h> *taps head*
<bqv[m]> but i would assume the next time i update i'll have the same amount of cache misses, no?
<bqv[m]> or is there just a chunk of commits that aren't allowed in unstable?
vykook has quit [Ping timeout: 240 seconds]
fragamus has joined #nixos
chagra_ has joined #nixos
erictapen has quit [Ping timeout: 260 seconds]
mexisme_ has quit [Ping timeout: 265 seconds]
xelxebar_ has quit [Remote host closed the connection]
erictapen has joined #nixos
vykook has joined #nixos