<notgne2>
Hey, I'm trying to build a system with a module cherry picked out of another nixpkgs source, this normally works but this time I'm getting `element link: validity error : IDREF attribute linkend references an unknown ID`, is there a way to disable the manual generation to get around this?
<notgne2>
this is also deployed with morph if that matters
<infinisil>
documentation.nixos.enable = false should be enough
<notgne2>
I think I tried that though, and it didn't work till I updated the unstable channel I pull the module and package from, and set man, info, and doc to disabled too on the documentation module
tobeportable has quit [Ping timeout: 272 seconds]
<dkjii>
abathur: on Gentoo, but using nix to build it
<dkjii>
abathur: I'm doing nix -iE
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/Jf97p
<dkjii>
and its installing the expression
<dkjii>
I'm compiling sharutil from the nixpkgs channel
<dkjii>
sharutils*
MmeQuignon has quit [Ping timeout: 272 seconds]
<dkjii>
it's an indirect dependency to nixos-install as far as i can see
Drakonis__ has quit [Quit: Leaving]
drakonis has joined #nixos
<MtotheM>
I can't figure out how to do the vim package override from the wiki. when i put it in my main nixOS config. it sais it doesn't exist.
<MtotheM>
yeah sorry, was more an expression than a question infinisil
<MtotheM>
so I've just setup a custom vimrc in my package declaration which works well. is this `<package>.customize` a pattern that works with all packages? and where can I find more information about it? let's say there is a program with multiple configuration files. how would it know?
<energizer>
colemickens: i'd appreciate a tip if you figure it out, hard to do zoom meetings atm
bqv has joined #nixos
lunatera has joined #nixos
lunaa has quit [Ping timeout: 256 seconds]
teej has quit [Ping timeout: 240 seconds]
jlpeters has quit [Ping timeout: 258 seconds]
betawaffle has quit [Ping timeout: 260 seconds]
bitonic has quit [Ping timeout: 265 seconds]
pingveno has quit [Read error: Connection reset by peer]
pingveno has joined #nixos
parsnip has quit [Ping timeout: 260 seconds]
MarkRBM has quit [Ping timeout: 272 seconds]
pkral has quit [Ping timeout: 240 seconds]
angerman has quit [Read error: Connection reset by peer]
emilazy has quit [Ping timeout: 246 seconds]
ProofTechnique has quit [Ping timeout: 265 seconds]
TheNumb has quit [Ping timeout: 260 seconds]
omnigoat has quit [Ping timeout: 256 seconds]
higherorder has quit [Ping timeout: 258 seconds]
angerman has joined #nixos
proofofkeags has quit [Remote host closed the connection]
wildsebastian has quit [Ping timeout: 240 seconds]
pingveno has quit [Max SendQ exceeded]
drakonis1 has joined #nixos
jbetz has quit [Ping timeout: 258 seconds]
emilazy has joined #nixos
NekomimiScience has quit [Ping timeout: 246 seconds]
georgyo has quit [Ping timeout: 240 seconds]
drakonis_ has quit [Read error: Connection reset by peer]
drakonis2 has joined #nixos
digitalgrease has quit [Ping timeout: 256 seconds]
claudiii_ has quit [Ping timeout: 246 seconds]
TheNumb has joined #nixos
MarkRBM has joined #nixos
h0m1 has quit [Ping timeout: 272 seconds]
drakonis1 has quit [Ping timeout: 260 seconds]
peel has quit [Ping timeout: 240 seconds]
h0m1 has joined #nixos
codygman has quit [Read error: Connection reset by peer]
betawaffle has joined #nixos
digitalgrease has joined #nixos
ProofTechnique has joined #nixos
georgyo has joined #nixos
NekomimiScience has joined #nixos
jlpeters has joined #nixos
pingveno has joined #nixos
codygman has joined #nixos
wildsebastian has joined #nixos
peel has joined #nixos
bitonic has joined #nixos
claudiii_ has joined #nixos
omnigoat has joined #nixos
angerman has quit [Ping timeout: 256 seconds]
higherorder has joined #nixos
pkral has joined #nixos
teej has joined #nixos
jbetz has joined #nixos
angerman has joined #nixos
parsnip has joined #nixos
bqv has quit [Ping timeout: 260 seconds]
<MtotheM>
There is syntax typos in almost all wiki pages I've read so far, and it really confused me in the start. can anyone contribute to fixing the wiki?
<infinisil>
MtotheM: Yeah just log in and edit the page
<infinisil>
There's also a channel for the wiki: #nixos-wiki
<pjt_014>
MtotheM: Add if you're not sure what's correct be sure to put such things in the talk pages. There will be a visible record then
<pjt_014>
also if you want to try things out then happyfacebot is happy to help
<pjt_014>
> 1 + 3 == 5
<{^_^}>
false
<pjt_014>
good bot
<pjt_014>
> (1 + 3 == 5) or (7 - 3 == 4)
<{^_^}>
attempt to call something which is not a function but a boolean, at (string):316:1
<pjt_014>
hm
jumper149 has quit [Quit: WeeChat 2.8]
<pjt_014>
> (1 + 3 == 5) || (7 - 3 == 4)
<{^_^}>
true
<MtotheM>
Sounds good, I'll make a note for later.
codygman has quit [Ping timeout: 246 seconds]
codygman has joined #nixos
<MtotheM>
I struggle to find the full set of options for things, how would I go about doing that? let's say I want to configure 'time'. I don't know what options it actually let's me set.
<MtotheM>
oh thanks. so there was a page for the options after all.
<pjt_014>
yup. and in the man page you may wanna search it with /^ *time\.
<pjt_014>
since 'time' is a common string
<pjt_014>
regex is `fun` (air quotes optional)
<MtotheM>
yeah
<MtotheM>
what about package specific options? I copied the example for 'vim_configurable` but it doesn't seem to exist in the options database. so I assume it only covers the core nix system.
<{^_^}>
[nixpkgs] @cdepillabout pushed 2 commits to haskell-updates: https://git.io/Jf9d2
proofofkeags has joined #nixos
xelxebar has quit [Remote host closed the connection]
xelxebar has joined #nixos
<cole-h>
pjt_014: I'm pretty sure simpson meant that `man configuration.nix` is only for NixOS system configuration, which includes running various servers (like nginx and even minecraft). For nixpkgs documentation, you need to refer to the manual or to the repo itself.
proofofkeags has quit [Ping timeout: 260 seconds]
whatisRT has quit [Ping timeout: 260 seconds]
sbetrium_ has quit [Quit: sbetrium_]
<simpson>
cole-h: Yes, words.
sbetrium has joined #nixos
<{^_^}>
[mobile-nixos] @samueldr opened pull request #164 → mruby-os: Fix native ARM build → https://git.io/Jf9dF
sigmundv has quit [Ping timeout: 258 seconds]
sbetrium has quit [Ping timeout: 246 seconds]
apolitico has joined #nixos
apolitico has left #nixos [#nixos]
orivej has joined #nixos
orivej has quit [Ping timeout: 246 seconds]
orivej has joined #nixos
whatisRT has joined #nixos
lawr3nce has quit [Quit: Leaving]
felixfoertsch23 has joined #nixos
felixfoertsch has quit [Ping timeout: 246 seconds]
felixfoertsch23 is now known as felixfoertsch
whatisRT has quit [Remote host closed the connection]
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #nixos
drakonis2 has quit [Ping timeout: 264 seconds]
m0rphism has quit [Ping timeout: 256 seconds]
drakonis2 has joined #nixos
proofofkeags has joined #nixos
cole-h has quit [Ping timeout: 264 seconds]
proofofkeags has quit [Ping timeout: 256 seconds]
cole-h has joined #nixos
<energizer>
,locate bin/alsa
<{^_^}>
Found in packages: jack2, alsaUtils, zita-alsa-pcmi
<energizer>
afaict none of those packages actually have bin/alsa
noobly has joined #nixos
whatisRT has joined #nixos
<samueldr>
that doesn't search for a full binary name, but for anywhere in the path matching bin/alsa
<samueldr>
so e.g. bin/alsamixer matche
<samueldr>
,locate bin alsa
<{^_^}>
Found in packages: jack2, alsaUtils, zita-alsa-pcmi
<cole-h>
I don't think there is an "alsa" binary
<samueldr>
(this might also fail the same way)
<samueldr>
yeah, cole-h, that's what I was going to continue with
<{^_^}>
[nixpkgs] @marsam pushed 3 commits to release-20.03: https://git.io/Jf9NP
<energizer>
ok so now my wifi interface has disappeared
<energizer>
after rebooting
<noobly>
Hi, I'm trying to install nix on guix, but I'm getting the 'cannot connect to daemon' error, as described here: https://github.com/NixOS/nix/issues/2401. Any tips? I've tried running with --no-daemon but I get the same error. I'm trying to install by $ sh <(curl https://nixos.org/nix/install)
<{^_^}>
nix#2401 (by mumblez, 1 year ago, closed): Nix single user install broken - 2.1
<energizer>
lovesegfault: talyz: are you using impermanence on real computers yet?
<colemickens>
energizer: I also have that some times :)
<colemickens>
energizer: xps 9370 for me
orivej has quit [Ping timeout: 240 seconds]
orivej has joined #nixos
<energizer>
colemickens: just rebooting solved that
lunatera has quit [Ping timeout: 260 seconds]
<energizer>
9360 for me
<cole-h>
energizer: 99% sure both are using it (or at least some variation on it)
drakonis2 has quit [Read error: Connection reset by peer]
endformationage has quit [Ping timeout: 256 seconds]
runciter_ has joined #nixos
<CMCDragonkai>
So I'm using fetchurl on a 18 GiB zip file. Now its finished downloading. And it's spending a lot of time in `installPhase`. The install phase is taking much longer than I thought it would be.
<CMCDragonkai>
clever: perhaps my install phase should just be a symlink
<CMCDragonkai>
clever: to the src
<CMCDragonkai>
clever: what do you think?
<clever>
CMCDragonkai: that would be much better
kenran has joined #nixos
<CMCDragonkai>
clever: what's the function to symlink the entire src directory?
<CMCDragonkai>
clever: like a symlink tree thing
<cole-h>
Why not just `ln -s $src $out/src` or something?
<CMCDragonkai>
in this case the `$src` will be pointing to a directory that contains many files, i'd want to replicate that flat list of files as symlinks in the $out
<clever>
CMCDragonkai: buildEnv requires the input dir to already have the right name, which is what cole-h's thing does
<clever>
CMCDragonkai: fetchurl produces a single file, not a dir of many files
<CMCDragonkai>
ah yes
<CMCDragonkai>
but i recently changed to fetchzip just now
<CMCDragonkai>
for this particular sub-problem
<clever>
fetchzip already produces the dir you want
<clever>
just use it naked
<CMCDragonkai>
but this needs to be derivation with name/version... etc?
<CMCDragonkai>
like an actual package
<CMCDragonkai>
and perhaps meta
<clever>
CMCDragonkai: fetchzip is a derivation
<CMCDragonkai>
can i put `version` and `meta` into fetchzip?
<clever>
yeah
<CMCDragonkai>
clever: is this just a matter of overrideattrs or do you mean `fetchzip { version = ...; meta= ...; }`
<clever>
CMCDragonkai: just plain fetchzip { version=; meta=; ... }
<CMCDragonkai>
and `name`
<clever>
name is auto-generated from the url, but you can still set your own name
orivej has quit [Ping timeout: 260 seconds]
<CMCDragonkai>
clever: it says there's an unexpected argument 'version'
orivej has joined #nixos
<CMCDragonkai>
clever: error: anonymous function at /nix/store/07s871hz02vb128z40y8sigxp95dpg2d-source/pkgs/build-support/fetchurl/default.nix:38:1 called with unexpected argument 'version', at /nix/store/07s871hz02vb128z40y8sigxp95dpg2d-source/pkgs/build-support/fetchzip/default.nix:17:2
<clever>
does it really need a version field?
<CMCDragonkai>
not really
retrocswl has joined #nixos
Acilim has joined #nixos
Acilim has joined #nixos
palo1 has joined #nixos
palo has quit [Ping timeout: 256 seconds]
palo1 is now known as palo
orivej has quit [Ping timeout: 258 seconds]
orivej has joined #nixos
kleisli has joined #nixos
lord| has quit [Read error: Connection reset by peer]
user_0x58 has quit [Quit: Leaving]
agsdheidjd has joined #nixos
<energizer>
afaict my headphones don't work until i run `sudo alsactl init`. ok fair enough. now i want to run that automatically, so i go ` services.xserver.displayManager.sessionCommands = ''${pkgs.alsaUtils}/bin/alsactl init '';` but that didn't work ( no sound). what could be going wrong?
<cole-h>
Maybe not running as root?
<energizer>
lightdm does run as root
<cole-h>
Does it work if you run it manually again?
<energizer>
ya
<cole-h>
Assuming you're also running pulseaudio, it could be that pulseaudio hasn't started at that point
<cole-h>
Since it's socket activated by default IIRC
<energizer>
interesting idea
<cole-h>
You could try switching to a tty, `systemctl start --user pulseaudio` and then logging in and seeing if that works
<energizer>
on second thought maybe you were right about root
<energizer>
cuz that script seems to be running as user
butterthebuddha has quit [Ping timeout: 260 seconds]
mtn has joined #nixos
alp has joined #nixos
MmeQuignon has joined #nixos
<mtn>
Random q: is it possible to source a virtual environment automatically when running `nix-shell`?
<mtn>
(By adding a line to my `shell.nix` or something)
<cole-h>
You could try adding something to your `shellHook` (I think that's the name0
<cole-h>
)
<mtn>
Awesome, that worked
<mtn>
Thanks cole-h!
<energizer>
mtn: why bother with virtualenv when you already have it's big sister, nix-shell
<energizer>
?
bahamas has quit [Ping timeout: 264 seconds]
<mtn>
I haven't set up things with poetry yet (using an annoying outside package I just got working with pip and not sure about making the switch since I may not use it for long), and I'm working in a codebase with packages that are in pip but not nixpkgs
<mtn>
On the other hand, maybe getting poetry working will take less time than I'm anticipating
linarcx has joined #nixos
<energizer>
if you have unusual packages it can take a little while to figure out what overrides they need, otherwise it's pretty quick
<chagra>
I wonder if I can find a configuration.nix with pacman under systemPackages
<pjt_014>
so maybe adding that flag is enough?
<chagra>
there are 7 contributors to that packages
<chagra>
there has to be someone using it
<chagra>
somehow
lsix has joined #nixos
<pjt_014>
chagra: use the nix show-... cmd above on bluez and you can see more examples of /etc being used. I imagine beween this and samba one of those shows how pacman should do it
<qyliss>
What does Pacman need at /etc?
<chagra>
it seems it needs a pacman.conf file and a pacman.d directory with a mirrorlist file under it
<qyliss>
Right, so what you'll need is a NixOS module that uses environment.etc to put those things there
<chagra>
I'm still going through what I need then i can find out what must go under environment.etc
<pjt_014>
keep throwing things at it till the err messages stop complaining, that's what I'd do anyway
<pjt_014>
not eleagant but it works
<chagra>
error: failed to initialize alpm library
retrocswl has quit [Quit: Connection closed for inactivity]
<chagra>
(could not find or read directory: /var/lib/pacman/)
<chagra>
finally hit the error
<chagra>
hmmm what a curious package, it was updated this april
<chagra>
i need to recheck the package expression
<chagra>
I'm 100% sure someone has to be using this
Shados has joined #nixos
<{^_^}>
[patchelf] @domenkozar merged pull request #209 → Add libasan build option and test print-needed → https://git.io/JfS2R
<{^_^}>
[nixos-search] @garbas opened pull request #97 → Fix title of the app before it loads → https://git.io/JfHmQ
andreas303 has quit [Ping timeout: 240 seconds]
<chagra>
after setting up /etc/pacman.conf and /etc/pacman.d/mirrorlist
<chagra>
and mkdir /var/lib/pacman
<chagra>
pacman just works!
tobiasBora has joined #nixos
szicari has quit [Ping timeout: 256 seconds]
<chagra>
This is a game changer, I can finally have another package manager to fall back on when I need a rare package
andreas303 has joined #nixos
<evils>
chagra: that's how i got into nix...
<tobiasBora>
Hello! I recently updated my rasp that is running gitea. The update was successful, but there was a major issue at the end: gitea seems to have lost all the database information (so for example I can't connect anymore, there is no more public projects...). I can still see the repo in /var/lib/gitea/repositories, and I'll try to see if the database is just removed or still lives somewhere. If anobody
<tobiasBora>
had that issue before, please let me know ;)
<chagra>
evils: I've been using NixOS for a year and I've been too spoiled by it. I can't leave it
<chagra>
but sometimes if I want to run some rare program
<evils>
package it, share the love
<chagra>
and I see it has an AUR epression already written
<chagra>
I just get lazy
<chagra>
I always write package
<chagra>
packages*
<chagra>
but sometimes I just wanna run something instantly
<chagra>
it's just a quality of life thing
szicari has joined #nixos
lord| has joined #nixos
jgeerds has quit [Ping timeout: 246 seconds]
<jtojnar>
pacman in buildFHSUserEnv?
<energizer>
i wonder how many packages aur has that nixpkgs doesnt
<srk>
aur2nix :)
<chagra>
the way I managed to set it up, you can't run it in your current environement fully. I just got it set up enough for pacstrap to work so I can set up an arch-nspawn
<chagra>
srk: aur2nix is a thing?
alp has joined #nixos
<srk>
chagra: no, just invented it :)
<evils>
srk++
<{^_^}>
srk's karma got increased to 18
<chagra>
honestly if that was a thing everyone here would just become so lazy lol
<asheshambasta>
Is it possible to use Jack instead of Pulseaudio on NixOS? I've been looking into some music production on nix but it seems that Jack -> Pulseaudio is the suggested route, which seems to be hard to configure. But then again, I'm new to Jack and wonder how "lower level" Jack is compared to PA.
<asheshambasta>
Yeah, I was wondering if I could just use Jack as my primary sound-service.
kleisli has quit [Ping timeout: 265 seconds]
<srk>
my desktop now has three soundcards so pulseaudio is more confy for daily use as I can switch them via pacmd instead of switching cables :D
<asheshambasta>
srk: supercollider seems to require Jack (AFAICS)
<kenran>
Is there an equivalent to something like 'writeShellScriptBin' that does "give me those files here as a nix derivation" without having to use a full-blown mkDerivation?
<srk>
asheshambasta: ah, right :) that was one of the reasons I was running it I guess - with tidal
orivej has quit [Ping timeout: 258 seconds]
<srk>
tidal can do without sc but only samples
orivej has joined #nixos
<asheshambasta>
musnix looks promising
<asheshambasta>
srk: so in your setup, you have jack sitting on top of ALSA and then pa on top of Jack?
<asheshambasta>
that is, you have pa -> jack -> alsa for audio?
<srk>
asheshambasta: yes, that was the case with my previous setup, currently I only have pa but want to revive that
<srk>
I didn't fiddle with alsa, only used qjackctl but there was no jack module back then
<kenran>
,locate bin mv
<{^_^}>
Found in packages: toybox, busybox, coreutils, klibcShrunk, coreutils-full
<srk>
what sucks is that nvidia drivers are not compatible with -rt patched kernel so it's either gpu or rt sound.. gotta plugin one more gpu so I can pass the second one to vm with ugly drivers
<evils>
,locate kenran
<{^_^}>
Couldn't find in any packages
<evils>
,locate
<{^_^}>
Use ,locate <filename> to find packages containing such a file. Powered by nix-index (local installation recommended) https://github.com/bennofs/nix-index
<srk>
kenran: symlinkJoin
chloekek_ has joined #nixos
fendor has quit [Remote host closed the connection]
<kenran>
srk: I'd already need an existing nix derivation for that, right?
<srk>
kenran: yes, most things are derivations tho :)
zupo has joined #nixos
bahamas has quit [Quit: leaving]
<{^_^}>
[nixos-search] @garbas merged pull request #97 → Fix title of the app before it loads → https://git.io/JfHmQ
<{^_^}>
[nixos-search] @garbas pushed to master « Fix title of the app before it loads (#97) »: https://git.io/JfHsw
Chiliparrot has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
knupfer1 is now known as knupfer
romildo has joined #nixos
noudle has joined #nixos
<{^_^}>
[nixos-search] @garbas pushed to fix-92 « List channel as button group »: https://git.io/JfHGm
<{^_^}>
[nixos-search] @garbas pushed to fix-92 « List channel as button group »: https://git.io/JfHGY
<{^_^}>
[nixos-search] @garbas opened pull request #98 → List channel as button group → https://git.io/JfHGO
<{^_^}>
[nixpkgs] @martinbaillie opened pull request #90181 → terraform-ls: init at 0.3.2 → https://git.io/JfHGG
zarel has quit [Ping timeout: 256 seconds]
zarel has joined #nixos
markus[m]2 is now known as Foooooo[m]
m0rphism has joined #nixos
orivej has quit [Ping timeout: 256 seconds]
orivej has joined #nixos
<ramses_>
Is there a reliable way to get the path "lib/python3.7" when building a python packages? So that my derivation would continue to work if the python version changes?
<kenran>
srk: I'm not a native speaker and probably explained my problem badly :) inside some directory, there are (non-nix) files, mostly CSV. Now I need them in a derivation I'm writing (it's a docker image via buildImage). How can I get them "in there"?
chloekek_ has quit [Ping timeout: 246 seconds]
<{^_^}>
[nixos-search] @garbas merged pull request #98 → List channel as button group → https://git.io/JfHGO
<{^_^}>
[nixos-search] @garbas pushed to master « List channel as button group (#98) »: https://git.io/JfHGo
<kenran>
I thought about creating another derivation that would build nothing and in its `installPhase` copy those local files only to $out. But then I thought there must already be a utility function for such things.
<clever>
ramses_: its an attribute of the python package
<ramses_>
clever: that seems to be exactly what I need, thanks!
<srk>
kenran: ${./myFileToCopyToStore}
maxter has quit [Ping timeout: 264 seconds]
<Shados>
kenran: you could write such a function pretty trivially on top of pkgs.runCommand
Darkmatter66_ has quit [Read error: Connection reset by peer]
cosimone has quit [Remote host closed the connection]
<{^_^}>
[nix] @edolstra pushed to master « Add Store::readDerivation() convenience function »: https://git.io/JfHGF
cosimone has joined #nixos
<kenran>
srk: thanks, I'm currently trying that. Shados: there's a server which at runtime seems to need some files (it's not mine). Those files are "there", and in the nix expression that describes the docker container I need them to be "next to it" as well.
Darkmatter66 has joined #nixos
<notgne2>
I'm trying to package some software that has a lot of "install(...DESTINATION /..." in the cmake files, what is a good way to work around this in Nix? I'm honestly not very familiar with cmake at all either
MmeQuignon has quit [Ping timeout: 272 seconds]
<notgne2>
also, what's the right way to make a PR to un-broken a package? `swiften` is broken (and marked as such) right now, but downgrading boost and fixing the flags in the preInstall script seems to bring it back to life
<asheshambasta>
srk: it seems like to get tidal (+ sc) working, I will need to have Jack run on top of Alsa and then decide whether or not I want to keep pa around for more "normal" use audio.
<asheshambasta>
Or have Jack -> Pa working as in the nixos wiki.
<asheshambasta>
or something like musnix.
<srk>
combination :) musnix if you want realtime
orivej has quit [Ping timeout: 246 seconds]
<srk>
some apps have direct jack support like mpv, not sure about browsers
drakonis has joined #nixos
<notgne2>
actually, looking at nixpkgs now I see that swift-im is packaged (same source) and working with the newer boost, that's weird
<Joestar79>
I'm trying to override kubernetes-helm version but I'm getting an error. Can anyone take a look please and help me find the problem? https://pastebin.com/ZLH8hpE4
evanjs has quit [Read error: Connection reset by peer]
romildo has left #nixos ["Leaving"]
fendor has quit [Ping timeout: 258 seconds]
fendor has joined #nixos
knupfer has quit [Ping timeout: 260 seconds]
drakonis_ has joined #nixos
knupfer has joined #nixos
evanjs has joined #nixos
fendor_ has joined #nixos
Rusty1 has joined #nixos
klntsky has quit [Remote host closed the connection]
drakonis has quit [Ping timeout: 256 seconds]
fendor has quit [Ping timeout: 264 seconds]
klntsky has joined #nixos
<{^_^}>
[nixos-search] @garbas pushed to fix-94 « handle 10000 package limit in the UI correctly »: https://git.io/JfHZ5
<{^_^}>
[nixos-search] @garbas opened pull request #99 → handle 10000 package limit in the UI correctly → https://git.io/JfHZd
EsperLily has quit [Remote host closed the connection]
alp has quit [Ping timeout: 272 seconds]
<asheshambasta>
srk: I think given how painful audio seems to be in Linux, I'd probably want to go for the simplest setup.
<asheshambasta>
(so far, I've been using Ableton Live/Reaktor for music production on a Mac)
alp has joined #nixos
<asheshambasta>
also it seems like Jack has issues with suspend/hibernate etc
<{^_^}>
[nixpkgs] @romildo opened pull request #90183 → gdk-pixbuf-xlib: fix the pkg-config file → https://git.io/JfHn4
<srk>
asheshambasta: another option is to launch jack manually, disabling pulseaudio autospawn, then pulseaudio -k to stop it, launch jack via qjackctl, start pulse again so it uses jack
<{^_^}>
[nixos-search] @garbas pushed to fix-43 « make nix-build work »: https://git.io/JfHnV
<{^_^}>
[nixos-search] @garbas opened pull request #100 → make nix-build work → https://git.io/JfHno
<asheshambasta>
so far I've been trying to do this https://nixos.wiki/wiki/JACK but it seems to me that qjackctl has changed (I don't see the options listed in the screenshot of the page), and loading the jack modules in pa seems to fail with some connection issues logged.
<Orbstheorem>
Hi. My /boot partition is full and nixos-rebuild switch fails. I have already run nixos-collect-garbage --delete-older-than 1d and it's still full. How can I remove the profiles using those kernels and cleanup my /boot partition?
FRidh has quit [Quit: Konversation terminated!]
asheshambasta has joined #nixos
FRidh has joined #nixos
<asheshambasta>
I think the Jack wiki page is outdated, I'll see if I can update it.
<asheshambasta>
In my case, it is clear that pa cannot connect to jack, not sure why.
<infinisil>
Orbstheorem: I usually just run `sudo rm -rf /boot/kernels/*`, then I rebuild switch
knupfer has joined #nixos
knupfer has joined #nixos
orivej has quit [Ping timeout: 256 seconds]
orivej_ has joined #nixos
doomlist3 has quit [Ping timeout: 264 seconds]
iyzsong has joined #nixos
<infinisil>
Or `nix-collect-garbage` first (so older generations are removed), then `sudo rm -rf /boot/kernels/*` (so a boot switch can succeed), then `nixos-rebuild switch` (which does a boot switch, this time only putting the kernels and initrds from the generations that you didn't remove)
omneh has joined #nixos
noudle has quit []
OmnipotentEntity has quit [Ping timeout: 260 seconds]
<Orbstheorem>
infinisil: Thanks ^^ I wasn't sure just removing the old kerneles was safe x)
<{^_^}>
[nixos-search] @garbas merged pull request #100 → make nix-build work → https://git.io/JfHno
<{^_^}>
[nixos-search] @garbas pushed to master « make nix-build work (#100) »: https://git.io/JfHC3
orivej has quit [Quit: No Ping reply in 180 seconds.]
plutes has quit [Ping timeout: 272 seconds]
orivej has joined #nixos
<MtotheM>
oh. I tried looking here yesterday, but got no results back. the page is so large it breaks my browsers search function. thanks @Shados
orivej has quit [Ping timeout: 246 seconds]
<{^_^}>
[nixpkgs] @7c6f434c opened pull request #90186 → mplayer: fix build by forcing -lfreetype → https://git.io/JfHlx
szicari has quit [Ping timeout: 256 seconds]
<{^_^}>
[nixpkgs] @rnhmjoj merged pull request #89159 → nixos/networking: check interface state files exist before acting on them → https://git.io/JfKJc
sigmundv has quit [Read error: Connection reset by peer]
<{^_^}>
[nixops] @adisbladis closed pull request #1313 → Add a withPlugins function to the NixOps derivation → https://git.io/JfTFI
justanotheruser has joined #nixos
chloekek has quit [Quit: WeeChat 2.7]
szicari has joined #nixos
plutes has joined #nixos
<tobiasBora>
Hello, after upgrading to nixos 20.03, my gitea's database is empty. Someone on the gitea's discord suggested that the upgrade of postgresql may be the issue, especially because I don't have any release.stateVersion in my project (when I installed the raspberry, I don't think this field was set automatically and I was not aware of it's existence). What do I need to do to make sure I get my old DB
<evils>
tobiasBora: i'm not familiar with gitea, but can you roll back nixos to a point where it works? then back up, maybe restore from that if you can't get anything else work?
<asheshambasta>
So, I've been through hell trying to get pa to work with Jack, using Jack as the sink. And I failed. :-)
<srk>
:(((
<srk>
asheshambasta: pa jack sink/sources cause trouble?
<asheshambasta>
paudio cannot connect to jack even though systemctl reports jackd is running, jack cannot use the "alsa" driver (saving the prefs. makes jack constantly fall back to "dummy")
<asheshambasta>
yeah, and I cannot list the jack sources or sinks for that matter.
agsdheidjd has quit [Ping timeout: 256 seconds]
<tobiasBora>
evils: I can try, but I think that I was laking space and I did some garbage collection, so I'm not sure what's left or not. But I also would like to understand how stateEpoch works (only one occurence in the manual??). For example, if I add now a stateEpoch, can it save my database?
<srk>
asheshambasta: will test on 20.09 later, it was working without any issues for me
<asheshambasta>
Thanks, I will also try testing this on my laptop after work. So far I've wasted an 1hr of work on this. I wouldn't mind using Jack for all things, tbh, but it seems like a lot in Linux has been built with pa in mind; ah fragmentation.
<wrl>
i always run pa on top of jack
<wrl>
with jack always running underneath
<wrl>
doesn't work if you've got soundcards dynamically showing up or leaving (usb hotplug generally)
<wrl>
but otherwise rock solid
<asheshambasta>
wrl: if you can share your related configuration it might help me.
<wrl>
asheshambasta: i'm on arch now but i had things workin on nixos
<wrl>
let me thinks
<wrl>
think*
<wrl>
jfc
fendor has joined #nixos
<wrl>
i have pulse set to have no dynamic configuration, just pa jack source/sink
<asheshambasta>
and also worrying is that in qjackctl, if I tinker with the settings and attempt to set the driver to "alsa" it keeps falling back to "dummy" without errors/warnings.
iyzsong has joined #nixos
<wrl>
i was using jackd as a system service
<wrl>
on nixos
<asheshambasta>
same here, jackd is started and running.
iyzsong has quit [Client Quit]
<asheshambasta>
Anyway, this looks like hell this weekend.
<goibhniu1>
are you just trying to get jack and PA to play nicely together?
<asheshambasta>
yeah, route pa audio to/from jack
<goibhniu1>
hrm, I've been doing that for years without issue
<goibhniu1>
did you follow the guide on the wiki?
<srk>
with the nixos module?
<asheshambasta>
mind sharing your nix config? I did follow the guide on the wiki.
<tobiasBora>
Ok great, I managed to get gitea to work again with a proper system.stateVersion = "18.09";! Now, my question is: if I don't want to keep old softwares (like postgresql 9.6 in that case) is there some script that can automatically moves me from one stateVersion to another stateVersion?
duairc_ is now known as duairc
Jackneill has joined #nixos
proofofkeags has quit [Ping timeout: 246 seconds]
<srk>
hmm, actually there *is* a jack service
<srk>
but it doesn't have multi-user.wants so it needs to be started manually it seems
<srk>
asheshambasta: ^
<asheshambasta>
srk: systemd is reporting it to already be running on my machine on boot.
<{^_^}>
[nixpkgs] @AndersonTorres pushed 2 commits to master: https://git.io/JfHRL
dingenskirchen has joined #nixos
rardiol_ has joined #nixos
rardiol has quit [Read error: Connection reset by peer]
<asheshambasta>
so to get jack to work with pa I need to do the following: 1. ensure pa "alsa-sink" modules are unloaded 2. ensure pa "jack-sink/source" modules are loaded, 3. jackd is running and reachable (which also seems to be an issue but I need to check))
<asheshambasta>
is there an easy way to tinker with these things without having to repeatedly reboot and/or start new sessions?
doomlist3 has joined #nixos
siers is now known as ij
<asheshambasta>
all of pa/jack etc seem to be "hideously" stateful.
orivej has joined #nixos
<srk>
you don't have to reboot unless switching kernel
dingenskirchen has quit [Quit: dingenskirchen]
<srk>
interestingly if I run jack via service I don't get audio but everything looks fine
<{^_^}>
[nixops] @adisbladis merged pull request #1368 → Remove the dependency on <nixpkgs> for flakes → https://git.io/JfHZc
armin_ has quit [Quit: Great minds discuss ideas. Average minds discuss events. Small minds discuss people.]
amerigo has joined #nixos
ArdaXi has quit [Ping timeout: 246 seconds]
ArdaXi has joined #nixos
<asheshambasta>
okay, if I restart pulseaudio so far in the same session, none of my applications manage to reconnect to it.
<asheshambasta>
I believe this is because paudio sets environment variables etc. to be found
dingenskirchen has joined #nixos
<{^_^}>
[nix] @matthewbauer opened pull request #3689 → WIP: Substitutions from different store dirs → https://git.io/JfHRM
genevino has joined #nixos
meh` has quit [Ping timeout: 240 seconds]
meh` has joined #nixos
eoli3n_ has quit [Ping timeout: 264 seconds]
<mdlayher>
hey all! i'm going live at https://twitch.tv/mdlayher to work on a NixOS channel metadata Prometheus exporter in Go that could be used to power status.nixos.org! feel free to stop by and chat!
Chiliparrot has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
<evanjs>
So is there any type of limit to diskSize in the context of this function?
<evanjs>
Trying to create disks beyond ~65GB (not sure what size is the max, exactly) and it will stall on `cptofs -p ${optionalString (partitionTableType != "none") "-P ${rootPartition}"} -t ${fsType} -i $diskImage $root/* /`
leungbk has joined #nixos
<clever>
evanjs: does the image need to be that big initially? why not resizefs after deploying?
<simpson>
I'm not sure if I like this yet, but I think it has everything I need; it's got the generation, the time that the generation was built, the release and state versions, and I can infer things like age and freshness using time().
<evanjs>
clever: yeah lassulus was wondering the same thing. Trying to encode sensible defaults for virtual appliances we are going to provision
<gchristensen>
simpson: nice! maybe should make a proper module for that
<evanjs>
Figured that might be easier than including "how to resize to our recommended specs are importing the OVA"
<evanjs>
Well, for ESXi, at least...
<clever>
evanjs: you can look at how the aws images resize themselves on boot
<clever>
evanjs: and then use something like qemu-img to just change the length of the disk image, without touching the contents any
virus_dave has joined #nixos
virus_dave has quit [Client Quit]
drakonis1 has joined #nixos
<simpson>
gchristensen: Hey, IMO this is the most proper that this module has ever been~ But yeah, I would love to get something like this upstream. I want to spend a few weeks with these metrics first and see whether they're even effective.
virus_dave has joined #nixos
drakonis_ has quit [Read error: Connection reset by peer]
<evanjs>
clever: do you mean aws stuff under nixos/modules/virtualisation?
<gchristensen>
simpson: please do, mine seems to have some sort of problem where it stops being correct ... but I haven't looked in to it. I'll adopt yours and see if I can chase it down
virus_dave has joined #nixos
virus_dave has quit [Client Quit]
<clever>
evanjs: yeah
<evanjs>
either way, it does seem uncomfortable to have an implicit limit like that. bah
<jboy>
anybody know whether upgrading to nix>=2.2 can break things when coming from an earlier version (like 2.1.1)?
<simpson>
Yikes. I'm not sure why that would happen. Maybe the textfiles get jammed up somehow? I'll keep an eye out for that.
virus_dave has joined #nixos
virus_dave has quit [Client Quit]
<gchristensen>
I'm not totally sure
drakonis has quit [Ping timeout: 272 seconds]
szicari has quit [Ping timeout: 256 seconds]
<gchristensen>
simpson: I'm dreaming of a services.prometheus.exporters.nixos.enable = true; which enables the textfile exporter and writes out
<evanjs>
clever: lol you didn't literally mean just "autoResize = true;" did you? :P Errr that and "boot.growPartition", I guess?
<clever>
evanjs: maybe, i forget how trivial it is to use outside of nixos
<simpson>
gchristensen: Yeah! That would be nice, and that's definitely along the lines of what I'm intending to get here. I'm hoping that these metrics will still be useful even if machines can nixos-rebuild themselves automatically.
<gchristensen>
good idea
<simpson>
Also mdlayher++ for writing new exporters.
evanjs has quit [Read error: Connection reset by peer]
<{^_^}>
mdlayher's karma got increased to 0b10
evanjs has joined #nixos
genevino has quit [Quit: Great minds discuss ideas. Average minds discuss events. Small minds discuss people.]
leungbk has quit [Ping timeout: 258 seconds]
leah2 has quit [Ping timeout: 260 seconds]
szicari has joined #nixos
<evanjs>
clever: alright I'll play around with that a bit and see what I can find. I wonder how/if this could event be addressed in the context of cptofs. Like even something like nix's "doing something that needs more than X MB of memory... this may fail" or whatever it was
<evanjs>
just something that tells the user "idk if this is going to work" :P
<clever>
evanjs: you could also use cptofs with a smaller image, then manually expand it in a second derivation
magnetophon has quit [Read error: Connection reset by peer]
<evanjs>
Like I totally agree that images like that are sort of insane, it would just be nice to have some clarity on what's happening, especially if something might fail, etc
<{^_^}>
#59867 (by dfordivam, 1 year ago, open): VirtualBoxOVA does not work for more than 64gb baseImageSize
<evanjs>
clever: lassulus found it lol
<MtotheM>
Both vim and tmux have built in ways to declare the configuration. but for those that don't. is the only way to handle this with home manager?
<simpson>
Yes. home-manager was originally developed because NixOS didn't handle this sort of per-user per-program configuration well.
<evanjs>
MtotheM: it really depends. Sometimes you can do a bit with overrides, but home-manager is definitely the best way to go for dotfile config and etc imo
cosimone_ is now known as cosimone
<evanjs>
it can definitely seem intimidating at first, for sure
leungbk has quit [Ping timeout: 264 seconds]
<simpson>
(There are other tools and proposals, but home-manager's gotten quite popular.)
<MtotheM>
Especilly if you can create containers on the fly using the same configuration files. I can see myself using that feature a lot to create small boxes you can ssh into.
<MtotheM>
having then pre-defined*
minall has joined #nixos
<minall>
Hello nixos community!... I'm interested in trying nixos, but I have some questions...
<betawaffle>
does anyone know if `ninja -j 0` is different from giving the actual number of cores/threads to `-j`? It seems to be behaving differently.
<minall>
When I want to use for example, a apache server, or mysql database, do I need to add it to my system wide configuration in NixOS, or just installing it as a package of that user, I can do a 'systemctl start apache2' for example, and it will work
<betawaffle>
also, is `$NIX_BUILD_CORES` what i should be using?
<betawaffle>
on my machine it seems to be zero, which doesn't seem like what i want
<simpson>
Minall: That's your choice. NixOS has tools to ease configuring those services on a systemwide level (e.g. binding port 80), but you can also use those packages from an unprivileged user if you like, and there's even some support for systemd user services.
kenran has joined #nixos
meh` has quit [Ping timeout: 256 seconds]
fendor has quit [Remote host closed the connection]
proofofkeags has joined #nixos
<betawaffle>
"The cores setting determines the value of NIX_BUILD_CORES. NIX_BUILD_CORES is equal to cores, unless cores equals 0, in which case NIX_BUILD_CORES will be the total number of cores in the system."
<betawaffle>
but in my case, it's not setting NIX_BUILD_CORES to 64, just zero
kenran has quit [Ping timeout: 246 seconds]
<betawaffle>
any idea what i'm doing wrong?
<simpson>
betawaffle: No idea offhand, but it appears that at least some nixpkgs expressions know that it can be zero, and explicitly check for zero before using it.
<betawaffle>
gchristensen: it's weird, i've got a clang build that doesn't run out of memory with -j 64, but does run out at -j 0
<gchristensen>
how many cores do you have?
<betawaffle>
not sure how many cores ninja is assuming in the case of 0
<betawaffle>
64
<betawaffle>
(threads)
<minall>
infinisil: Because, in another distro, like arch, I just install apache2 and 'systemctl start apache2' it. Pretty fast, and I though the only way to add apache2 in nixos was doing it system wide in the configuration, and a nix rebuild switch, which will install a lot of new packages and waste some time... But If I can also do the first, installing
<minall>
apache with nix-env -i and systemctl start, and I'll have my server started, then there's no problem
leungbk has quit [Ping timeout: 240 seconds]
magnetophon has quit [Read error: Connection reset by peer]
<betawaffle>
gchristensen: and nproc reports 64 at least, so that's good
<simpson>
Minall: The time of building the system configuration is usually not that much, overall, and that'd be the only difference. Like, it's gonna be the *same* new packages either way, in Nix, right? Because Nix packages don't really care which user requested them.
<infinisil>
Minall: Adding `services.mysql.enable = true;` and running `nixos-rebuild switch` isn't slow, it installs exactly what's needed, nothing more
<asheshambasta>
srk: were you able to get audio to work with the jackd service?
<selfsymmetric-mu>
Hey, I'm looking for a package that can get me `libpq.so.5` in my nix-shell. I have one in my `/nix/store` but it's under `lnfh7mm8vw8sdmp1nccqmk0alvl6jm74-postgresql-9.6.11-lib` and that `-lib` is throwing me off. What should I put in my `buildInputs` and in my `LD_LIBRARY_PATH`?
<clever>
selfsymmetric-mu: what happens if you just put postgresql in buildInputs?
<{^_^}>
Found in packages: postgresql95.lib, postgresql96.lib, postgresql_10.lib, postgresql_11.lib, postgresql_12.lib
Chiliparrot has joined #nixos
<clever>
selfsymmetric-mu: and then link with -lpq
orivej has quit [Ping timeout: 256 seconds]
<infinisil>
So -lib corresponds to the .lib output
<infinisil>
And things in buildInputs automatically have their .lib output added
orivej has joined #nixos
<infinisil>
So as clever said, just adding postgresql in buildInputs should work
<{^_^}>
[nixpkgs] @FRidh merged pull request #90174 → python: don't use optimizations on Darwin → https://git.io/JfHLX
<{^_^}>
[nixpkgs] @FRidh pushed to staging-next « python: don't use optimizations on Darwin »: https://git.io/JfHz4
<srk>
asheshambasta: no, only by running it via qjackctl, need to investigate. I've used the same command line args as qjackctl did via services.jack.jackd.extraOptions and it hooks on correct soundcard but no sound
leungbk has joined #nixos
<asheshambasta>
I see, thanks. I'm still struggling even after `unload-module module-alsa-card`, using qjackctl I cannot seem to be able to set the driver to 'alsa' instead of 'dummy'.
chagra has quit [Ping timeout: 256 seconds]
user_0x58 has joined #nixos
<srk>
asheshambasta: my is running via alsa, only had to specify correct soundcard via extraOptions = [ "-dalsa" "-dhw:Solo" "-r48000" "-p1024" "-n2" ];
nikita` has joined #nixos
<srk>
for service to start correctly
zupo has joined #nixos
<betawaffle>
is 183 seconds a good clang build time?
<{^_^}>
[nixpkgs] @jlesquembre opened pull request #90198 → kubectx: re-pack, switch from bash to go → https://git.io/JfHzV
<asheshambasta>
I see. I feel quite lost at this point tbh.
<selfsymmetric-mu>
clever: What's -lpq?
<selfsymmetric-mu>
infinisil: Oh okay. Weird, I have postgresql in my buildInputs already…
<clever>
selfsymmetric-mu: that tells gcc to search for either libpq.so or libpq.a, and link to it
<selfsymmetric-mu>
Oh I see. I'm in Haskell, maybe that's different. I'll take a look.
<{^_^}>
[nixpkgs] @jeremyschlatter opened pull request #90199 → python3Packages.iterm2: init at 1.14 → https://git.io/JfHzr
<{^_^}>
#23454 (by nh2, 3 years ago, closed): nix-build can't build Haskell executables that depend on postgresql-libpq
<selfsymmetric-mu>
Closed as Can't Repro.
<srk>
asheshambasta: does it work with pulse disabled?
<catern>
is there a way to disable color output from "nix search"?
<asheshambasta>
srk: I've unloaded the alsa module in PA, so effectively, would that mean pa is disabled? Or do I need to kill the daemon?
<asheshambasta>
also I'm not sure which 'daemon' qjackctrl is showing me.
<selfsymmetric-mu>
I replace `postgresql` with `postgresql.lib` and added `${pkgs.postgresql.lib}/lib` for now.
<srk>
asheshambasta: when I try unload alsa-sink I get Module module-alsa-sink not loaded. and it looks like it's still uses the soundcard as it is visiable in configuration tab of pavucontrol
<asheshambasta>
and it seems like pactl cannot load the jack-sink module since it cannot connect to jack.
<srk>
qjackctl detects the system one if it's running
<srk>
(after restart)
magnetophon has quit [Read error: Connection reset by peer]
Chiliparrot has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
<asheshambasta>
I see, I've not rebooted.
<asheshambasta>
Maybe that is the issue.
<asheshambasta>
using pavucontrol I see no output devices but dummy.
zupo has quit [Remote host closed the connection]
<srk>
no need for reboots I would say, only if you're trying to get it start just right on boot
kreyren has quit [Remote host closed the connection]
<srk>
asheshambasta: what's the output of pactl load-module module-jack-sink channels=2 ?
magnetophon has joined #nixos
kreyren has joined #nixos
<srk>
for me it outputs a number and sink appears in pavucontrol
<asheshambasta>
Quite an unhelpful "module initialisation failed"
<srk>
I would try it step by step, get jack running first, then adding pulse
<clever>
asheshambasta: more output may wind up in the journal for the service running pulseaudio
kreyren has quit [Remote host closed the connection]
<phirsch>
asheshambasta: What works for me: pactl unload-module module-alsa-card; jack_control start . With module-jackdbus-detect loaded, PA then automatically loads module-jack-sink amodule-jack-source. Jack set to alsa driver. Make sure to unmute and adjust volume using alsamixer.
kreyren has joined #nixos
kreyren has quit [Remote host closed the connection]
<asheshambasta>
jack_control start fails for me with a DBus error: `DBus exception: org.jackaudio.Error.Generic: Failed to open server`
<srk>
phirsch: oh, good point about alsamixer
kreyren has joined #nixos
kreyren has quit [Remote host closed the connection]
kreyren has joined #nixos
<phirsch>
This is with services.jack.jackd.enable=false
<srk>
This sound device does not have any controls. :D
<srk>
can't even mpv -ao jack with jackd.enable = true
<phirsch>
srk: are you sure jack is configured correctly? (Haven't followed the full conversation. Jackdbus log output is in ~/.log/jack/jackdbus.log . Have you looked at /proc/asound/cards ?
Joy43 has quit [Remote host closed the connection]
civodul has quit [Ping timeout: 272 seconds]
<asheshambasta>
brb, desperate reboot attempt.
asheshambasta has quit [Remote host closed the connection]
OmnipotentEntity has joined #nixos
OmnipotentEntity has joined #nixos
OmnipotentEntity has quit [Changing host]
omneh has quit [Ping timeout: 246 seconds]
pamplemousse has joined #nixos
asheshambasta has joined #nixos
<asheshambasta>
okay, reboot, as one would've guessed, didn't help.
<asheshambasta>
This is incredibly frustrating. :-()
<phirsch>
asheshambasta: I remember seeing similar (identical?) errors from PA when running JACK as a system service. Starting JACKdbus via jack_control as described above works for me.
chloekek_ has joined #nixos
coderobe has quit [Ping timeout: 256 seconds]
teto has quit [Quit: WeeChat 2.8]
<phirsch>
Joy43: Mitigating dependency hell through transactional declarative installation of isolated software components yielding reproducible setups.
<phirsch>
Joy43: (That's not an official statement, just my personal understanding.)
<asheshambasta>
phirsch: I now have audio via Jack but PA still cannot see jack
<asheshambasta>
I'll disable the service and try with what you suggest
<phirsch>
asheshambasta: Do you have module-jackdbus-detect loaded?
<phirsch>
asheshambasta: This is AFAIK the NixOS PA default setup.
<asheshambasta>
yeah
<asheshambasta>
I'm seeing something like `Unable to contact D-Bus session bus: org.freedesktop.DBus.Error.NotSupported: Unable to autolaunch a dbus-daemon without a $DISPLAY for X11` in the ogs now
<asheshambasta>
logs*
magnetophon has quit [Read error: Connection reset by peer]
<phirsch>
asheshambasta: Did you disable the service? The message seems to indicate that the daemon is being started outside of an X11 session.
chrispickard has quit [Remote host closed the connection]
chrispickard has quit [Remote host closed the connection]
MtotheM1 has joined #nixos
<MtotheM1>
thanks
coderobe has joined #nixos
<phirsch>
srk: Nice.
<phirsch>
asheshambasta:
muslimm has joined #nixos
muslimm has quit [Max SendQ exceeded]
muslimm has joined #nixos
muslimm has quit [Max SendQ exceeded]
muslimm has joined #nixos
muslimm has quit [Max SendQ exceeded]
<phirsch>
* asheshambasta:Haven't been running JACK as realtime - forgot to add myself to the group first... Reboot has to wait for some more days...
<asheshambasta>
at this point I just wish I could do without pulseaudio and use things like my browser audio etc. without needing to do all of this.
muslimm has joined #nixos
muslimm has quit [Max SendQ exceeded]
MtotheM has quit [Ping timeout: 240 seconds]
<asheshambasta>
this looks like pulseaudio trying to connect to dbus.
<asheshambasta>
anyway, giving up on this for now, its too frustrating.
justanotheruser has quit [Ping timeout: 256 seconds]
<simpson>
MtotheM1: Nix itself will only operate on /nix, and nix-env will only touch a few specific .nix-*/ directories in your homedir. NixOS will leave homedirs alone, except where requested; you can also assume that anything custom you put in /etc will be okay.
coderobe has joined #nixos
<simpson>
MtotheM1: But, TBH, I would look at NixOS as *projecting* out the state of the system; the userland is projected from the Nix store onto your login session. The Nix store can hold many userlands; the trick is to realize that "the whole system" is mostly just a few clever symlinks.
selfsymmetric-mu has left #nixos ["gone to the land of dead hiccups and extinguished light bulbs"]
<{^_^}>
[nixpkgs] @peti pushed 439 commits to haskell-updates: https://git.io/JfHaY
leungbk has quit [Ping timeout: 256 seconds]
meh` has joined #nixos
drakonis has joined #nixos
doomlist3 has joined #nixos
minall has quit [Remote host closed the connection]
<jared-w>
Out of curiosity, if I want to save /nix/store in docker across various stuff in CI, would it be better to mount /nix as a volume or just /nix/store? The containers themselves are "mega-containers" and use their own nix; they're more akin to VM images than "real containers". I also don't have to worry about parallelism.
vandenoever has quit [Read error: Connection reset by peer]
vandenoever has joined #nixos
vandenoever has quit [Changing host]
vandenoever has joined #nixos
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « haskell-polysemy-plugin: drop obsolete overrides »: https://git.io/JfHVz
Chiliparrot has joined #nixos
cript0nauta has quit [Remote host closed the connection]
<energizer>
i have i/o errors reading from dev sda1 (the "/" partition), leading to boot failure. i'm guessing this is hopeless and i should just reinstall?
<{^_^}>
[nixpkgs] @primeos pushed 2 commits to release-20.03: https://git.io/JfHVi
<{^_^}>
[nixpkgs] @SCOTT-HAMILTON closed pull request #90030 → baobab: Fix missing icons on KDE → https://git.io/JfS3l
OmnipotentEntity has quit [Read error: Connection reset by peer]
OmnipotentEntity has joined #nixos
OmnipotentEntity has joined #nixos
OmnipotentEntity has quit [Changing host]
<MichaelRaskin>
At that point I would start evaluating whether you want to physically install a new storage device first… (not clear from one sentence, but worth investigating)
alp has joined #nixos
<{^_^}>
[nixpkgs] @jonringer pushed 8 commits to python-unstable: https://git.io/JfHV1
<energizer>
MichaelRaskin: what would i check?
<MichaelRaskin>
Well, what type of the errors these are…
gentauro has quit [Read error: Connection reset by peer]
alexherbo2 has joined #nixos
pamplemousse has quit [Ping timeout: 256 seconds]
gentauro has joined #nixos
<alexherbo2>
hi
meh` has quit [Ping timeout: 256 seconds]
<{^_^}>
[cabal2nix] @peti closed pull request #375 → Allows for package definition in package.dhall → https://git.io/fA0aV
<alexherbo2>
I have the following error trying experimenting with nix-build
<alexherbo2>
> error: anonymous function at /home/alex/root/projects/configuration/etc/nixos/pkgs/applications/networking/browsers/surf/default.nix:1:1 called without required argument 'gtk', at /nix/store/cmi14x6qb80v1ivg73r1r8y28c4dh93d-nixos-20.09pre228622.029a5de0839/nixos/lib/customisation.nix:69:16
<{^_^}>
error: syntax error, unexpected ':', expecting ')', at (string):316:133
<{^_^}>
[nixpkgs] @Mic92 pushed 3 commits to release-19.09: https://git.io/JfHVj
<energizer>
i dont have another disk atm, so i'm gonna try to reinstall. `nixos-generate -c configuration.nix -f iso` says The option `fileSystems./.fsType' has conflicting definitions, in `/nix/var/nix/profiles/per-user/root/channels/nixos/nixos/modules/installer/cd-dvd/iso-image.nix' and `/home/energizer/nixos-hosts/gamma/hardware-configuration.nix'.
dingenskirchen has quit [Remote host closed the connection]
dingenskirchen has joined #nixos
greizgh has quit [Quit: greizgh]
<energizer>
why does iso-image.nix have an opinion about my root fsType?
* energizer
wonders why everything talks about a "live CD" instead of a more modern format like HD-DVD
<MichaelRaskin>
I guess because CD and DVD actually got some use
<cransom>
i thought hd-dvd lost to bluray
<cransom>
you could say livebd, but i have only one machine in the house that has a working, external drive.
pamplemousse has joined #nixos
<simpson>
energizer: Space vs temporal access modes. Technically, an HD-DVD can be reburnt a few hundred times, which makes up for it, but otherwise there's just a lot of unneeded space.
domogled has joined #nixos
<simpson>
That said, doesn't at least Debian produce a series of DVD ISOs?
chagra has quit [Read error: Connection reset by peer]
rschulman has joined #nixos
<energizer>
twas a joke, im actually wondering why it doesnt say live usb :)
chagra has joined #nixos
<cole-h>
energizer: You could try `nix-build '<nixpkgs/nixos>' -A config.system.build.isoImage`
<{^_^}>
[nixpkgs] @sorki opened pull request #90205 → haskellPackages: Unbreak few, add me as a maintainer → https://git.io/JfHwM
domogled has quit [Quit: domogled]
omneh has quit [Ping timeout: 260 seconds]
orivej has quit [Quit: No Ping reply in 180 seconds.]
<energizer>
cole-h: what's the difference between that and nixos-generate -fiso?
<{^_^}>
[nixpkgs] @peti pushed 2 commits to haskell-updates: https://git.io/JfHoZ
orivej has joined #nixos
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JfHoW
civodul has joined #nixos
omneh has quit [Ping timeout: 258 seconds]
captn3m0 has joined #nixos
leungbk has joined #nixos
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage2nix: gcodehs and ttn-client can't be build because of broken dependencies »: https://git.io/JfHoz
captn3m0- has quit [Ping timeout: 258 seconds]
OmnipotentEntity has joined #nixos
OmnipotentEntity has joined #nixos
OmnipotentEntity has quit [Changing host]
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JfHoV
knupfer1 has joined #nixos
pamplemousse has quit [Ping timeout: 260 seconds]
knupfer has quit [Ping timeout: 256 seconds]
knupfer1 has quit [Ping timeout: 260 seconds]
das_j has quit [Quit: killed]
ajs124 has quit [Quit: killed]
Scriptkiddi has quit [Quit: killed]
mrpi has quit [Quit: killed]
brainopia has joined #nixos
<brainopia>
Hello, I'm trying to use services.elasticsearch and specify version as 7.7.1:
Scriptkiddi has joined #nixos
ajs124 has joined #nixos
mrpi has joined #nixos
das_j has joined #nixos
<brainopia>
I've tried package = services.elasticsearch7 // { version = "7.7.1" }, but it still uses the defealt 7.5.1
<brainopia>
default*
<{^_^}>
[nixpkgs] @peti pushed 11 commits to haskell-updates: https://git.io/JfHo9
<stoile>
brainopia: I think you have to override the src with overrideAttr.
OmnipotentEntity has quit [Ping timeout: 256 seconds]
szicari has quit [Ping timeout: 272 seconds]
<eyJhb>
Trying to use znapzend service in NixOS for backing my encrypted dataset (rpool is set so all datasets are encrypted), to my rtank (which has the same encryption fun), but it fails with ` cannot receive new filesystem stream: zfs receive -F cannot be used to destroy an encrypted filesystem or overwrite an unencrypted one with an encrypted one`
<eyJhb>
Any ideas?
jet_ has left #nixos ["Leaving"]
<bqv>
eugh
Chiliparrot has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
<eyJhb>
Weird that ZFS send/recv does not work nicely with encryption...
<eyJhb>
I would have thought it did
nikita` has joined #nixos
iqubic has joined #nixos
vandenoever has quit [Read error: Connection reset by peer]
<stoile>
FRidh: I think the apsw version needs to be in sync with sqlite, so the cherry-pick of 3.32.x might be a problem.
<MtotheM1>
why wouldn't it? it has native support for it eyJhb
vandenoever has joined #nixos
vandenoever has joined #nixos
vandenoever has quit [Changing host]
<iqubic>
So, I have the following ZFS settings set in my configuration.nix. http://dpaste.com/1JFVK6N How can I change how often the ZFS Snapshots are made, and how many are kept?
<eyJhb>
MtotheM1: apparantly having src and dst is not supported, as far as I can read because of some key fun...
user_0x58 has joined #nixos
kenran has quit [Ping timeout: 256 seconds]
<cole-h>
iqubic: `man configuration.nix` and search for `services.zfs.autoSnapshot` and see what other options it supports.
<{^_^}>
[cabal2nix] @peti pushed to master « doc: remember my live stream notes »: https://git.io/JfH6o
orivej_ has quit [Ping timeout: 260 seconds]
<iqubic>
Like, sure, the data type for the number of snapshots at each range to keep is signed integer, but I worry that values that are less than or equal to 0 will mess up the system.
jb55 has quit [Remote host closed the connection]
jb55 has joined #nixos
<eyJhb>
Any solution to this would be nice. src and dest are encrypted, same key
codygman has quit [Read error: Connection reset by peer]
alp has joined #nixos
codygman has joined #nixos
<jasom>
iqubic: it looks like it will install a timer that does nothing if you have keep set to 0, because the program it runs will exit immediately if keep is 0: do_new_snapshots(datasets, interval) if keep > 0
<iqubic>
Yeah. I know.
<jasom>
iqubic: I'm guessing a negative value will be bad though
<iqubic>
I suppose so.
codygman has quit [Read error: Connection reset by peer]
codygman has joined #nixos
leungbk has quit [Ping timeout: 256 seconds]
codygman has quit [Read error: Connection reset by peer]
codygman has joined #nixos
leungbk has joined #nixos
knupfer has quit [Quit: knupfer]
knupfer1 has joined #nixos
knupfer1 is now known as knupfer
omneh has joined #nixos
nikita` has quit [Ping timeout: 265 seconds]
slack1256 has joined #nixos
tokudan has quit [Remote host closed the connection]
<samueldr>
but a right answer is what tokudan said,
<energizer>
so if i want to avoid relying on environment info, what should i do?
<tokudan>
not a complete answer though :)
<samueldr>
energizer: what is the goal?
rickhull has joined #nixos
knupfer has quit [Quit: knupfer]
<energizer>
samueldr: whenever i install nixos, i have to mess around with channels imperatively and i dont like it, so i want to avoid that
<samueldr>
ah
<samueldr>
sorry, I don't know a "good" way, but what I do is mess with NIX_PATH so nixpkgs is what I want and extrapolate from there, I won't share exactly how because I don't know that it's "good"
<energizer>
is flakes a solution to this?
<samueldr>
it is _a_ solution
<samueldr>
(I don't have experience with it though)
<energizer>
"In the next blog post, we’ll talk about typical uses of flakes, such as managing NixOS system configurations, distributing Nixpkgs overlays and NixOS modules, and CI integration." https://www.tweag.io/blog/2020-05-25-flakes/
<chloekek_>
When building LLVM with CMake, and then building it again later with more features enabled, it won’t rebuild from scratch.
<chloekek_>
I suppose with Nix that cannot be avoided indeed, because Nix invokes CMake without any object files in scope.
MmeQuignon has quit [Ping timeout: 256 seconds]
rickhull has left #nixos [#nixos]
<chloekek_>
It would require reimplementing CMake in Nix, and generating fine-grained derivations, but that’s not what it does currently, so I’ll just rebuild it from scratch.
<MtotheM1>
can I not reference options inside of a string? I have this `"{networking.hostName}"` but it sais 'networking' is an undefined variable. despite me having declared it above.
leungbk has quit [Remote host closed the connection]
alp has quit [Remote host closed the connection]
leungbk has joined #nixos
alp has joined #nixos
vykook has quit [Ping timeout: 256 seconds]
<jasom>
chloekek_: plus there is no guarantee that the result of building for target X, then adding target Y and rebuilding will be the same as building target Y, then adding target X and rebuilding, so the derivation would have to depend on the ordering as well as the list of targets.
dxtr has quit [Ping timeout: 246 seconds]
alp has quit [Remote host closed the connection]
<chloekek_>
jasom: LLVM + AVR would just take LLVM as an input. Would have a different input hash, so no problem.
alp has joined #nixos
drakonis_ has joined #nixos
<jasom>
chloekek_: you could make that possible by saving off the entire build tree in your LLVM derivation, and then using that as an iput for building with +AVR; but obviously the LLVM in nixpkgs doesn't do taht.
<chloekek_>
Right.
codygman has quit [Read error: Connection reset by peer]
leungbk has quit [Quit: ERC (IRC client for Emacs 27.0.91)]
virus_dave has quit [Client Quit]
virus_dave has joined #nixos
virus_dave has quit [Client Quit]
virus_dave has joined #nixos
virus_dave has quit [Client Quit]
virus_dave has joined #nixos
virus_dave has joined #nixos
virus_dave has quit [Client Quit]
virus_dave has joined #nixos
virus_dave has quit [Client Quit]
vykook has quit [Ping timeout: 256 seconds]
virus_dave has joined #nixos
virus_dave has quit [Client Quit]
justanotheruser has joined #nixos
virus_dave has joined #nixos
vykook has joined #nixos
drakonis1 has joined #nixos
vykook[m]1 is now known as Vykook[m]
drakonis_ has quit [Ping timeout: 260 seconds]
agsdheidjd has quit [Ping timeout: 260 seconds]
codygman has quit [Read error: Connection reset by peer]
vykook has quit [Ping timeout: 258 seconds]
codygman has joined #nixos
alp has quit [Ping timeout: 272 seconds]
vykook has joined #nixos
muslimm has joined #nixos
gustavderdrache has quit [Quit: Leaving.]
muslimm has quit [Max SendQ exceeded]
muslimm has joined #nixos
muslimm has quit [Max SendQ exceeded]
muslimm has joined #nixos
muslimm has quit [Max SendQ exceeded]
muslimm has joined #nixos
muslimm has quit [Max SendQ exceeded]
muslimm has joined #nixos
muslimm has quit [Max SendQ exceeded]
<infinisil>
bqv: Feel free to propose a PR to make it work with flakes too
<dgpratt>
I have the nix package manager configured on Ubuntu. I notice that nix puts some things in ~/.nix-profile/share that (I think) might otherwise go in ~/.local/share.
<dgpratt>
is this happening because if I were on full-blown NixOS, that would be the place to put such things?
<bqv>
infinisil: well, how? the problem is that pkgs isn't plumbable, but would you be ok with it being?
<bqv>
or well no, i guess that's not the problem
<bqv>
but i can't think of any other solution off head
<dgpratt>
in short, I'm trying to understand the scheme at work here and how best to get things like fonts installed with nix-env where other things expect to find them
justanotheruser has quit [Ping timeout: 240 seconds]
vykook has quit [Ping timeout: 246 seconds]
<infinisil>
bqv: Tbh I have no idea, especially since with the issue I linked, I plan to only provide an overlay. But the overlay internally uses its own pinned nixpkgs versions
<infinisil>
Though maybe that's no problem with flakes?