mmmattyx has quit [Quit: Connection closed for inactivity]
steadmon has quit [Quit: leaving]
<clever>
spease: nix-env will generate that automatically when you install something
philr has joined #nixos
<spease>
@clever what command do I need to use?
<clever>
spease: nix-env -iA nixos.hello
<clever>
just install anything, and it creates a profile
noudle has quit []
shibboleth has quit [Quit: shibboleth]
<spease>
@clever Thanks, that fixes that. Unfortunately looks like I'm still getting a "error: file 'darwin' was not found in the Nix search path" :-/
<clever>
having a profile wont change the contents of $NIX_PATH
<spease>
I uninstalled my nix install, then had to reinstall a few times because the uninstall instructions weren't accurate (and differed in various places). I'm guessing that's what's causing this mess. I've been trying to surgically fix it...
<clever>
why did you uninstall it?
<spease>
I was going to reinstall it with brew, but then it turned out that brew cask didn't seem to exist anymore
<clever>
nix should always be managed by nix
<clever>
installing it with another package manager tends to break the sandbox stuff
<spease>
ahh
<jlv>
Wait. If Nix has to be installed with Nix, how to you install Nix if you don't have Nix?
<clever>
jlv: the install script will unpack a tar of a /nix/store that has nix and all of its deps
<spease>
@clever Any idea how nix-darwin could install successfully, but I still get "error: file 'darwin' was not found in the Nix search path (add it using $NIX_PATH or -I)"?
<clever>
spease: did you try logging out and back in?
<spease>
just opened a new window. brb.
spease has quit [Remote host closed the connection]
<nilsirl[m]>
Hi, I've been trying to connect to a wireguard server, but when I run `sudo wg-quick up ./config.conf`, I loose internet connectivity and nothing works. I have to disable wireguard for my internet to work again. I believe this is failure is linked to nixos because out of my 2 attempts to get wireguard to work on this nixos system, none of the have worked and I've been able to connect with the exact same wireguard
<nilsirl[m]>
configuration on my phone and with a different configuration some time ago on an arch system.
<nilsirl[m]>
I'm not using nixos' builtin wireguard support, just `wg-quick` directly from the command line and have nothing in my network settings
Darkmatter66 has quit [Quit: ZNC 1.7.5 - https://znc.in]
Darkmatter66 has joined #nixos
rajivr has joined #nixos
justanotheruser has quit [Ping timeout: 260 seconds]
<nicolas[m]1>
There's many open issues regarding Wireguard on NixOS. If you tunnel your entire traffic through Wireguard then it doesn't work AFAIK
D_ has quit [Ping timeout: 260 seconds]
D_ has joined #nixos
<{^_^}>
[nix] @chkno opened pull request #3932 → Don't try to parse signature check as commit timestamp → https://git.io/JJ7s7
<nilsirl[m]>
nicolas: 😂
<nilsirl[m]>
So I suppose there aren't any fix. Are there any open issues for this?
<nilsirl[m]>
after rereading your message, I guess the correct message would be, what are the open issues regarding wireguard? #51258 seems interesting
marcusr has quit [Remote host closed the connection]
marcusr has joined #nixos
kalbasit has joined #nixos
matthuszagh has joined #nixos
<matthuszagh>
hello. anyone know of a good guide for setting up networking for qemu/kvm? i'm running ubuntu in a virtual machine and can't get the networking to work
<matthuszagh>
i have another vm that runs windows 10 where i used -net nic and it works fine
<matthuszagh>
but that doesn't seem to work for me with a ubuntu guest. started reading about this a little and it sounds like -device virtio-net is a better way to do it anyway
<kalbasit>
anyone using Brave? I can't seem to find the Sync menu on mine
<matthuszagh>
or if not a guide, someones dotfiles where this works
zangi has quit [Quit: WeeChat 2.7.1]
<{^_^}>
[nixpkgs] @aanderse merged pull request #95231 → nixos/mysql: run postStart as an unprivileged user → https://git.io/JJyjW
<spease>
Is $NIX_PATH supposed to be pointing to root channels while .nix-defexpr etc points to user channels?
proofofkeags has joined #nixos
<bqv>
basically, yeah
markus1189 has quit [Ping timeout: 256 seconds]
Alaura has joined #nixos
Alaura has quit [Max SendQ exceeded]
Alaura has joined #nixos
Alaura has quit [Max SendQ exceeded]
Alaura has joined #nixos
Alaura has quit [Max SendQ exceeded]
ardumont has joined #nixos
waleee-cl has quit [Quit: Connection closed for inactivity]
worldofpeace_ has joined #nixos
knerten2 has joined #nixos
<spease>
Is there documentation anywhere of the current setup that's needed in ~/.zshrc and/or ~/.profile for nix (particularly nix-darwin) to set NIX_PATH properly? If I use the NIX_PATH from the manual install instructions I can use darwin-rebuild, but it reverts as soon as I launch a new shell
veleiro has quit [Remote host closed the connection]
veleiro has joined #nixos
mounty has joined #nixos
spease has joined #nixos
<spease>
Figured it out. It turns out that there was a `. '/nix/var/nix/profiles/default/etc/profile.d/nix-daemon.sh'` in /etc/zshrc that overrides the darwin `/etc/zshenv` . `nix-env -q` output still doesn't show the installed packages though, even though I can interact with the binaries or remove them and switch and have them disappear.
sigmundv has quit [Ping timeout: 256 seconds]
matthusz` has left #nixos ["ERC (IRC client for Emacs 27.1)"]
<slashzero>
Hi, I'm just getting started with NixOS, so pardon my ignorance. I just did a fresh install inside of VirtualBox. However, whenever I set `services.xserver.enable = true` I see an error "error 9 while decompressing xz file" when it tries to build the mesa-drivers derivation.
<spease>
Ahh - looks like it shouldn't since I have all my packages installed with nix-darwin's configuration file. 🤦
<energizer>
slashzero: that makes me think it ran out of memory
<hyper_ch>
nilsirl[m]: you want to tunnel the traffic from your phone through your nixos system?
<nilsirl[m]>
hyper_ch: nah 😂, I was saying that it worked on my phone but didn't work on my desktop system running nixos (suggesting that the problem is not linked to my wireguard configuration file or server)
<hyper_ch>
nilsirl[m]: ah... ok :)
<slashzero>
energizer: I don't think that's it. The VM has 3.5GiB free and 8GiB of swap.
dansho has quit [Remote host closed the connection]
dansho has joined #nixos
<spease>
slashzero Are you sure the file is downloading completely and that it's still the right file on the server? It looks like an interrupted download has caused that error for people based on a google search
<{^_^}>
[nixpkgs] @nagisa opened pull request #95392 → odaconverter: init at 21.7.0.0 → https://git.io/JJ7CL
e has quit [*.net *.split]
growpotkin has quit [*.net *.split]
evils has quit [*.net *.split]
philr has quit [*.net *.split]
user_0x58 has quit [*.net *.split]
troydm has quit [*.net *.split]
samrose has quit [*.net *.split]
Guest46902 has quit [*.net *.split]
n9nes- has quit [*.net *.split]
zal has quit [*.net *.split]
lewo` has quit [*.net *.split]
LnL has quit [*.net *.split]
buffet has quit [*.net *.split]
omneh has quit [*.net *.split]
stree has quit [*.net *.split]
Unode has quit [*.net *.split]
raoul has quit [*.net *.split]
edcragg has quit [*.net *.split]
lord| has quit [*.net *.split]
samhza has quit [*.net *.split]
ktosiek has quit [*.net *.split]
jD91mZM2 has quit [*.net *.split]
Streetwalrus has quit [*.net *.split]
judson has quit [*.net *.split]
jhuizy has quit [*.net *.split]
cransom has quit [*.net *.split]
m4ts has quit [*.net *.split]
dweller has quit [*.net *.split]
reivilibre has quit [*.net *.split]
sjourdoi- has quit [*.net *.split]
tobiasBora has quit [*.net *.split]
drewc has quit [*.net *.split]
statusbot has quit [*.net *.split]
dminuoso has quit [*.net *.split]
tpanum has quit [*.net *.split]
Philonous has quit [*.net *.split]
runciter has quit [*.net *.split]
electrocat has quit [*.net *.split]
switchy has quit [*.net *.split]
WeblordPepe has quit [*.net *.split]
oleks has quit [*.net *.split]
pie_ has quit [*.net *.split]
deni has quit [*.net *.split]
dfgg has quit [*.net *.split]
djanatyn has quit [*.net *.split]
vikingman has quit [*.net *.split]
mxco86 has quit [*.net *.split]
averell has quit [*.net *.split]
dominikh has quit [*.net *.split]
callahad87074 has quit [*.net *.split]
S0rin has quit [*.net *.split]
Uma has quit [*.net *.split]
Czen has quit [*.net *.split]
edwtjo has quit [*.net *.split]
lohfu has quit [*.net *.split]
kwork has quit [*.net *.split]
aminechikhaoui has quit [*.net *.split]
rotaerk has quit [*.net *.split]
nore has quit [*.net *.split]
mcint has quit [*.net *.split]
FireFly has quit [*.net *.split]
Aleksejs has quit [*.net *.split]
maurer has quit [*.net *.split]
mk_modrzew has quit [*.net *.split]
rkallos has quit [*.net *.split]
djahandarie has quit [*.net *.split]
Nafai has quit [*.net *.split]
moredhel has quit [*.net *.split]
srk has quit [*.net *.split]
hyperfekt has quit [*.net *.split]
averell has joined #nixos
mxco86 has joined #nixos
callahad87074 has joined #nixos
dominikh has joined #nixos
Czen has joined #nixos
S0rin has joined #nixos
kwork has joined #nixos
rotaerk has joined #nixos
nore has joined #nixos
aminechikhaoui has joined #nixos
Uma has joined #nixos
lohfu has joined #nixos
rkallos has joined #nixos
mcint has joined #nixos
edwtjo has joined #nixos
FireFly has joined #nixos
srk has joined #nixos
djahandarie has joined #nixos
mk_modrzew has joined #nixos
Nafai has joined #nixos
Aleksejs has joined #nixos
maurer has joined #nixos
hyperfekt has joined #nixos
moredhel has joined #nixos
endformationage has quit [Quit: WeeChat 2.7.1]
lord| has joined #nixos
dansho has quit [Ping timeout: 264 seconds]
<_habnabit>
how do you go from a path or store to the package that you need to install to get it? specifically trying to install nixos-rebuild on a non-nixos system so i can rebuild a system remotely
aw has quit [Quit: Quitting.]
spacefrogg has quit [Quit: Gone.]
aw has joined #nixos
user_0x58 has joined #nixos
evils has joined #nixos
philr has joined #nixos
growpotkin has joined #nixos
LnL has joined #nixos
buffet has joined #nixos
n9nes- has joined #nixos
omneh has joined #nixos
lewo` has joined #nixos
samrose has joined #nixos
zal has joined #nixos
Guest46902 has joined #nixos
troydm has joined #nixos
Unode has joined #nixos
stree has joined #nixos
ktosiek has joined #nixos
judson has joined #nixos
jD91mZM2 has joined #nixos
Streetwalrus has joined #nixos
edcragg has joined #nixos
jhuizy has joined #nixos
oleks has joined #nixos
pie_ has joined #nixos
drewc has joined #nixos
dminuoso has joined #nixos
reivilibre has joined #nixos
raoul has joined #nixos
sjourdoi- has joined #nixos
WeblordPepe has joined #nixos
Philonous has joined #nixos
electrocat has joined #nixos
tpanum has joined #nixos
tobiasBora has joined #nixos
deni has joined #nixos
switchy has joined #nixos
djanatyn has joined #nixos
runciter has joined #nixos
dweller has joined #nixos
dfgg has joined #nixos
cransom has joined #nixos
m4ts has joined #nixos
statusbot has joined #nixos
samhza has joined #nixos
ktosiek has quit [Remote host closed the connection]
spacefrogg has joined #nixos
slashzero has quit [Remote host closed the connection]
proofofkeags has quit [Remote host closed the connection]
<bqv>
pulseaudio uses all of them
<bqv>
select with pavucontrol
<bqv>
or similar
redmp has joined #nixos
<redmp>
what's the best way to pass overridden nixpkgs & custom dependencies from the default.nix to shell.nix?
<redmp>
(i'd like these things to be available eg. so that i can tell the shell environment to include cabal-install or so)
edk_ has joined #nixos
leothrix has quit [Ping timeout: 256 seconds]
<redmp>
or, asked another way, if this doesn't sound right, what's the right way to approach this problem?
<redmp>
up until now i put everything into default.nix and used isNixShell to decide whether to produce a drv or an environment, but this bit me today and so i'm trying to separate them out.. there's a lot of additional complexity it seems, however, in separating these files
<energizer>
redmp: my current pattern is to have default.nix take arguments, and have shell.nix pass those arguments to it
<redmp>
ahh... energizer something like `{ makeEnv ? false }:` in default.nix?
<energizer>
no
<energizer>
just run shell.nix when i want a shell, and default.nix when i dont
<redmp>
with the pinning/overrides/dependencies in which of them?
cmk_zzz_ is now known as cmk_zzz
<energizer>
in the shell has been working so far for me
Thra11 has quit [Quit: WeeChat 2.8]
<redmp>
how do you pass arguments from shell.nix to default.nix when you do nix-build?
<redmp>
since nix-build calls default.nix you'd have to `import ./shell.nix {}`? that seems the opposite of what you said a bit ago
<simpson>
I wonder what the structure of your workflow is like. I generally keep shell.nix for the tools that I need for development, and default.nix for the build of the project being developed; there's no overlap. nixpkgs is pinned by setting the NIX_PATH environment variable.
<simpson>
I use direnv to coordinate setting up nix-shell and NIX_PATH, but it can also be done explicitly and manually.
<redmp>
my setup: only have `default.nix` and end it with `if nixpkgs.lib.inNixShell then drv.env else drv` .. this works great with nix-build and nix-shell, but i found out today, doesn't do what i want when imported for deployment with nixops
<simpson>
All that said, yeah, importing default.nix from shell.nix isn't bad, and it works. It would cause my direnv to reload constantly, so I don't do it.
<simpson>
Ah, interesting. To build and test projects, I normally incant nix-build and then poke at the result/ symlink.
<spease>
I've been trying to create a wrapper for the OS X system clang in order to be able to build with it. However, cmake doesn't seem to detect it. Right now I'm doing `stdenv = if pkgs.stdenv.isDarwin then pkgs.overrideCC pkgs.stdenv (pkgs.callPackage ./clang-wrapper.nix { pkgs = pkgs; }) else pkgs.stdenv;`
<spease>
Does that seem right?
<redmp>
i don't use direnv; i'd like to have the nix pin version in the file, but maybe what i'm doing is silly
<redmp>
xd1le: if you boot into a livecd, you can experiment with mount options rapidly `mount -o <options> <device> <mountpoint>` .. this might be easier than experimenting with your configuration.nix.. once you find the right options, write them in your nix file and apply
<dminuoso>
xd1le: yeah
<redmp>
(assuming you're trying to do this with your root partition)
evanjs has quit [Read error: Connection reset by peer]
<pickfire>
Hi, I am new to using nix. I just started out with home-manager. I tried setting `home.language.base = "en-US.UTF-8"` but why do I still get `/home/ivan/.nix-profile/bin/man: can't set the locale; make sure $LC_* and $LANG are correct` when I tried to `man home-configuration.nix`?
<{^_^}>
[nixpkgs] @utdemir opened pull request #95409 → dockerTools.streamLayeredImage: Store the customisation layer as a tarball → https://git.io/JJ76U
<felschr[m]>
Is it possible to make it use python3?
ml| has quit [Ping timeout: 264 seconds]
<immae>
felschr[m]: the line you quoted is the one, you could replace neovim with neovim.override({ python = python3 }) in your systemPackages path to avoid that
<nature>
felschr[m]: https://nixos.wiki/wiki/Vim you have some indications about vim and python3, but neovim probably has the same thing where you do neovim.override
<felschr[m]>
Doesn't `python ? null` mean that it's usually not set?
<srhb>
felschr[m]: I think it's actually in the neovim wrapper.
<immae>
felschr[m]: it’s not set by default, but if it remained null the derivation wouldn’t build. So something must override the null value
<nature>
I have the following in my configuration.nix: https://paste.rs/GRD, my problem is that I still don't have any extensions in chromium aftern rebuilding...
<felschr[m]>
oh, I see. It's probably line 132 in `wrapper.nix`?
<felschr[m]>
I didn't noticed that the `withPython` argument is set to `true` by default.
<srhb>
felschr[m]: And indeed, with neovim = wrapNeovim neovim-unwrapped { withPython = false; } -- the py2 dep goes away
<srhb>
felschr[m]: (No idea whether it works though)
<felschr[m]>
Neat. Thanks, I'll try my luck.
ManiacOfMadness has joined #nixos
<{^_^}>
[nixpkgs] @mweinelt merged pull request #95086 → nixos/manual: add a section about enabling OpenCL for Intel GPUs → https://git.io/JJDqk
<Unode>
hi everyone, I have a bit of a strange issue with networking and DNS resolution when using a VPN (openconnect).
<Unode>
A minute or so after the VPN connection is established, some commands resolve hostnames through the VPN while others through the external connection.
<{^_^}>
[nixpkgs] @CMCDragonkai opened pull request #95419 → pythonPackages.aiojobs: init at 0.2.2 → https://git.io/JJ79e
<{^_^}>
[nixpkgs] @Rizary opened pull request #95420 → nar-serve: init at 0.1.0 → https://git.io/JJ79a
<hexa->
Izorkin: you once asked about librenms here, what's the latest on that?
<Raito_Bezarius>
Is there some docs on how NixOS builds its initramfs?
<Raito_Bezarius>
Is there a way to add my own stuff in the ramdisk?
mariatsji has quit [Remote host closed the connection]
<immae>
Raito_Bezarius: nixos/modules/system/boot/stage-1.nix (note that there are recent changes so "add your own stuff in the ramdisk" may have different response depending on the version and on what you want to put in it - secrets or normal files or ...)
<Raito_Bezarius>
immae: 20.03+
<Raito_Bezarius>
It's not really secret but it's a custom full disk decryption logic
<Raito_Bezarius>
Thanks immae
<Raito_Bezarius>
immae++
<{^_^}>
immae's karma got increased to 22
mariatsji has joined #nixos
<immae>
Then take into account that it will change a bit when you upgrade to 20.09
<Raito_Bezarius>
Alright, that works
cosimone has joined #nixos
mariatsji has quit [Ping timeout: 260 seconds]
cosimone has quit [Remote host closed the connection]
<siraben>
I think that's the cause at least, since I have the error when building zathura on Darwin;
<siraben>
`error: nm: invalid argument --`
mariatsji has joined #nixos
<niso>
niksnut: i'm currently tyring to build a repo and realized that i'd have to pass stuff like nixpkgs down to the indivdual flakes (maybe with "requires" if this is still uptodate?)
<niso>
niksnut: hydra specifies it's own nixpkgs version in the flake, means i'd have to individually upgrade nixpkgs (e.g. to 20.09) on a per-package base
<niso>
niksnut: thus i was wondering if there is something like a reference repository which already solves gotchas like that
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
zupo has joined #nixos
<niksnut>
generally you don't pass nixpkgs to your input flakes
n9nes- is now known as n9nes
<niksnut>
instead you use the overlays or nixos modules provided by those flakes
<niksnut>
and plug them into your top-level nixpkgs
<niso>
niksnut: means packages shouldn't use flakes.nix instead of default.nix?
<balsoft>
niksnut: While we're on this subject, do you have any recommendations for projects which do want to provide users with tested versions of nixpkgs, e.g. like https://github.com/input-output-hk/haskell.nix ? Specifying those versions as inputs is a bad idea (since inputs are strict)
<Mic92>
balsoft: you can pin nixpkgs in this case no?
<balsoft>
Mic92: Hmm, what do you mean?
<niso>
Mic92: exaclty what i was looking for, thanks
<balsoft>
haskell.nix provides users with two versions of nixpkgs (unstable and release) and users may choose to use one of those or another version entirely (in which case they're on their own).
<niso>
niksnut: thanks for your help too! i was looking for inputs.something.follows = "";
<balsoft>
Mic92: ah, you mean pinning them without flakes? That was my thought too, but using two mechanisms of updating dependencies seems overcomplicated.
<niksnut>
you can say something like: inputs.nixpkgs.url = github:NixOS/nixpkgs/<rev>;
<niksnut>
in your flake.nix
<niso>
niksnut: exactly, but this means that i'd have to change this line 100 times for my 100 packages
<niso>
niksnut: with follows i can make the 100 packages depend on the 1 main package
<niso>
niksnut: s/1 main package/1 main flake
<Mic92>
niso: you also don't need to pin nixpkgs in flakes
<freezeboy>
any idea about a way to package multiple variants of the same library with multiple flags ?
sangoma has quit [Ping timeout: 256 seconds]
nij has joined #nixos
turlando has quit [Ping timeout: 256 seconds]
iyzsong has joined #nixos
<freezeboy>
librealsense could have options for With(out)Cuda, With(out)Examples, With(out)AdditionalDrivers ... if we combine all of them it really gets complicated...
<nij>
If I plan to purchases several thinkpads, varying from old models like x200 to new models like x1carbon, how likely would my configuration.nix work for all of them out of the box?
<nij>
(my current config.nix works for x220)
mariatsji has quit [Ping timeout: 240 seconds]
<immae>
freezeboy: the usual way is to add { withCuda ? false, withExamples ? false, withAdditionalDrivers ? false } in the signature of the function
<freezeboy>
immae yes but should I propagate all variants as a pkgs attribute ?
<immae>
That’s you to decide. Only the propagated variants will get built by hydra
disasm has joined #nixos
<immae>
you could put "common" variants like librealsense-cuda, librealsense-full for instance
<adisbladis>
Only create attributes for what makes sense for a large number of people imo
<adisbladis>
It's hard to say exactly what that means
<freezeboy>
yes ... that's the problem I have no idea which versions make sense for most people
<adisbladis>
Re what Hydra builds it's worth knowing that it doesn't build unfree packages, so no cuda variant will be built on hydra anyway
<{^_^}>
[nixpkgs] @jtojnar pushed commit from @dasj19 to master « quadrapassel: 3.36.02 -> 3.36.05 »: https://git.io/JJ7xH
wolfshappen has quit [Ping timeout: 240 seconds]
<nij>
Asked above when people were busy :| ..
<nij>
I have a configuration.nix that works for my thinkpad x220.
<nij>
How like would it also work for other thinkpad machines, varying from old to new models?
<freezeboy>
@nij
<nij>
Or should I expect to make one config.nix for each machine.
<adisbladis>
nij: It should just work
<freezeboy>
nij difficult to say, it really depends on harware support
alp has quit [Ping timeout: 272 seconds]
<freezeboy>
for hight level software support it should be ok
<nij>
adisbladis: that's my impression, but I saw someone kept different files for different machines.. I'm a bit worried.
<nij>
freezeboy: I see
<adisbladis>
Barring hardware specifics everything should just work the same
<nij>
the hardware.config is autogenerated while I installed nixOS
<srhb>
nij: I wouldn't worry too much. Most of that hardware is really just kernel support, especially if you don't have to deal with special GPUs
wolfshappen has joined #nixos
<nij>
I guess that file would be different automatically on different machines?
<srhb>
nij: Yeah, especially when using auto generated partition identifiers
<nij>
srhb: That's a relief to hear :)
<nij>
yay nixos rocks
<srhb>
nij: The usual approach is to have one small top level configuration.nix file which has hardware/machine specific information, and imports the "common config"
<srhb>
nij: When your systems diverge, that is.
<srhb>
Which is fine and works great.
<nij>
Super neat
polezaivsani has joined #nixos
<nij>
Thanks :)
<nij>
Maybe another question..
<nij>
I just purchased a Brother's scanner that requires a driver to run..
<nij>
I'm on arch, and just realized that Brother provided drivers for debian and rpm..
<nij>
fortuantely it would work on arch.. I just have to extract the driver and put it at the right location
<nij>
But how would this work on nixOS?
freezeboy has quit [Ping timeout: 245 seconds]
<srhb>
nij: Usually you'd have to package it to some degree if it isn't already.
<srhb>
And then use one of the nixos options to inject that driver/plugin into sane somehow
<srhb>
Disclaimer: I know very little about scanners :P
<nij>
srhb: no worries
ddellacosta has quit [Ping timeout: 260 seconds]
<manveru>
nij: `hardware.sane.extraBackends`
ddellacosta has joined #nixos
<manveru>
and `hardware.printers.ensurePrinters.model` i think
<manveru>
though not sure if you need that for scanners
<{^_^}>
[nixpkgs] @jtojnar pushed commit from @dasj19 to master « gdk-pixbuf-xlib: 2019-10-19-unstable -> 2020-06-11-unstable »: https://git.io/JJ5TQ
<vivasvat>
question, has emacs 27 been made available in the repos?
<NobbZ[m]>
Currently it's only available through the overlay AFAIK.
knupfer has quit [Remote host closed the connection]
bbarker has joined #nixos
knupfer has joined #nixos
<bbarker>
I'm wondering if it is possible in nix's hackage (haskell package) support to somehow annotate that certain Haskell packages require system dependencies ... so I don't have to figure it out every time i include one in a project. Would seemingly be a good use of Nix as well
<bbarker>
just an example: the haskell package 'digest' seems to require zlib
<clever>
bbarker: if its properly defined in the cabal file, then cabal2nix should expect pkgs.zlib to exist, as an input
<bbarker>
clever, can you give an example of that behavior (a good cabal file that does this)? I'll start to raise the issue upstream
<clever>
bbarker: basically, you just want a extra-libraries: field in the cabal file
<clever>
bbarker: also, i think the library is called libz.so, linked with -lz, so you would want `extra-libraries: z` and `pkgs.z`
inkbottle has joined #nixos
<clever>
you also need to be careful you dont depend on zlib.cabal
<bbarker>
that was already with stack build --verbose
yoctocell[m] has joined #nixos
proofofkeags has quit [Ping timeout: 264 seconds]
<clever>
bbarker: `stack build` cant use the deps info from each package, and it ignores what nix has already built
<clever>
bbarker: so you must re-create the list of buildInputs everything needs in your shell.nix file, this happens every time somebody tries using stack
<clever>
bbarker: i find it MUCH simpler to ignore stack, and use plain cabal (old cabal)
jumper149 has joined #nixos
<bbarker>
clever, technically i was using `stack --nix build`, sorry forgot to include that
zaeph has quit [Ping timeout: 256 seconds]
<bbarker>
but i guess the problem may still apply?
<clever>
bbarker: but that just runs a shell.nix file, which must have zlib in the buildInputs\
<bbarker>
ah i see
<clever>
the stack method isnt aware of any native deps for the packages
<bbarker>
i'll give cabal a try then, thanks
<clever>
one reason i avoid stack
<clever>
you need to cabal2nix your project into nix, and then run nix-shell on the .env of the project
<{^_^}>
[nixpkgs] @vcunat pushed to master « qt5: switch back to 5.12 on darwin »: https://git.io/JJ5LF
ManiacOfMadness has joined #nixos
__monty__ has joined #nixos
rembo10 has joined #nixos
whald has quit [Quit: Leaving]
<manveru>
hmm
<manveru>
> let x = runCommand "x" {} "echo yay > $out"; y = builtins.readFile x; in builtins.hashFile "sha256" (builtins.seq y x)
<{^_^}>
cannot read '/nix/store/bsvmliw3a8bxzxy1skbqpp12vfxhpgsy-x', since path '/nix/store/yllqhj8j62ir8nlhmy53ya0nic4yv392-x.drv' is not valid, at (string):320:50
<manveru>
hashFile is strange..
gthm has joined #nixos
<{^_^}>
[nixpkgs] @primeos merged pull request #95327 → cmake: Add a patch to fix a regression since CMake 3.18 → https://git.io/JJH2x
<x39>
Hello there, recently switched over to a dualboot-system using nixos and windows and facing problems here. The NixOS installation keeps on crashing at random all few minutes. Sadly ... the interweb is quite unhelpful in assisting me to how to get to the actual cause. So i hope that somebody here could hint me towards what i have to do (installation
<x39>
is pretty much 1:1 from the ISO with the only changes to nix-env introduced being a default-firefox and git installation). Running System with *MSI X470 Gaming Plus* mobo, *AMD Ryzen 7 2700X* CPU, *GTX 1080 TI Zotac Mini* GPU and round about 6 drives (1 NVME, 2 SSD, 3 HDD) and am pretty much a rookie with nixos for the kinda obvious crashing
<x39>
problem. Thanks in Advance for any guidance (though do note that testing methods out is obviously slow as i have to reboot into NixOS etc.)
<kini>
x39: it may help to know what exactly you mean by "crashing".
Pidgeotto has joined #nixos
Pidgeotto has quit [Excess Flood]
Pidgeotto has joined #nixos
Pidgeotto has quit [Excess Flood]
<x39>
that is the point where it gets triggy, i suspect a kernel panic but i am not experienced enough with nixos to diagnose. The behavior observed is that the whole system does a reboot
<x39>
sometimes with a keystroke done, sometimes during idle, sometimes while clicking
<feffe[m]>
anyone know how the `staging` and `staging-next` tends to work? How often is staging typically merged into next and how long in there until master? What I've found so far is the rfc https://github.com/NixOS/rfcs/blob/master/rfcs/0026-staging-workflow.md, which claims that (as far as I understand it) checks to merge every 6 and 12 hours respectively for `staging` and `staging-next`. I've been waiting for
<feffe[m]>
`9f911ab1b588cedfd5918ed5e09870648a528d57` to hit master but I don't konw when I should expect that to happen.
<feffe[m]>
nij:
<feffe[m]>
* nij: stuff that needs to patch something after it's built, like vscode live share for example
<NobbZ[m]>
In the first month I really missed the ability to bump from AUR and ABS temporarily, though in the meantime I consider overrides and overrideAttrs even nicer
<feffe[m]>
nij: yeah at least in my experience, there is a work around but it's very much agains the nixos way of doing things
<nij>
NobbZ[m]: Do you feel that if you want to configure anything, you can (at least theoretically)?
<NobbZ[m]>
Everything I want to configure is configured as I want it.
<nij>
feffe[m]: Do you mean when you have to put an .rc file?
<nij>
for example?
<nij>
NobbZ[m]: Great!!!
<NobbZ[m]>
And as I have configured everything through Home-manager I can easily share and reproduce between my systems
<feffe[m]>
no, vs code live share needs to patch the binary afaik, or something along those lines at least
rardiol has joined #nixos
user_0x58 has joined #nixos
<nij>
NobbZ[m]: I'm quite envious
<nij>
hope I will be at that level soon
<feffe[m]>
so anything that needs to change something in the derivation after the fact can be a pain to work with, unless someones already prepared a derivation with that in mind
<nij>
feffe[m]: hmm.. haven't encountered that
<nij>
feffe[m]: Oh so if you want (and can) you can write a derivation to solve your problem?
<typetetris>
Looking at this poetry2nix stuff I think, thats how we should handle haskell in nix too.
<feffe[m]>
yeah, haven't done it much myself but I think the goto method would be to fork nixpkgs and make the changes needed
x39-nixos has joined #nixos
dbmikus has joined #nixos
dingenskirchen has quit [Quit: dingenskirchen]
dingenskirchen has joined #nixos
<x39-nixos>
could puke in buckets right now ... sitting here for minutes already and nothing happened yet whilst it took sometimes mere seconds :( never thought i would appreciate a system-crash ^^'
user_0x58 has quit [Ping timeout: 260 seconds]
<typetetris>
x39-nixos: Have you looked at the dmesg-output after the fact? Maybe looked at journalctl output of the system, it might write messages before it crashes. Also do you have other linux distros running on that system? Have there been kernel parameters you needed to run them crash free?
<nij>
feffe[m]: :quit
nij has quit [Quit: leaving]
<NobbZ[m]>
feffe nope, override or overrideAttrs is usually preferred
<{^_^}>
[nixpkgs] @worldofpeace merged pull request #94878 → nixos/tests: add test for firejail → https://git.io/JJXLI
<{^_^}>
[nixpkgs] @worldofpeace pushed 2 commits to master: https://git.io/JJ5OB
<manveru>
`plug = import ./vimrc/plugins.nix;` should be `plug = import ./vimrc/plugins.nix { inherit stdenv, writeText };`
<itshaydendev>
Should it?
<itshaydendev>
Ohhh yeah
<itshaydendev>
I see
<manveru>
oh, without the `,` :)
<manveru>
usually `cannot coerce a function to a string` means you aren't giving something you try to interpolate enough arguments to turn it from function to string
sangoma has joined #nixos
<exarkun>
How do I give a vm built with nixos-build-vms more space to which to unpack docker images?
<itshaydendev>
`plug = import ./vimrc/plugins.nix { inherit stdenv writeText };` got me `syntax error, unexpected '}', at /etc/nixos/packages/neovim/vimrc.nix:8:64`
<itshaydendev>
would it not be callPackage?
<jared-w>
`plug = import ./vimrc/plugins.nix { inherit stdenv writeText; };`, it needs a semicolon after writeText
<itshaydendev>
ah
<manveru>
oh... sorry
<manveru>
another copy&paste mistake :(
<itshaydendev>
No worries!
<manveru>
callPackage would work, if you had it in that file...
<itshaydendev>
It's building this time! \o/
<jared-w>
I only make that mistake like 20 times a week
<itshaydendev>
Ayyyyyy, it's working!
x39 has joined #nixos
<manveru>
awesome :D
<itshaydendev>
Time to go find all my plugins in vimPlugins lol
<itshaydendev>
Or prefetch the urls to add 'em myself
<{^_^}>
[nixpkgs] @peti pushed 2 commits to haskell-updates: https://git.io/JJ5Or
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JJ5OK
<jared-w>
That's honestly why I never fully "nixified" quite a bit of my dotfiles. I just use home-manager to symlink everything into the right directory and do that
MichaelRaskin has joined #nixos
x3963 has joined #nixos
CodeSpelunker2 has joined #nixos
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage2nix: update list of broken builds to fix evaluation »: https://git.io/JJ5OX
<x3963>
sooo ... having tried both staying in tty for minutes (no crash observed) and setting boot.crashDump.enable to true, i still cannot seem to get a proper info about what is going on
x39 has quit [Ping timeout: 245 seconds]
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JJ5OM
<x3963>
now reading through the interwebs, journalctl came up and it shows me a few distinct errors which ... i cannot really do anything with as google spits out many things but nothing useful
CodeSpelunker has quit [Ping timeout: 256 seconds]
<exarkun>
I guess there's always `qemu-img resize ...`
itshaydendev has quit [Quit: WeeChat 2.8]
x3963 has quit [Ping timeout: 245 seconds]
<{^_^}>
[nixpkgs] @Ma27 opened pull request #95446 → lib/modules: improve error-message for undeclared options if prefix contains no options → https://git.io/JJ5Ox
<AndroUser>
https://pastebin.com/x8xAXHnY this is the output of `journalctl -p 3` and something tells me that something with the coredump stuff ain't working with just those few options 😅 still not sure what the cause is, though the Hw error is kinda New in there (previously it complained about amd_kvm not being enabled)
<{^_^}>
[nixpkgs] @worldofpeace pushed 2 commits to master: https://git.io/JJ530
georgyo[m]1 has joined #nixos
alexherbo2 has quit [Ping timeout: 240 seconds]
ddellacosta has quit [Quit: WeeChat 2.8]
domogled has joined #nixos
rajivr has quit [Quit: Connection closed for inactivity]
<georgyo>
I recently noticed that NixFlakes has merged into the master branch. Which to me means it's fairly close (maybe a few months) to a release.
<georgyo>
While using the nix from the master branch, there is quite a few changes to behavior.
ddellacosta has joined #nixos
<worldofpeace>
georgyo: yes, I do believe there is a release planned of nix that will have these experimental features
<{^_^}>
[nixpkgs] @peti pushed 2 commits to haskell-updates: https://git.io/JJ53K
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JJ53X
<energizer>
how do i kill all running builds?
domogled has quit [Client Quit]
domogled has joined #nixos
<georgyo>
worldofpeace: Interesting. Is there a loose time line for that? It may change some plans for things I'm building at work.
alexherbo2 has joined #nixos
<symphorien[m]>
energizer: pkill -9 -G nixbld ?
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage2nix: update list of broken builds »: https://git.io/JJ539
<energizer>
symphorien[m]: sounds good, any reason i'll regret this?
<symphorien[m]>
you were compiling chromium and it was 99% done ?
x39 has joined #nixos
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/JJ537
<energizer>
nope didnt work
<symphorien[m]>
run it as root
<x39>
sooo ... great ... my mobile phones IRC chat is not even relyable enough to keep connected ... did somebody by chance respond to my IRC message with the pastebin link about journalctl output?
<energizer>
i did
domogled has quit [Client Quit]
<worldofpeace>
georgyo: I'm not sure there is a loose time line that I'm aware of, I'd have to ask Eelco in #nixos-dev
domogled1 has joined #nixos
<symphorien[m]>
there is a link to logs in the topic, x9
<symphorien[m]>
x39*
<x39>
(y)
alp has quit [Ping timeout: 272 seconds]
domogled1 is now known as domogled
domogled has quit [Client Quit]
domogled has joined #nixos
<georgyo[m]1>
worldofpeace: I see, this makes it a lot harder to plan, but we will have to make do.
<x39>
to now answer the question typetetris, journalctl output is here: https://pastebin.com/x8xAXHnY had no linux distro running on this specific combination of HW yet (besides using virtualbox) thus NixOS being the first running here (used to use Ubuntu at work for years at my old job)
cript0nauta has quit [Remote host closed the connection]
criptonauta_ has joined #nixos
<typetetris>
x39: Maybe try mcelog to see, what the hardware failures are
<worldofpeace>
georgyo: I did mention it to eelco in #nixos-dev for you. Not sure what exactly the new release will change for you, but any experimental feature shouldn't be relied on in any production environment just yet
kalbasit_ has quit [Ping timeout: 240 seconds]
<x39>
trying in a minute
<typetetris>
x39: Also try not starting any x server but stay on the console and see if the problems prevail there.
domogled has quit [Read error: Connection reset by peer]
domogled1 has joined #nixos
<x39>
they are UX related, that much i already suspect ... because on terminal, after 20 minutes nothing happened whilst in UX, 20 minutes without a system crash are kinda a luckshot
<georgyo[m]1>
worldofpeace: I don't plan to rely on (or use) the experimental features at all, however the master branch changes the behavior of all the other commands as well.
<exarkun>
How do I give a vm built with nixos-build-vms more memory?
x39 has quit [Remote host closed the connection]
<georgyo[m]1>
worldofpeace: similarly, `nix search` is very powerful and after the release I'll be telling users to do `nix-env -qaP | grep foo` again
AndroUser has joined #nixos
domogled1 is now known as domogled
domogled has quit [Client Quit]
<typetetris>
x39: So find the xservers error log and look that up. Maybe the wrong driver is loaded or some such. For me it is still in `/var/log/X.0.log` or some such.
domogled has joined #nixos
<typetetris>
Did you try to set `services.xserver.videoDrivers` to include `nvidia`?
<worldofpeace>
georgyo: ahh, understood
<{^_^}>
[nixpkgs] @FRidh pushed commit from @andersk to master « meson: Add upstream patch to stop failing on skipped tests »: https://git.io/JJ5sZ
domogled has quit [Client Quit]
domogled1 has joined #nixos
x39 has joined #nixos
AndroUser has quit [Ping timeout: 260 seconds]
xelxebar has quit [Remote host closed the connection]
<x39>
Well... My nixos Desktop dual boot is rendered useless thanks to those System crashes... So not sure if "more frustrating" is correct 😂😂
<jasom>
just because it's been a while since I've done this; what's the best way if I want to build a package that already exists in nixos, but with added buildInputs and make flags? I can just copy the whole expression and make an overlay, but maybe there's something better? (in this specific case it's building squashfs-tools with LZO support).
<niso>
bqv: i'll check them out. i'm currently considering on writing an overlay for nix, which converts the current-flakepath into an absolute one, i think that way i could do it without them?
quinn has joined #nixos
<{^_^}>
[nixpkgs] @tfmoraes opened pull request #95450 → python3Packages.vtk9: init at 9.0.1 → https://git.io/JJ5ZV
gv has quit [Quit: gv]
<niso>
bqv: ah, you were refering to git submodules?
domogled has quit [Client Quit]
Rusty1 has joined #nixos
domogled1 has joined #nixos
griff_ has quit [Quit: griff_]
domogled1 is now known as domogled
<bqv>
Yes
domogled has quit [Client Quit]
<niso>
bqv: i see, guess i could just store the flakes with the projects source instead. The only disadvantage would be maintenance/refactoring
ddellacosta has quit [Quit: WeeChat 2.8]
karaboshi has joined #nixos
<karaboshi>
hello, I want to add a python dependency to my shell.nix but it's not available in the nixpkgs. Where can I read / learn about making this work?
<_habnabit>
how is a nixos module supposed to execute e.g. chown in ExecStartPre? ${pkgs.coreutils.bin}/bin/chown gives me an error of "error: attribute 'bin' missing". /run/current-system/sw/bin/chown ? i thought systemd would use $PATH anyway
cole-h has quit [Quit: Goodbye]
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<manveru>
it's `pkgs.coreutils`, no `.bin` needed
zupo has joined #nixos
<_habnabit>
manveru, other packages have a bin attribute.. when do you use it?
<manveru>
when it's a split package, you almost never have to use it explicitly
<_habnabit>
infinisil, oh i can't use preStart because it does ExecStartPre without a + prefix.. i need the + because it sets User/Group on the service and the mkdir needs to run as root. i guess i could make a second unit for that
<{^_^}>
[nixpkgs] @jonringer merged pull request #92997 → librealsense: refactor to allow use of cuda, and prepare for python integration → https://git.io/JJOvH
<{^_^}>
[nixpkgs] @jonringer pushed commit from @freezeboy to master « librealsense,python3Packages.pyrealsense2: refactor to add cudaSupport and python bindings »: https://git.io/JJ5BD
diamondburned has quit [Read error: Connection reset by peer]
diamondburned has joined #nixos
veleiro` has joined #nixos
zupo has joined #nixos
justanotheruser has quit [Ping timeout: 260 seconds]
veleiro has quit [Ping timeout: 256 seconds]
zupo has quit [Ping timeout: 240 seconds]
zupo has joined #nixos
mbrgm_ has joined #nixos
mbrgm has quit [Ping timeout: 260 seconds]
mbrgm_ is now known as mbrgm
eoli3n has quit [Ping timeout: 246 seconds]
<W1lkins>
Is there a way to use a single configuration option from "master" (unstable?) with an otherwise stable (20.03?) configuration?
eoli3n has joined #nixos
justanotheruser has joined #nixos
<buffet>
hey, i rebuild my server recently, using --target-host, and i couldnt boot it afterwards anymore (failed to mount /mnt-root). it might have something to do with me using -I, is there any info on this? i cant do much testing rn sadly
<energizer>
W1lkins: i dont think so. you'd have to look in the code and see what that config option does, and the implement it yourself
<buffet>
i think you could add `<nixpkgs-unstable/nixos/modules/...>` to imports somewhere W1lkins
<buffet>
with nixpkgs-unstable = fetchTarball {...} { }