<thoughtpolice>
gchristensen: `nix-build -A hello && nix copy --to file://$PWD/cache-test ./result && nix run --store file://$PWD/cache-test nixpkgs.hello -c hello` seems to work OK for me
<thoughtpolice>
(from the root of the nixpkgs checkout, in my case)
justanotheruser has joined #nixos
acarrico has quit [Ping timeout: 276 seconds]
apache8080 has joined #nixos
ximun has quit [Ping timeout: 246 seconds]
ximun has joined #nixos
oida has quit [Ping timeout: 268 seconds]
<drozdziak1>
Does Nix know how to deduplicate CFLAGS? I'm battling a long argument list problem for some time now but I notice many flags are -I's that repeat 2-3 times
<gchristensen>
thoughtpolice: maybe I should ask you my question a bit more directly ... I'd like to run nix-build like this https://gist.github.com/grahamc/fa7b3cf853f1ab8daa71c0cf467fbb9c and give this nix-build a fake remote builder: where `ssh` in PATH is secretly a shim to call nix-store --serve --write with a second tempdir it can build "remotely" in
hlz has joined #nixos
ddellacosta has quit [Ping timeout: 265 seconds]
abathur has joined #nixos
jsoo_ is now known as jsoo
<thoughtpolice>
Oh, *that* one I would have no clue about. Interesting idea, though; is this for r13y or something?
andreas303 has quit [Remote host closed the connection]
andreas303 has joined #nixos
<jumper149>
That doesn't seem to fix it :(
ddellacosta has quit [Ping timeout: 265 seconds]
<jumper149>
tomberek: I am using this option, just once for the domain and then using the certificate in different services
tv has joined #nixos
<jumper149>
waaaait, maybe enableACME does exactly the same? :D
<tomberek>
jumper149: i'd check the generated nginx.conf and the file it is pointing to. Make sure it's properly formatted, etc. (it's easy to have extra blurbs from openssl accidentially in the file)
marcusr_ has quit [Remote host closed the connection]
<tomberek>
i use enableACME often and it JustWorks(tm)
marcusr has joined #nixos
<jumper149>
Ah yes now I remember the problem that I was having before dnsProvider and webroot are mutually exclusive (from security.acme.certs.domain)
<jumper149>
That's why I was saying that nginx tries an HTTP challenge
<tomberek>
check what is written to the nginx.conf file, the double quotes or string interpolation may be doing something odd
<jumper149>
Yes will do that now, one thing after another :D
justanotheruser has quit [Ping timeout: 272 seconds]
<matthewcroughan>
Mic92: do you have any nix-shells that could be used to automate qemu logic?
marcusr has quit [Read error: Connection reset by peer]
<rmcgibbo[m]>
omasanori: bisecting sage is going to take a looooong time
<rmcgibbo[m]>
godspeed
<matthewcroughan>
I wanna make a nix-shell that takes two arguments, does some logic with them and downloads the result of the first argument to the /nix/store.
marcusr has joined #nixos
<matthewcroughan>
then runs qemu with variable flags
<matthewcroughan>
ideally this shell would work the same under macos and any linux distro ;D
ahmed_elgabri has joined #nixos
zebrag has quit [Quit: Konversation terminated!]
<omasanori[m]>
<rmcgibbo[m] "omasanori: bisecting sage is goi"> Sure, it tooks 7 hours/step on my laptop and 10 minutes/step on a Xeon Silver machine in my institute (which I can use when there is no other jobs).
<omasanori[m]>
* latter
ml| has quit [Ping timeout: 264 seconds]
apache8080 has quit [Ping timeout: 276 seconds]
<rmcgibbo[m]>
heh
<enjoinedmot[m]>
I know someone who is working on qemu with nix
<rmcgibbo[m]>
nice
<matthewcroughan>
enjoinedmot[m]: got their shells? :D
<enjoinedmot[m]>
no... I don't even know the details, just noticed your chat... I could circle back with more info maybe
zupssss has joined #nixos
<tomberek>
can i get a review/second-look on 114432?
ahmed_elgabri has quit [Ping timeout: 240 seconds]
<tomberek>
omasanori[m]: i encounter that with tensorflowWithCuda builds (ouch)
<raboof>
if a nix store path is a root but I'm pretty sure it shouldn't be, what do I do to further diagnose? It's not linked directly in /nix/var/nix/gcroots/auto , but perhaps indirectly. pretty sure it's not part of any profile.
avaq has joined #nixos
jbetz has quit [Ping timeout: 260 seconds]
pingveno has quit [Read error: Connection reset by peer]
jbetz has joined #nixos
pingveno has joined #nixos
alexherbo2 has quit [Ping timeout: 240 seconds]
malook has quit [Read error: Connection reset by peer]
Khetzal_ is now known as Khetzal
<{^_^}>
[nixpkgs] @dotlambda pushed 197 commits to python-unstable: https://git.io/JtbVk
<sephii>
I’d like to remove old generations on a network with nixops. The help says `delete-generation` takes one argument GENERATION which is "number of the desired configuration". Should I put the number of the generation I want to keep, or the generations I want to delete?
Guest67960 has quit [Quit: authenticating]
Guest67960 has joined #nixos
Guest67960 is now known as JJJollyjim
<ikwildrpepper>
sephii: afaik the number of the generation you would like to delete
JJJollyjim has quit [Client Quit]
<ikwildrpepper>
yeah, looking at the code, it confirms that it is consistent with the name of the option and its description
<sephii>
ikwildrpepper: so if I have 50 generations and I only want to keep the latest one I need to execute the command in a loop right? There’s no shortcut like `+3` to keep 3 generations?
<ikwildrpepper>
sephii: haven't used the feature in a while, let me check quickly :)
ahmedelgabri has joined #nixos
<ikwildrpepper>
sephii: I thought there was a feature to do such a thing, but, if I look at the options and the code, it looks like the delete-generation only accepts int's currently, which means it can only delete single generations currently
zups has joined #nixos
<sephii>
ikwildrpepper: ok, I did it with a shell loop, thanks :)
<ikwildrpepper>
it uses internally the nix-env --delete-generations command, so it probably wouldn't be too complicated to add :)
<{^_^}>
[nixpkgs] @dotlambda pushed 917 commits to staging: https://git.io/JtbVj
grumble has quit [Quit: K-Lined]
<yaymukund>
anyone had luck setting up systemd-boot w a luks-encrypted volume?
<yaymukund>
my nixos install won't boot. i've rebooted via install media and mounted everything as it was before i ran nixos-install. how can i reinstall it with an updated configuration.nix?
<aforemny>
,locate stropts.h
<yaymukund>
ooo, i think i just run nixos-install again
<{^_^}>
Found in packages: zig, z88dk, framac, zfs.dev, musl.dev, boomerang, emscripten, zfsUnstable.dev
<aforemny>
jimkooch: Note that this is only a rough draft. You'll want to remove unused dependencies from buildInputs, update the configurePhase and meta section to turn it into something usable.
<woffs>
Is there a reason why tor-browser-bundle-bin is still at 10.0.9 (while tor-browser-linux64-10.0.9_en-US.tar.xz not downloadable)? Recent version is 10.0.12.
<qyliss>
yaymukund: nix-collect-garbage -d will only clean up your user profiles if run as non-root
<qyliss>
to gc _everything_ you need to run it as root, but it doesn't require it because it can still find some garbage even if it doesn't have root privs
<{^_^}>
[nixpkgs] @adisbladis opened pull request #114463 → emacs.pkgs.telega: Apply patch to always prefer fixed store path to server → https://github.com/NixOS/nixpkgs/pull/114463
bitmapper has joined #nixos
avo has joined #nixos
avo has quit [Client Quit]
risson_ is now known as risson
thc202 has joined #nixos
m4ts has quit [Ping timeout: 264 seconds]
ddellacosta has joined #nixos
fendor has joined #nixos
ddellacosta has quit [Ping timeout: 264 seconds]
m4ts has joined #nixos
m4ts has quit [Ping timeout: 240 seconds]
m4ts has joined #nixos
brown121407 has joined #nixos
<sterni>
Graf_Blutwurst: seems like sbt fails to build with jre = jre8;
<sterni>
err nvm
<brown121407>
Hi! I installed MS Teams from Nix (I'm using just the package manager; the host operating system is Guix), but when I try to start it it says: /home/seamas/.nix-profile/bin/teams: error while loading shared libraries: libffmpeg.so: cannot open shared object file: No such file or directory
<sterni>
Graf_Blutwurst: that shell.nix works perfectly fine for me
<brown121407>
I tried with both ffmpeg adding ffmpeg to the Nix profile, but it didn't help.
<brown121407>
Has anyone else encountered this issue?
<jdelStrother>
Is pkgs/development/ruby-modules/runtests.sh broken for everyone or just me? I get an error about "called with unexpected argument 'stdenv'", but it seems like it's been broken for a very long time
<sterni>
maralorn: did you have a look at haskellPackages.cabal-install yet?
<sterni>
maralorn: seems like the 8.10 set doesn't like the new version from stackage-nightly and throws
<sterni>
i. e. assertion ((super).cabal-install.version == "3.2.0.0") failed at configuration-ghc-8.10.x.nix
<maralorn>
sterni: As a matter of fact I have a fix that I am testing locally now. And I accidentally commited that fix in the above push so I needed to amend the commit.
<sterni>
maralorn: nice, I've been meaning to have a look at it, but was too tired yesterday
<maralorn>
sterni: My problem is that from my experience I don‘t know if we should build cabal-install 3.4.0.0 for ghc 8.10 or stay with 3.2. But I am trying to build 3.4 now and see how it goes.
meh` has joined #nixos
zupo has joined #nixos
aexoxea has joined #nixos
aexoxea has quit [Client Quit]
aexoxea has joined #nixos
exoaoxe has joined #nixos
<sterni>
maralorn: the removal of sandbox and v1-sdist is pretty substantial I guess
exoaoxe has quit [Client Quit]
dev_mohe has quit [Quit: dev_mohe]
jdelStrother has quit [Quit: Ping timeout (120 seconds)]
<maralorn>
sterni: Yeah, so all the more a reason to update, right?^^
bnjmnt4n[m] has joined #nixos
<sterni>
maralorn: I guess we might as well if we want to have 3.4 in 21.05 so we notice any issues as early as possible on unstable?!
<maralorn>
sterni: I don‘t know. I don‘t expect a lot of breakage tbh.
bdju has quit [Read error: Connection reset by peer]
<maralorn>
But right now I am having trouble with getting it to build.^^
bdju has joined #nixos
<maralorn>
Because it needs a newer random, and because everyone loves property tests, if you bump random you rebuild the tests of the wohl ecosystem …
<sterni>
:|
<sterni>
I think nothing is as agonizing as haskell building all 500 modules of a package _again_ for the tests
<Taneb>
If it's packaged properly it shouldn't
<Taneb>
However nixpkgs' Haskell package sets builds all the packages both without and with profiling information
<maralorn>
Taneb: What does it have to do with profiling?
<simpson>
sterni: The npm generate.sh is agonizing if you have a slow network. There might be other subsystems that need to check every package like that, but IMO npm's the worst.
<Taneb>
The "building all 500 modules twice" thing sterni mentioned
<maralorn>
Oooh, it does that??
<Taneb>
Yeah
<sterni>
Taneb: it also happens when I build stuff locally with cabal-install
<sterni>
seems to be a default setting
<Taneb>
sterni: huh
<maralorn>
I thought sterni was talking about, when you compile the project successfully but a test breaks, and then you do something to fix the test but you always need to recompile because it's in the same derivation …
<sterni>
maralorn: that as well, but also all modules are compiled twice with tests
<Taneb>
Is this your packages or someone else's?
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<rauno>
When deploying with nixops, does the nix-serve provided http subtituter must be in nix.conf or it can be defined in nixops parameters also? It doesnt seem to work with --option defining
aexoxea has quit [Quit: Goodbye!]
<Reventlov>
I want to modify a checkPhase using an overlay to add some flags to some command line, but this checkPhase references other stuff from the package (and, in particular, a pythonEnv defined with some let ... in)
<Reventlov>
How can I do that, easily ?
slby[m] has joined #nixos
<slby[m]>
What does skipping suscpicious writable file mean?
<sterni>
Reventlov: that's not possible without modifiying the derivation, if you want something like that you have to create a similar (if not equal) pythonEnv yourself
<Ericson2314>
the nixpkgs cross stuff doesn't use any --sysroot right now, because it works the same way as the native stuff
<Reventlov>
sterni: meh, ok, thanks :)
<Ericson2314>
My favorite is the LLVM cross, where LLVM Clang compiler-rt libc++ are all built separately. If GCC would do the same, we could radically simplify cc-wrapper
<sterni>
Reventlov: I wonder if it'd be feasible to just use pythonEnv as nativeBuildInput as well instead of python
<sterni>
would avoid the interpreter confusion
<sterni>
but could be detremental to cross compilation possibly? not sure
ddellacosta has joined #nixos
dev_mohe has joined #nixos
jimkooch has joined #nixos
m4ts has quit [Ping timeout: 256 seconds]
<jimkooch>
I lost my $USER .nix-profile symlink to store. How do I fixed this? Is deleting my username from configuration.nix the only way?
m4ts has joined #nixos
<jimkooch>
so .nix-defexpr and .nix-profile
<jimkooch>
have broken links in them... i read a howto to delete user in configuration.nix, but before is there an easier way to re-link my user to store?
ddellacosta has quit [Ping timeout: 276 seconds]
<clever>
jimkooch: defexpr is automatically updated when using nix-channel
<clever>
jimkooch: and .nix-profile should repair itself when you use nix-env
<clever>
but if the old state was lost, then it wont have the old things you had installed
<jimkooch>
clever: ok.. after `nix-env -i xxx`, my ~/.nix-profile/ is re-link. But my ~/.nix-defexpr/channels is still broken link. How to fix this broken link?
<jimkooch>
I have androidstudio-4.4.1 which I launch via `nix-shell -p android-studio -I nixpkgs=https://github.com/NixOS/nixpkgs/archive/7138a338b58713e0dea22ddab6a6785abec7376a.tar.gz --run android-studio`. How do, I instead install that and available in my profile rather than nix-shell as shown?
meh` has quit [Ping timeout: 265 seconds]
<kunrooted[m]>
how am I supposed to tweak and patch suckless utilities in nixOS while using home-manager? I've heard I can provide config.h as a patch or something, but, any examples on how to do so?
<avn>
Folks! Anyone use nix+cuda? (on ubuntu, not on nixos)
vidbina has quit [Ping timeout: 276 seconds]
<rednaZ[m]>
Also, am I the only one who thinks identifying a package by its name in https://nixos.org/manual/nixpkgs/stable/#sec-allow-unfree is a stupid idea? Names can change and, worse, are not unique. What would be a better alternative?
<{^_^}>
[nixpkgs] @peti pushed 436 commits to haskell-updates: https://git.io/JtbpJ
Lorenzoas3ii[m] has quit [Quit: Idle for 30+ days]
a7p[m] has quit [Quit: Idle for 30+ days]
Hurttila has quit [Ping timeout: 240 seconds]
jimkooch has quit [Quit: leaving]
chisui has quit [Ping timeout: 240 seconds]
<tomberek>
avn: i've fought the cuda battle, what do you need?
jimkooch has joined #nixos
jimkooch has quit [Client Quit]
zupo has joined #nixos
awmv has joined #nixos
<avn>
tomberek: How I can hint application, where it can load libcuda.so?
<numkem>
I've been setting up prometheus on my homelab in nixos rather than docker and while I still set services.prometheus.exporters.node = true it doesn't create the systemd unit
afreak_ has quit [Ping timeout: 264 seconds]
<tomberek>
avn: i've just dealt with this yesterday. I find myself adding this to the environment and it seems to work well (fixup the versions as needed): LD_LIBRARY_PATH="${pkgs.stdenv.cc.cc.lib}/lib:${pkgs.cudatoolkit_11_0}/lib:${pkgs.cudnn_cudatoolkit_11_0}/lib:${pkgs.cudatoolkit_11_0.lib}/lib:${pkgs.linuxPackages.nvidia_x11}/lib";
<tomberek>
on ubuntu you may want to point it at the native libraries if you have them installed
respawn_ has quit [Read error: Connection reset by peer]
<viric>
How would one use the nix cross compilers (or nix compilers) targetting a usual sysroot?
<avn>
tomberek: yup. At least probbaly we have two problems overlapping -- one with libcuda lookup, second one was libgst_foo_cuda.so itself (is a GStreamer plugin using cuda)
<viric>
/lib/ld-linux.so, etc.
sangoma has quit [Ping timeout: 246 seconds]
<tomberek>
avn: i know i've gotten the gstreamer CUDA working, it took a few more paths... but i do recall it worked (though sadly for my video workflow it wan't much of a speedup)
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<tad-lispy[m]>
Hello! I'm proud to finally set up NixOS on my laptop to a working state. There are few challenges still. One is multi display setup. I use Gnome on Wayland and have two monitors (built-in and external). In Gnome settings I set scaling on the external to 200%. It is also my primary monitor. Now GTK apps look correct, but non-GTK apps (Firefox, Electron based apps, OBS Studio) are scaled incorrectly on the built-in screen.
<tad-lispy[m]>
Everything is too big.
<tad-lispy[m]>
On Ubuntu with X I got good results by enabling fractional scaling, even though I only use integer values. Somehow it scales all apps correctly.
i1l has quit [Quit: Quit]
i1l has joined #nixos
<sterni>
tad-lispy[m]: non-GTK in wayland or xwayland applications?
davidv7 has quit [Read error: Connection reset by peer]
<tad-lispy[m]>
How do I know?
strobelight has joined #nixos
<sterni>
checking the process monitor if they are child processes of Xwayland is what I always do
<sterni>
idk if there's a better way
<tad-lispy[m]>
👀
<dutchie>
my favourite way to check wayland v xwayland is to launch xeyes and see if they follow my pointer over that window
<sterni>
gotta remember that one
<tad-lispy[m]>
I have no xeyes. What's the package?
<sterni>
tad-lispy[m]: xorg.xeyes
<tad-lispy[m]>
They do follow.
<tad-lispy[m]>
And you are right - they don't follow over alacritty, which is not GTK+, right?
<sterni>
tad-lispy[m]: okay, DPI issues with Xwayland are relatively common
<sterni>
tad-lispy[m]: try having a .Xresources somewhere with the content `Xft.dpi: 96`
<sterni>
tad-lispy[m]: and do an xrdb -load <file>
<tad-lispy[m]>
Done. Nothing happened. Do I need to restart something?
<sterni>
tad-lispy[m]: you can try it anyways just to be sure, close all X11 applications, kill Xwayland and restart them again
<sterni>
tad-lispy[m]: alacritty has native wayland support, does it scale correctly?
<tad-lispy[m]>
Ok, be back...
<tad-lispy[m]>
Yes, it does.
<tad-lispy[m]>
Done, but I see no difference.
<tad-lispy[m]>
I put the file in `~/.Xresources`.
<sterni>
tad-lispy[m]: did you xrdb -load it?
<tad-lispy[m]>
Yes, before killing Xwayland.
<tad-lispy[m]>
When I killed it the DM died and I had to log in again.
<sterni>
I assume the Xresources fix is not it then
<Yaniel>
for a moment I thought you were talking about D&D there
<tad-lispy[m]>
Haha
<tad-lispy[m]>
No, she's fine.
<dutchie>
does xwayland load ~/.Xresources by default? i vaguely remember having to put an explicit `xrdb -load ~/.Xresources` in my .xsession/.xinitrc files
<sterni>
tad-lispy[m]: at least we now know what the issue is right :/
<sterni>
tad-lispy[m]: the thread mentions that that is only available when running inside Xwayland for some reason
<sterni>
tad-lispy[m]: so you'd need to get an X11 terminal temporarily
<sterni>
i guess ?!
<slby[m]>
Yaniel: Thanks I just copied the contents of the shell script and then it recognized it had make.
<{^_^}>
[nixpkgs] @adisbladis opened pull request #114486 → gst_1_all.gst-plugins-{bad,good}: Remove gstreamer dev output from closures → https://github.com/NixOS/nixpkgs/pull/114486
<tad-lispy[m]>
Finished reading the issue and not sure where to go from here.
<tad-lispy[m]>
One thing that stands out is this `org.gnome.mutter` schema missing. I ran into this before.
afreak_ has joined #nixos
<tad-lispy[m]>
As I said on Ubuntu I had same problems and then enabled fractional scaling. Even with same, integer values all apps were scaled correctly.
<tad-lispy[m]>
So I was trying to do this, following Arch Wiki.
vidbina has joined #nixos
malook has joined #nixos
<tad-lispy[m]>
Any ideas about this missing schema?
<sterni>
tad-lispy[m]: have you tried setting the schema from within xwayland, e. g. by temporarily using xterm for that?
<tad-lispy[m]>
You mean same command (`gsettings set org.gnome.mutter experimental-features '["scale-monitor-framebuffer"]'`) but from `xterm` running in a Wayland session?
<tad-lispy[m]>
Same effect, no such schema.
<sterni>
okay, unfortunate
jimmiehansson has quit [Remote host closed the connection]
<sterni>
the issue mentioned something about it only being available in X?!
<tad-lispy[m]>
Interestingly the xterm is absolutely tiny on the 4K monitor.
AmandaC_ has quit [Ping timeout: 260 seconds]
<avn>
tomberek: one more mandatory variable -- GST_REGISTRY_1_0=/dev/null
<avn>
may be eveno worhth to make "no registry" is a default on nixos
<tad-lispy[m]>
As far as I understand someone says that `xrandr` has read-only access to Xwayland. So I guess that's why you can't set the scaling?
AmandaC_ has joined #nixos
<radvendii>
is there a way to build a Mac .app bundle with nix? in particular collecting all of the libraries the binary depends on would be nice. a bit like the windows dll hell problem we run into when cross-compiling for windows
sangoma has joined #nixos
<tad-lispy[m]>
As for the mutter schema considering that Mutter is a wayland compositor it wouldn't make sense to be only available in Xorg, right?
<radvendii>
I got around it on windows by making all the libraries static but that's not working on mac
<NieDzejkob>
So, openjdk15's build is broken. Hydra has a commit ID for last working build, but the range of commits between working and broken is huge and includes a merge between staging and staging-next. Is there a way to compute which commits actually changed the derivation?
alexherbo2 has quit [Quit: Ping timeout (120 seconds)]
sangoma has joined #nixos
ahmedelgabri has quit [Ping timeout: 265 seconds]
<tad-lispy[m]>
Switching. It takes time (unexpectedly). Will get back once I restart.
<rmcgibbo[m]>
I've been running it with that patch rebased onto 20.09, so maybe the different build environment between that and unsable is bad. I'll take a look.
<drozdziak1>
How do I wrap PYTHONPATH in a python package?
<drozdziak1>
I have a shell that includes tmuxp and docker-compose, but docker-compose pollutes the path for tmuxp
<drozdziak1>
and tmuxp gets an error because it looks for pyaml in py3 territory
<shla>
rmcgibbo[m], have you tried it with linuxPackages_latest? w/o latest it works fine.
justanotheruser has joined #nixos
<NieDzejkob>
drozdziak1: grep for 'wrapProgram.*PYTHONPATH' for examples
justan0theruser has quit [Ping timeout: 272 seconds]
samhza has quit [Remote host closed the connection]
samhza has joined #nixos
<cocreature>
Hi, I have a case where I want to patch nixpkgs via a patch file rather than modify a derivation via an override (because the thing I want to change is not exposed and I would need to duplicate a lot of the current derivation). What’s the best way of applying a patch to the result of fetchTarball?
qbit has quit [Quit: leaving]
<hexa->
override postPatch
<hexa->
or patches really
rajivr has quit [Quit: Connection closed for inactivity]
sangoma has quit [Ping timeout: 240 seconds]
<cocreature>
hexa-: oh does the builtin have a patches attribute? I couldn’t see it in the docs
<ajs124>
cocreature: ah. hm. you could append something to preConfigure instead of overwriting? that's not significantly less messy, but maybe slightly?
<cocreature>
ajs124: yeah I can find a way to make it work more or less reasonably but a patch file to just patch the actual buildMk seemed clearer here as a short term solution. long-term I’d just like to upstream something into nixpkgs to make that easier
<ldlework>
Uncompressing NVIDIA Accelerated Graphics Driver for Linux-x86_64 460.39/nix/store/cfxvfmsgqqp92cish7ahf8gdylmpp4yk-NVIDIA-Linux-x86_64-460.39.run: line 699: /build/makeself.bysINdYB/xz: No such file or directory
<ldlework>
/nix/store/cfxvfmsgqqp92cish7ahf8gdylmpp4yk-NVIDIA-Linux-x86_64-460.39.run: line 690: /dev/tty: No such device or address
<ldlework>
/nix/store/rf2s4dz3g8p8n4p306k2nxg97411kixy-builder.sh: line 10: 6 Terminated sh $src -x
<ldlework>
Something to do with this maybe...
<{^_^}>
[nix] @Ericson2314 opened pull request #4580 → Restore now-working build-remote-content-addressed-fixed test → https://github.com/NixOS/nix/pull/4580
<le0taku>
I have noticed that something on NixOS seems to reset my home directory permissions on every reboot. Does anyone here know what could be the cause of this?
malook has quit [Quit: malook]
sangoma has joined #nixos
saschagrunert has quit [Remote host closed the connection]
<ldlework>
builder for '/nix/store/l4al92fndd3qkpv6pn73s79hrqz75blr-sddm-sugar-light.drv' failed with exit code 1
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « git-annex: update to new version 8.20210223 »: https://git.io/JtNqz
redmp has joined #nixos
hyiltiz has joined #nixos
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage2nix: update list of broken packages to fix evaluation errors on Hydra »: https://git.io/JtNq1
<ldlework>
oh thank god it just needed a full reboot
<tpw_rules>
where can i find information on writing an update script?
sangoma has quit [Ping timeout: 240 seconds]
ahmedelgabri has joined #nixos
Qwerky has joined #nixos
hiro98 has joined #nixos
vidbina has joined #nixos
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « hackage2nix: update list of broken packages to fix evaluation errors on Hydra »: https://git.io/JtNYD
<tpw_rules>
oh sure, if this is just to use it personally then something like that would work
<tpw_rules>
what i've done in those sorts of circumstances is just get the old derivation from the commit history
<jonreeve[m]>
How do I put that derivation in the file above? I don't think I can just copy-paste it in, can I?
<tpw_rules>
one way would be to remove cryptopp from the argument list then prefix the mkDerivation with: let cryptopp = import ./my_download_cryptopp.nix; in
<{^_^}>
jart/cosmopolitan#56 (by heikkilevanto, 8 hours ago, open): Segfault on Debian Linux, after adding a file to redbean
<rmcgibbo[m]>
shla: I think the issue is that when the ena PR was merged, linuxPackages_latest was 5.10, and it built okay. But now it's linux 5.11 and it doesn't build.
vrinek has quit [Ping timeout: 272 seconds]
<yaymukund>
`nix-shell '<home-manager>' -A install` - does this import home-manager and then run its `install` command, which presumably copies or symlinks things to your home directory?
<yaymukund>
I'm on day 1 of nix (already got it running! so cool!), so just trying to gain an intuition :)
<NieDzejkob>
at face value, it takes the install -A ttribute of the home-manager channel and gives you a shell with the dependencies of install available
<NieDzejkob>
but there's a shellHook, so it can also run some other commands. Not sure if it actually does, I don't use home-manager
<yaymukund>
ah gotcha
<yaymukund>
is there an alternative, or do you just maintain dotfiles as you would in a non-nix distro?
<praduca>
hey, anyone here use treesheets? i'm been unable to use it, it freezes at satart showing an alert sayng that an assertion failed... it wont even shut down
<praduca>
don't know if it is something with my installation, but nix is all about this being not relevant, so... if someone confirms that it is broken, i will fill an report
justanotheruser has joined #nixos
<cole-h>
Might as well file it anyways.
strikerlulu has quit [Ping timeout: 264 seconds]
justan0theruser has quit [Ping timeout: 272 seconds]
<{^_^}>
[nixpkgs] @peti pushed to haskell-updates « haskell-brittany: disable the freakin' test suite »: https://git.io/JtNlw
ahmedelgabri has joined #nixos
aleph- has joined #nixos
ahmedelgabri has quit [Ping timeout: 246 seconds]
apache8080 has joined #nixos
philr has quit [Ping timeout: 272 seconds]
cole-h has quit [Ping timeout: 265 seconds]
<jumper149>
greizgh++
<{^_^}>
greizgh's karma got increased to 1
<tad-lispy[m]>
Regarding the previously discussed issue with scaling Xwayland apps on monitors with different DPI, I've opened an issue here: https://github.com/NixOS/nixpkgs/issues/114514
<{^_^}>
#114514 (by tad-lispy, 4 minutes ago, open): No such schema “org.gnome.mutter”
erasmas has quit [Quit: leaving]
fendor_ has quit [Read error: Connection reset by peer]