<bqv[m]>
Ankhers: ah, gotcha. yeah, either use the binaries within the same FHS env, or if you want to use them outside, build a derivation for them or manually patchelf them
<immae>
bqv[m]: you mentionned flakes above, is it something that is already (partially) implemented, or is it only draft at the moment?
<Ankhers>
bqv[m]: I figured that would be the case. Thanks.
<bqv[m]>
immae: i've been trying to work that out myself. as i understand it, it's actually possible to use them, but it's still experimental so subject to breakage?
<immae>
ok
<immae>
It’s actually interesting as far as I understand the ideas
<bqv[m]>
i'm going to try and switch my desktop to using flakes, i'll let you know how i go, i guess
<immae>
Thanks
<{^_^}>
[nixpkgs] @petabyteboy opened pull request #82361 → jetbrains.jdk: build from source → https://git.io/Jv6LV
<manveru>
afaik nixos still won't build with nixFlakes because nixos-option won't compile against it :|
<bqv[m]>
in some of my digging i thought i came across a fix for getting all those utils to work
<bqv[m]>
no idea where though
<immae>
manveru: I don’t mind bugs, I’m just trying to get started somewhere for now
zeta_0 has quit [Quit: rcirc on GNU Emacs 26.3]
<infinisil>
immae: Pretty much unchanged actually
Rusty1 has quit [Read error: Connection reset by peer]
<qyliss>
It's also worth noting, btw, that twice as many PRs have been merged this week than were opened.
<qyliss>
So we are making progress.
<infinisil>
Rounded still 106/186 days
<immae>
ok thanks infinisil
<infinisil>
I'll share the script in a bit
<immae>
qyliss: I can see that, it’s actually quite impressive
learnday20 has joined #nixos
<qyliss>
At this rate we'll be at 0 PRs in nine weeks :P
<qyliss>
(That's not going to happen of course, but it is definitely getting better)
mexisme_ has quit [Ping timeout: 256 seconds]
o1lo01ol1o has quit [Ping timeout: 240 seconds]
<Vikingman>
someone can help me fix this
<Vikingman>
/bin/bash: bad interpreter: No such file or directory
<qyliss>
This month, 1795 PRs have been merged, and 470 have been opened.
<qyliss>
vikingman: NixOS doesn't have a /bin/bash
<qyliss>
You should use /usr/bin/env bash
<immae>
Thanks infinisil
aleph- has joined #nixos
<immae>
qyliss: well, if noone opens any PR in the next month, then we’re good :D
<immae>
(actually, I think "init" PR’s will still remain, mostly)
zdm has quit [Quit: WeeChat 2.7.1]
<qyliss>
init PRs can be hard to review if you don't use the software
<immae>
yes
<Vikingman>
thanks qyliss
<Vikingman>
it fixed the script
<qyliss>
otoh, if I saw a review from a non-committer, and the code looked okay at a glance, and I could verify that the software built, I'd merge an init PR
<qyliss>
A problem for me, at least, is that when I go looking to merge some PRs I generally go for volume
<immae>
what do you mean by "volume"?
<qyliss>
I mean that if something isn't immediately ready for merging I just look for something else that is
<qyliss>
because I'm trying to just merge as many PRs as possible
<immae>
ok
<qyliss>
(e.g. I just merged about 10)
<qyliss>
If I stopped to review every PR, I might not get any merged.
<qyliss>
My current balance is way off. I should be reviewing more than I am. But it's tricky to get right.
<{^_^}>
[nixpkgs] @alyssais pushed commit from @romildo to master « matcha: 2020-03-01 -> 2020-03-11 »: https://git.io/Jv6tm
<{^_^}>
[nixpkgs] @bhipple opened pull request #82367 → [20.03] rainicorn: remove broken package that has been abandoned upstream → https://git.io/Jv6t3
zeta_0 has joined #nixos
<{^_^}>
[nixpkgs] @alyssais merged pull request #82367 → [20.03] rainicorn: remove broken package that has been abandoned upstream → https://git.io/Jv6t3
<{^_^}>
[nixpkgs] @alyssais pushed commit from @bhipple to release-20.03 « rainicorn: remove broken package that has been abandoned upstream »: https://git.io/Jv6tc
<adisbladis>
This is completely unscientific, but eval time goes up quite drastically
<bqv[m]>
fair enough
<danderson>
if I install a package and its closure contains a package with vulnerabilities, nix will refuse to install it, right? (unless I tell it to ignore the vulns)
<qyliss>
I believe so.
<danderson>
hm. Need more nixos boxes to test.
<adisbladis>
bqv[m]: From ~7.4-7.5 seconds > ~8.0-8.4 seconds (I know this interval is quite large)
<adisbladis>
bqv[m]: With that in mind I don't really think it's worth it.
<bqv[m]>
yeah no worries, i'll note and close then
<adisbladis>
It's a shame though :/
<danderson>
confirmed, nixos also refuses to build a closure with insecure packages \o/
claudiii has quit [Quit: Connection closed for inactivity]
<{^_^}>
[nixpkgs] @cdepillabout pushed 2 commits to haskell-updates: https://git.io/Jv63O
kenran has joined #nixos
domogled has joined #nixos
<bqv[m]>
Can someone on unstable check the version of nheko quick for me?
<bqv[m]>
Actually…
<bqv[m]>
> pkgs.nheko.version
<{^_^}>
"0.6.4"
<hoijui>
could it be interesting for sourcehut, to go into a strong collaboration with [git-bug](https://github.com/MichaelMure/git-bug), to basically use it as a back-end to stroe the issue/todo data directly in the repo?
<gentauro>
I begin to the the following error message after my latest upgrade to the stable 19.09 channel: "warning: ignoring the user-specified setting 'auto-optimise-store', because it is a restricted setting and you are not a trusted user"
<gentauro>
is there a way to disable that warning?
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
o1lo01ol1o has joined #nixos
civodul has joined #nixos
m4ts has quit [Quit: bye]
o1lo01ol1o has quit [Ping timeout: 265 seconds]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
Tucky has joined #nixos
<NobbZ[m]>
gentauro: disable that setting for your user or make the user trusted.
<NobbZ[m]>
`nix.trustedUsers` can do the latter for you.
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
<gentauro>
NobbZ[m]: I would rather disable the feature (I can read something about create hard links instead of symbolic amirite?)
<gentauro>
NobbZ[m]: if I look in `/etc/nix/nix.conf` the `auto-optimise-store = false` is already disabled
<gentauro>
so why all the warnings?
malSet has quit [Read error: Connection reset by peer]
<NobbZ[m]>
Optimising the store is doing exactly that. Hardlinking files in the store that are identical.
<symphorien>
ehmry: there is dex, but it does not test icons
<gentauro>
NobbZ[m]: so you are saying I should enable it?
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
<gentauro>
like this `nix.extraOptions = "auto-optimise-store = true";` in the `/etc/nixos/configuration.nix`?
<NobbZ[m]>
I do it by hand regularly. I like to have control. Though in your case it seems as if you have enabled that setting for your user, not system wide. That's why the warning is there.
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has quit [Read error: Connection reset by peer]
<gentauro>
but it's disabled cos I got some strange errors when rebuilding and switching
<gentauro>
(I think it was the generation of the /etc/nix/nix.conf file)
kenran_ has joined #nixos
kenran has quit [Quit: leaving]
<LinuxHackerman>
gentauro: how do you sandbox your nix-shells?
o1lo01ol1o has quit [Remote host closed the connection]
<gentauro>
LinuxHackerman: always like this -> `alias nsp='nix-shell --pure'`
<srhb>
gentauro: The size of that is just about the size of the store since it contains hardlinks to every path in order to do the dedup via "optimise"
<srhb>
But it's not _extra_ space used.
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
<srhb>
There's nothing _wrong_ with just doing occasional optimise instead of automatic though, it just means you'll be using more space between optimise runs, and doing more writes.
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
o1lo01ol1o has joined #nixos
malSet has joined #nixos
chloekek has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
<gentauro>
srhb: OK, I will give it a try
malSet has quit [Read error: Connection reset by peer]
bahamas has quit [Ping timeout: 268 seconds]
lsix has joined #nixos
greymalkin has joined #nixos
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
<LinuxHackerman>
gentauro: that doesn't do any sandboxing, only uses a clean process environment
<gentauro>
srhb: :o, it saved my 5 GB of SSD !!!
<gentauro>
srhb: ME GUSTA !!!
malSet has joined #nixos
<srhb>
gentauro: I think I'm currently saving like.. 100G or something. I'll check :P
<srhb>
gentauro: Haskell derivations are especialy amenable to optimisation due to many duplicated paths. If you gc less often you'll probably see the same thing, and save writes :)
<srhb>
(If you have the space for it, that is)
<{^_^}>
[nixpkgs] @vcunat pushed commit from @tokudan to master « firefox-bin: 73.0 -> 74.0 (from PR #82257) »: https://git.io/Jv6cs
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has quit [Read error: Connection reset by peer]
psyanticy has joined #nixos
<{^_^}>
[nixpkgs] @NeQuissimus pushed 4 commits to release-19.09: https://git.io/Jv6cP
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
<{^_^}>
[nixpkgs] @NeQuissimus pushed 5 commits to release-20.03: https://git.io/Jv6c1
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
jason_ has joined #nixos
hoijui has quit [Ping timeout: 272 seconds]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
helozjisky has quit [Ping timeout: 256 seconds]
nschoe has joined #nixos
<{^_^}>
[nixpkgs] @FRidh opened pull request #82392 → octoprint: 1.3.12 -> 1.4.0, use python3 → https://git.io/Jv6ch
<Cadey>
i've noticed that the docker-containers family of options isn't documented in the nixos manual, how would I go about documenting it in the manual?
<{^_^}>
[nixpkgs] @FRidh closed pull request #82294 → Octoprint 1.4.0 with Python 3 (WIP; help needed) → https://git.io/JvKoX
sigmundv has joined #nixos
<manveru>
Cadey: similar to the containers doc i'd imagine
karantan has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<nakkle>
Hi there! I'm having a weird problem, where after a rebuild that disabled `virtualization.libvirtd`, there's still some lingering libvirt dnsmasq process. I can kill this process, but after a reboot they are restarted by systemd, belonging to libvirtd.service. I can't figure out how that's even happening because there are no .system files belonging to libvirts (as you'd expect when you disable the nixos option). Does anyone know what
<{^_^}>
#82083 (by sylvainf911, 3 days ago, open): kvm module exception and VirtualBox modules issue in linux 4.19.108
<emilsp>
I am actually quite enjoying my time off from testing across far too many platforms for once :)
<genesis>
alias Massoud="echo Massoud: commandant not found"
<emilsp>
But if I was in a dire need of a working KVM, I'd just reboot in an earlier state or try out an unstable kernel.
cmacrae` has quit [Read error: Connection reset by peer]
cmacrae` has joined #nixos
<ottidmes>
emilsp: I believe I saw someone else yesterday with a similar KVM issue, they used 4.19.107, in which it still works
misterwhatever has joined #nixos
<emilsp>
how would I specify that version in my nix config?
misterwhatever has quit [Remote host closed the connection]
<adisbladis>
emilsp: Overall I'd highly recommend running the latest kernel on Ryzen.
<ottidmes>
emilsp: if you can just upgrade to the latest kernel, definitely do that, but otherwise, the easiest way to get an older kernel is AFAIK to pin nixpkgs to a commit that still had it at that version
kolbycrouch has quit [Quit: Leaving]
o1lo01ol1o has quit [Remote host closed the connection]
<exarkun>
Hmm it works without `inherit defaultCrateOverrides` which worked before ... but then I refactored
<{^_^}>
[nixpkgs] @B4dM4n opened pull request #82397 → openjfx: respect NIX_BUILD_CORES in nested cmake → https://git.io/Jv6lr
<ottidmes>
emilsp: Just in case you need it, this could be how you downgrade to the kernel 4.19.107: { boot.kernelPackages = (import "${builtins.fetchGit { url = "https://github.com/NixOS/nixpkgs.git"; ref = "refs/heads/nixos-20.03"; rev = "9239f78bbf7a7359ea55c858ca9a94bb977dd26f"; }}" { }).linuxPackages_4_19; }
<exarkun>
I am reading generate crate2nix code and it does `buildRustCrate.override { ...}` which led me to believe `override` is something you can use on functions
<eyJhb>
How do I remove all the other commits, I based my branch on nixos-20.03..
<fendor__>
manveru, dumb questions incoming: where should the nix.conf be located? I see /etc/nix/nix.conf. Also, is there some documentation for that file?
<Taneb>
eyJhb: I think you can do "git rebase HEAD --onto nixos-20.03" but now I'm worried because I'm not that good and I don't want to accidentally break your checkout
<adisbladis>
Also please use `git cherry-pick -x` for backports so we have a reference to the commit on master
<manveru>
fendor__: you can either use that one, or make a per-user at ~/.config/nix/nix.conf
<manveru>
fendor__: the documentation is... pretty spread out :|
sigmundv has quit [Ping timeout: 256 seconds]
<manveru>
but `man nix.conf` shows most of it
Rusty1 has joined #nixos
<fendor__>
nice, thank you!
Rusty1 has quit [Remote host closed the connection]
<manveru>
you can also add it to your nixos config using `nix.extraOptions`
rogue_koder_ has quit [Remote host closed the connection]
<{^_^}>
[nixpkgs] @danieldk opened pull request #82404 → buildRustCrate: only link build deps into build script → https://git.io/Jv682
<ottidmes>
It boots fine and that secrets-initrd.gz seems to work fine too (although at the point of nixos-install it indeed does not exist)
<ottidmes>
Is it created at a later moment?
<ottidmes>
It is just a scary warning to get after an install
sigmundv has joined #nixos
werner291 has joined #nixos
RRedcroft has joined #nixos
<FireFly>
hmm, is there a sensible way to map volume controls to performing the corresponding actions in pulseaudio? I tried to map up services.acpid.handlers, but can't think of a clean way to deal with the handlers being run system-wide, but pulseaudio is configured with systemWide = false
ixxie has quit [Ping timeout: 265 seconds]
<ottidmes>
Hmm stupid of me, I checked /mnt/boot not /mnt/boot/grub, so I think it just warns about it missing in /boot/grub, which makes sense during installation, since its actually in /mnt/boot/grub. Will test it next time I am in an installer
veske has quit [Quit: This computer has gone to sleep]
veske2 has quit [Quit: This computer has gone to sleep]
veske has joined #nixos
veske2 has joined #nixos
v88m has quit [Ping timeout: 258 seconds]
<LinuxHackerman>
FireFly: you may be interested in the sound.mediaKeys.enable option, but it's very dependent on which desktop/window manager you use whether that option is actually what you want
<LinuxHackerman>
FireFly: for i3 and sway, I have keybindings like `bindsym XF86AudioRaiseVolume exec --no-startup-id "pactl set-sink-volume @DEFAULT_SINK@ +2%"`
<FireFly>
ah, the problem for me is that on at least this ideapad, the volume keys seem to emit acpi events, but not actually emit X11 media key keystrokes
<{^_^}>
[nixpkgs] @adisbladis pushed 2 commits to release-20.03: https://git.io/Jv64o
<FireFly>
I guess something like that would work, I wonder if there's a handy tool for synthesizing key events already.. but otherwise doing something akin to what xcape does (but without listening for modifier keydown/keyup events) would presumably work
karantan has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<Yaniel>
idk, I'd assume there is
bvdw has quit [Read error: Connection reset by peer]
bvdw has joined #nixos
<FireFly>
oh, duh, xdotool can simulate keypresses
* FireFly
tries
karantan has joined #nixos
<Yaniel>
on yeah, it can also get you banned from certain games :(
linarcx has quit [Ping timeout: 255 seconds]
<Yaniel>
probably not a problem on a laptop though
<Raito_Bezarius>
Is there a way to build QEMU VMs without having the user networking SLIRP on the top of it?
<Raito_Bezarius>
I do nix-build '<nixpkgs/nixos>' -A vm -I nixos-config=…
<Raito_Bezarius>
I wonder where I could change the parameters of the vm?
rivercat has quit [Quit: Lost terminal]
<{^_^}>
[nixpkgs] @grahamc opened pull request #82405 → pythonMinimal: don't include site-customise → https://git.io/Jv6BZ
bhipple has joined #nixos
skykanin has joined #nixos
<FireFly>
\o/ that works, using xdotool to emit corresponding key events in the acpi handlers, and then just mapping the keys as usual
<FireFly>
thanks :)
wfranzini has quit [Remote host closed the connection]
ash__ has joined #nixos
wfranzini has joined #nixos
ash__ is now known as ash_braker
<ottidmes>
I really ought to know this, but seem to be missing something, how do I copy a directory contents, not the directory itself, to another directory, cp -r dir1 dir2 results in dir2/dir1, and cp -r dir1/ dir2 results in dir2/dir1, and cp -r dir1/* dir2, wouldn't that fail if there are no contents in dir1?
Izorkin has quit [Ping timeout: 255 seconds]
<skykanin>
So I'm trying to run my haskell project, but I'm getting a freeglut error. System deps are handeled by nix and haskell packages are managed by cabal. Here is the shell.nix and project.cabal files https://pastebin.com/zxyqh8JRhttps://pastebin.com/EMFTDYeu. I'm on my laptop with intel integrated graphics, currently using mesa drivers with gallium (iris) enabled. I don't have any problem running this
<skykanin>
on my desktop with nvidia drivers.
sascha-grunert has quit [Remote host closed the connection]
<skykanin>
The error I'm getting is "freeglut (GameOfLife): ERROR: Internal error <FBConfig with necessary capabilities not found> in function fgOpenWindow"
<skykanin>
yes
<skykanin>
Yes
Izorkin has joined #nixos
stolyaroleh has joined #nixos
zupo has joined #nixos
<{^_^}>
[nixpkgs] @mmahut opened pull request #82406 → quorum: init at 2.5.0 → https://git.io/Jv6Bx
<{^_^}>
[nixpkgs] @flokli merged pull request #82373 → gitstatus: Fix build of the internally used libgit2 fork → https://git.io/Jv6qK
<LinuxHackerman>
Raito_Bezarius: you can either set config.virtualisation.qemu.networkingOptions in the VM's configuration, or set the QEMU_NET_OPTS environment variable for the run-nixos-vm script
<RRedcroft>
can i have multiple environment.systempackages in different config files? trying to put this into a "python.nix" file, but it complains about type https://pastebin.com/raw/yzNhJeSg
sigmundv has quit [Ping timeout: 258 seconds]
<LinuxHackerman>
RRedcroft: yes
<LinuxHackerman>
RRedcroft: you probably want python-with-my-packages rather than my-packages :)
<{^_^}>
[nixpkgs] @vcunat pushed commit from @andir to staging-next « firefox: 73.0.1 -> 74.0 (#82276) »: https://git.io/Jv6R8
karantan has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
wfranzini has quit [Quit: wfranzini]
cosimone has joined #nixos
<bennofs[m]>
does nix support building the nix binary itself completely statically?
<blackriversoftwa>
bennofs[m]: the neat thing with nix is it's usually pretty easy to try for yourself without worrying that you'll break anything! Try using `overrideAttrs` on the `nix` derivation from nixpkgs, and setting `dontDisableStatic = true;`
<lukego>
I made a nixos installer on Linux with 'sudo cp nixos-graphical-19.09.2201.7d31bbceaa1-x86_64-linux.iso /dev/sda' but trying to boot it just shows "Missing operating system" from the BIOS. any ideas?
Makaveli7 has quit [Quit: WeeChat 2.4]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
<pie_[bnc]>
clever: is there no way to do something like nix-store -l on a failed build?
malSet has quit [Read error: Connection reset by peer]
kenran_ has quit [Ping timeout: 258 seconds]
<{^_^}>
[nixpkgs] @GuillaumeDesforges opened pull request #82410 → scikit-survival: init at 0.11 → https://git.io/Jv60C
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
skykanin has quit [Quit: WeeChat 2.7.1]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
<NobbZ[m]>
lukego: If it is visible in udiskies UI or whatever else my current DE uses, then I usually hit the eject/power off button, just to be sure. But is your stick really `/dev//sda`?
<NobbZ[m]>
* lukego: If it is visible in udiskies UI or whatever else my current DE uses, then I usually hit the eject/power off button, just to be sure. But is your stick really `/dev/sda`?
<lukego>
Yeah. at least when I insert it dmesg prints messages about a 30GB volume sda being added. root disk is an nvme
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
<clever>
pie_[bnc]: nix doesnt save the output for failed builds
<pie_[bnc]>
LnL: i mean, i have it in my scroll, but i was wondering...
cmacrae` has quit [Ping timeout: 240 seconds]
<maddo>
jtojnar: oh I see
<maddo>
is there a way I could technically use the 3.36?
chagra_ has quit [Quit: WeeChat 2.7.1]
<maddo>
or git software in general, since I also need to test things from master
<jtojnar>
GNOME is not really supported outside unstable channel any way
<jtojnar>
I recommend using nixos-unstable channel directly
<jtojnar>
I would say unstable is misnomer, we should really rename unstable → fresh & release → stale like LibreOffice does
kenran has joined #nixos
jakobrs has joined #nixos
chagra has joined #nixos
wolfshappen has joined #nixos
<jakobrs>
Is it possible to create a script that ensures that a certain store path is live while the script is running?
<jakobrs>
Like `nix-shell -p /nix/store/path`, only that requires that the path is already realised
rardiol has joined #nixos
flodeskum has quit [Remote host closed the connection]
cosimone has quit [Quit: Quit.]
<ottidmes>
jakobrs: sounds like you already give the answer, or I misunderstood you, but why not wrap the call in a script that runs nix-store --realise to make sure those paths exist?
veske has joined #nixos
veske2 has joined #nixos
<jakobrs>
nttidmes: nix-store --realise could in theory cause a race condition if the path is garbage collected immediately after
<ottidmes>
jakobrs: you might want to use nix-store --query --deriver to get the .drv of those paths instead, cause I read nix-store --realize won't build them locally otherwise
flodeskum has joined #nixos
eacameron has joined #nixos
<ottidmes>
jakobrs: sure, but a lot of things can happen, question is, is that race condition likely to happen and worth putting the effort in to fixing?
kenran has quit [Ping timeout: 258 seconds]
veske has quit [Client Quit]
veske2 has quit [Client Quit]
<ottidmes>
jakobrs: there is nix-shell --repair, which probably does what you want
<jakobrs>
ottidmes: I realise it's not too important. fwiw I just found out `nix run` will realise the paths automatically. Though it can't be used like the nix-shell shebang.
<symphorien>
jakobrs: the garbage collector checks that a path is not opened by a running program before removing it
ddellacosta has joined #nixos
orivej has quit [Ping timeout: 268 seconds]
orivej has joined #nixos
dingenskirchen has quit [Ping timeout: 245 seconds]
<ottidmes>
I remember reading a trick about this, I will try and search the logs, but let me ask anyway, I have certain packages that should have a root, but are garbage collected, is there a way to make sure some packages are not garbage collected without putting them in say systemPackages
<Raito_Bezarius>
I'm using encryptedLinksTo but for some reason, it's creating records with malformed IP addresses
<Raito_Bezarius>
like 192.168.105.0.0.8
aveltras has joined #nixos
<immae>
ottidmes: I suppose something like `nix-env -p /some/path/to/a/dummy/location -i package` could work?
<ottidmes>
immae: I want this to be part of my NixOS configuration, not an external call
<immae>
Ah
<immae>
Then an empty derivation with the said package as buildInputs ?
<ottidmes>
immae: something like that, if I can't find the remark in the logs soon, I guess I will experiment with that, but I believe buildInputs get filtered for those that are actually used, but you could probably force that somehow
<{^_^}>
[nixpkgs] @timokau merged pull request #82416 → [20.03] sage: fix test suite with docutils 0.15 → https://git.io/Jv6zq
<{^_^}>
[nixpkgs] @timokau pushed 2 commits to release-20.03: https://git.io/Jv6gk
<immae>
I would do something like runCommand { ... } "echo ${package}" I guess :p
<{^_^}>
[nixpkgs] @teto pushed commit from @marsam to master « nushell: 0.10.0 -> 0.11.0 »: https://git.io/Jv62T
chloekek has quit [Quit: WeeChat 2.7]
stolyaroleh has quit [Quit: Konversation terminated!]
<boogiewoogie[m]>
is there a way to list just the attribute names in a set? I wanna find the path to my currently used kernel derivation without directly searching the store, but evaluating `linuxPackages` wasn't as helpful as I thought. :P https://hastebin.com/erimizoneh
<ottidmes>
boogiewoogie[m]: attrNames?
teto has joined #nixos
<boogiewoogie[m]>
ottidmes: oh, could have thought of that.. thanks! :D
RRedcroft has quit [Ping timeout: 258 seconds]
v88m has joined #nixos
shafox has quit [Remote host closed the connection]
<lovesegfault>
is it alright to nix-collect-garbage while I have a nix-shell open?
<gchristensen>
yep
* lovesegfault
does it
<lovesegfault>
gchristensen++
<{^_^}>
gchristensen's karma got increased to 228
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
<evanjs->
Anybody else having issues cross compiling libwebsockets? I know peterhoeg made a pr to reinit the package with cmake in nativeBuildInputs, but it looks like some tests are failing when I try to compile it with e.g. pkgsCross.armv7l-hf-multiplatform
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
chagra has quit [Quit: WeeChat 2.7.1]
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
Ariakenom has quit [Quit: WeeChat 2.7]
Diogo418 has quit [Quit: Ping timeout (120 seconds)]
Diogo418 has joined #nixos
iqubic` has quit [Ping timeout: 265 seconds]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
smeag0l has joined #nixos
iqubic` has joined #nixos
chagra has joined #nixos
iqubic` has quit [Remote host closed the connection]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
malSet has joined #nixos
malSet has quit [Read error: Connection reset by peer]
bvdw has quit [Write error: Connection reset by peer]
<excelsiora>
gchristensen: how do I run tests for nixops again?
bvdw has joined #nixos
kenran has quit [Ping timeout: 256 seconds]
Jackneill has quit [Remote host closed the connection]
<eyJhb>
ANyone else having trouble with vim-go?
<raboof>
I have a script/project that I can run in a 'nix shell', and now I'd like to package it up so that it runs as a LXC linux container. I saw nix-generators which looks interesting. there's also nixpkgs/nixos/modules/virtualisation/lxc.nix , but if I understand this correctly that is for running LXC contains on NixOS, not for running NixOS in an LXC container - does that sound right?
<raboof>
(my development machine is NixOS, I'd like to deploy to a proxmox cluster which AFAICS is some Debian-based product)
<excelsiora>
can anyone tell me how to run tests for nixops? we're trying to add mypy...
<bqv[m]>
excelsiora: totally unrelated but, i spotted your name in peripheral vision and my first thought was hogwarts
<boogiewoogie[m]>
hey, it seems that `resolved` can't currently be used with dns-over-tls as there's no option for it and `services.resolved.extraConfig = "DNSOverTLS=yes"` isn't enough. is this due to https://github.com/systemd/systemd/issues/9397 and is there an option planned for when systemd >243 is hitting nixos?
<{^_^}>
systemd/systemd#9397 (by shibumi, 1 year ago, closed): RFE: Certificate checking for Resolveds DNS over TLS feature
orivej has joined #nixos
knupfer has quit [Remote host closed the connection]
<ldlework>
adisbladis: if you get a moment can you help me in #nixos-emacs
psyanticy has quit [Quit: Connection closed for inactivity]
<Raito_Bezarius>
is there a way to have some precise private IP on NixOps?
<Raito_Bezarius>
other than 192.168.(index + 105 % 256).(index % 256)
<raboof>
hmm, is it expected that 'nixos-generate' would recompile all kinds of stuff, like xorg and the jre, instead of getting them from cache? I'm running the nixos-unstable channel.
<ldlework>
I'm so confused. I don't have the word "leo" in any of my nix configuration, and it's not in nix-env and set catting out my emacs binary, I get:
<{^_^}>
[nixpkgs] @alyssais pushed commit from @marsam to master « heroku: 7.38.1 -> 7.39.0 »: https://git.io/Jv6ri
<ldlework>
tilpner: thanks man
<raboof>
tilpner: looks like I'm a little behind on unstable, at 82b54d490663b6d87b7b34b9cfc0985df8b49c7d, but not much. I'll update
flodeskum has quit [Remote host closed the connection]
<raboof>
oh no that's actually uptodate even
<tilpner>
raboof: Ahh, it's the lxc profile
<tilpner>
raboof: It imports profiles/docker-container, which imports profiles/minimal, which sets environment.noXlibs = mkDefault true;
<tilpner>
And that causes the cache misses you're seeing
<ldlework>
argh, the package is still there somehow
<judson_>
Anyone have advice on dealing with the long process names that happen with NixOS?
<tilpner>
raboof: If you prefer a larger container to rebuilding locally, set that back to false
<raboof>
tilpner: ah, so because java was built with X support and I'm not building for a target without X support it's rebuilding. that makes sense then
<judson_>
<nixstore>/<hash>-cmd <nixstore>/<hash>-file gets up to hundreds of bytes quickly, and for e.g. ps it's mostly noise.
<raboof>
tilpner: thanks a lot, I'll just rebuild then (seems neat to have it 'minimal' like that), good to understand what's going on though ;)
<tilpner>
judson_: In htop, press p and those long paths disappear. That's the only time it annoyed me
flodeskum has joined #nixos
<judson_>
Our htop reports different processes, tilpner :)
<judson_>
p chops out some of those, but leaves, e.g. a ton of chromium processes
<ldlework>
wat
<ldlework>
error: packages '/nix/store/ggrfid3vpf94zd8zfbp9s6gawkk10wms-home-manager-path/LICENSE.txt' and '/nix/store/bkrkm7fia7f3vfjk7j9hplag9r7l4bp9-dotnet-sdk-2.2.103/LICENSE.txt' have the same priority 5;
civodul has joined #nixos
<ldlework>
This is gonna be a NixOS day, I can feel it :(
<Orbstheorem>
How can I use a file as a derivation source ?
<Orbstheorem>
I get error "do not know how to unpack source archive ..." when I just specify the path :(
<{^_^}>
[nixpkgs] @veprbl pushed commit from @amurrayc to release-20.03 « blender: use PYTHONPATH (#82341) »: https://git.io/Jv6rb
<cole-h>
Why do you have `[] ++`? Removing that might solve your problem.
zeta_0 has joined #nixos
shibboleth has joined #nixos
<Orbstheorem>
Same problem after removing it ^^
<Orbstheorem>
That variable has a default just bellow (l.80) to a list of paths.
hoijui has quit [Ping timeout: 268 seconds]
<tilpner>
Orbstheorem: Setting src to a list might trigger behaviour that assumes src is a single thing
<tilpner>
Orbstheorem: You're already trying to use srcs below, srcs = cfg.schroot_setup_d; behave differently?
<tilpner>
*+does
kolbycrouch has quit [Ping timeout: 268 seconds]
<Orbstheorem>
Well, if I recall correctly, the builder cannot use any other path from the nix store that wasn't declared in nixpkgs, am I right?
<ldlework>
GNU Emacs 28.0.50
<ldlework>
yay
ng0_ has joined #nixos
ng0_ has quit [Client Quit]
<tilpner>
Orbstheorem: That's technically wrong, but you mean the right thing. I think it should still work because srcs = ...; adds it to the derivation inputs
rajivr___ has quit [Quit: Connection closed for inactivity]
<Orbstheorem>
Wait, won't the interpolated strings be resolved by nix when making the closure for the derivation?
<tilpner>
I think so, but verify to be sure
Diogo4189 has joined #nixos
kolbycrouch has joined #nixos
hoijui has joined #nixos
airgap has joined #nixos
Diogo418 has quit [Ping timeout: 260 seconds]
<airgap>
non-nixos nix questions ok here?
reactormonk has joined #nixos
<FireFly>
I believe so
<ar>
yup
<airgap>
roger that. just installed nix, but i'm getting this error running anything: /nix/store/4b6f6apsk9k9649iml240w4by5qxli7h-bash-4.4-p23/bin/bash: error while loading shared libraries: libstdc++.so.6: cannot open shared object file: No such file or directory
<airgap>
colemickens: Orbstheorem: haven't traced it yet. this was a quick setup and thought *maybe* this was a known issue that had a quick answer. google yielded nothing
<{^_^}>
[nixpkgs] @Ekleog pushed commit from @symphorien to release-20.03 « nixos/btrfs: make autoScrub not prevent shutdown or suspend »: https://git.io/Jv6Kr
<{^_^}>
[nixpkgs] @Ekleog merged pull request #76739 → nixos/dovecot: add an option to enable mail_plugins → https://git.io/JeADN
<butterthebuddha>
NobbZ[m] Some pip packages have executables right?
ravndal has joined #nixos
o1lo01ol1o has joined #nixos
<NobbZ[m]>
butterthebuddha: If you are talking about libraries, you should instead properly depend on them in your projects, if though you are talking about executables/scripts you should use `poetry2nix` or `pypi2nix` or similar tools to create a derivation (if not yet available) and install that.
absoluutely__ has joined #nixos
linarcx has quit [Ping timeout: 260 seconds]
<butterthebuddha>
NobbZ[m] Can I use home-manager to manage these derivations?
<butterthebuddha>
It doesn't look like the package I want to install (leccap) exists
<butterthebuddha>
in nixpkgs
Jackneill has joined #nixos
<NobbZ[m]>
Yes, you can. Just put them in an overlay.
civodul has quit [Quit: ERC (IRC client for Emacs 26.3)]
abathur has quit [Ping timeout: 256 seconds]
rb2k has joined #nixos
<mcwitt>
On darwin, trying to build an R environment with a package compiled from source, getting linker about missing `-liconv`. Investigating, found a MWE that also fails: `nix-shell -p libiconv pkgconfig --run 'pkg-config --cflags libiconv'` results in "Package libiconv was not found in the pkg-config search path". Any ideas? Does this just work differently on darwin?
<NobbZ[m]>
mcwitt: As I have the same error on nixOS, something else seems to be missing in that command…
<{^_^}>
[nix] @edolstra pushed to flakes « Change the lock file to a graph »: https://git.io/Jv66I
zupo has joined #nixos
gustavderdrache has quit [Quit: Leaving.]
chloekek has quit [Ping timeout: 265 seconds]
mexisme_ has quit [Ping timeout: 240 seconds]
<drakonis>
oh, nice.
<mcwitt>
NobbZ[m]: Thanks for checking. Actually, I'm not sure if that specific example should work, since it sounds like `pkg-config` needs to get added to `nativeBuildInputs` for things to get set up correctly. But following the example on the wiki (https://nixos.wiki/wiki/C#pkg-config) I have the same issue, and also notice that $PKG_CONFIG_PATH is empty in the shell.
<mcwitt>
(had to change cryptsetup since it's not darwin-compatible, used libiconv instead)
rb2k has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
rb2k has joined #nixos
Jackneill has quit [Remote host closed the connection]
absoluutely__ has quit [Remote host closed the connection]
absoluutely_ has joined #nixos
absoluutely_ has quit [Remote host closed the connection]
absoluutely_ has joined #nixos
absoluutely_ has quit [Remote host closed the connection]