AllanEspinosa has quit [(Ping timeout: 252 seconds)]
Fare has quit [(Ping timeout: 246 seconds)]
<NixOS_GitHub>
[nixpkgs] basvandijk opened pull request #21032: whisper: upgrade to the same version as the other graphite components (release-16.09...fix-graphite_web-whisper) https://git.io/v1rfb
AllanEspinosa has joined #nixos
latk has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
erasmas has quit [(Quit: leaving)]
herzmeister has quit [(Quit: Leaving)]
herzmeister has joined #nixos
markus1199 has joined #nixos
markus1189 has quit [(Ping timeout: 244 seconds)]
anelson has quit [(Ping timeout: 245 seconds)]
anelson has joined #nixos
digitus has quit [(Quit: digitus)]
estewei has quit [(Quit: Quit)]
mizu_no_oto has joined #nixos
xwvvvvwx has joined #nixos
<xwvvvvwx>
Hey :)
<xwvvvvwx>
I have a really quick question about packing stuff up for nix
<xwvvvvwx>
when I run nix-hash --type sha256 <path_to_unpacked_dir>
<xwvvvvwx>
the hash that it outputs is longer than the one in the existing derivation
<xwvvvvwx>
and it is not accepted by nix
Seichi has quit [(Read error: Connection reset by peer)]
<xwvvvvwx>
how should I compute the hash of the output dir?
Seichi has joined #nixos
michalrus has joined #nixos
<xwvvvvwx>
ok figured it out, I have to use --base32 to get the compact representation
<joepie91>
xwvvvvwx: fwiw, the fastest option is to just let it fail with the wrong hash
<joepie91>
since it'll tell you what hash it expected
<xwvvvvwx>
yes, but it's a binary package
<xwvvvvwx>
so I did that for the linux x86 package
<xwvvvvwx>
but I can't for darwin / linux32
Fare has joined #nixos
AllanEspinosa has quit [(Ping timeout: 252 seconds)]
xwvvvvwx has quit [(Quit: WeeChat 1.6)]
jsgrant has quit [(Ping timeout: 260 seconds)]
jsgrant has joined #nixos
systemfault has joined #nixos
chris-bacon has joined #nixos
xwvvvvwx has joined #nixos
<xwvvvvwx>
Hey again :)
<xwvvvvwx>
So I was just checking out my commit against the contributing.md before submitting
<xwvvvvwx>
and there is a line that says I have to run `nix-shell -p nox --run "nox-review wip"`
kampfschlaefer has quit [(Ping timeout: 258 seconds)]
<xwvvvvwx>
but when I run this I get the following output: No uncommit changes. Did you mean to use the "--against" option?
<chris-bacon>
hello.. I am trying to install google chrome (not chromium) on nixos, and am also using grsec. I run into a well-known SUID error but I am still not getting to work. Can anyone help
<xwvvvvwx>
what is nox and how do I run it against a commit?
Seichi has quit [(Quit: Konversation terminated!)]
<Leo`>
Why does llvm-3.8 provides libLLVM.so but llvm-3.7.1 does not? Strange.
zraexy has quit [(Ping timeout: 244 seconds)]
redmq has quit [(Ping timeout: 252 seconds)]
<gchristensen>
anyone around with rebuild capabilities on hydra?
slack1256 has joined #nixos
filterfish has quit [(Remote host closed the connection)]
filterfish has joined #nixos
<Leo`>
Are there someone who develops with OCaml on NixOS out there?
Wizek_ has joined #nixos
slack1256 has quit [(Remote host closed the connection)]
<avn>
Leo`: I tried to play with some ocaml stuff. What questions you have?
chris-bacon has quit [(Quit: Page closed)]
mudri has quit [(Quit: WeeChat 1.6)]
jarlg has quit [(Quit: WeeChat 1.5)]
filterfish has quit [(Read error: Connection reset by peer)]
filterfish has joined #nixos
<Leo`>
avn: Do you typically use OPAM or Nix to install dependencies?
AllanEspinosa has joined #nixos
wak-work has joined #nixos
<avn>
Leo`: I tried only nix. I thinked about something generating .nix from OPAM, but exists solution not looks ideal
<Leo`>
avn: Ok. Thing is I don't really understand how tu use OCaml modules through Nix.
<Leo`>
I usually use Findlib (ocamlfind) to locate dependencies, but it seems unable to locate libraries installed through Nix.
glines has quit [(Ping timeout: 244 seconds)]
<avn>
Leo`: it works at least for lwt and js_of_ocaml stuff (ppx and runtime)
<Leo`>
I mean I only did something as simple as installing ocaml, findlib and batteries (for example) with nix-env, but ocamlfind ocamlc -package batteries still fails to find it.
mizu_no_oto has quit [(Quit: Computer has gone to sleep.)]
<Leo`>
(Which doesn't surprise me since Nix doesn't seem to create links to the libraries in ocamlfind's search path... But I don't understand how it's supposed to work then.)
<clever>
Leo`: you generaly only put libraries in nix-shell -p, they should never be installed with nix-env
<avn>
Leo`: I tried myEnvFun, but nix-shell should work as well. Not sure about nix-env
<Leo`>
Ah it works with nix-shell indeed.
<Leo`>
Mhhhhhh.
derjohn_mobi has quit [(Ping timeout: 250 seconds)]
<Leo`>
So when using nix-shell, special envvars are defined for ocamlfind...
<Leo`>
Interesting.
<clever>
you can see what gets set by doing "env > a.txt" outside of the shell, and "env > b.txt" inside the shell
<clever>
then just diff -u a.txt b.txt
<Leo`>
But for system (C) libraries, using nix-env links them to ~/.nix-profile/lib so that the toolchain can find them.
<Leo`>
Why not doing the same thing with Findlib?
derjohn_mob has joined #nixos
<clever>
thats not how nix is meant to be used
<clever>
different versions of the libs will conflict, and ruin the entire benefit of using nix
<Leo`>
Mh. Ok.
<avn>
Leo`: actually nix-env is an ad-hoc stuff in nixos
<Leo`>
I guess I'll get used to it. :p
<Leo`>
avn: Ad-hoc?
<avn>
Leo`: `nix-env -i` is ad-hoc ;) When you forget to install something, and want it before next nixos-rebuild
<Leo`>
Mh.
<Leo`>
Ok.
<avn>
but it my own truth ofc ;)
<Leo`>
I use it way too much then. :P
<Leo`>
But I guess that's a common thing to do when you transition from “standard” distributions. :p
<avn>
I personally build all (or most) things at nixos-rebuild (and have them defined in configuration.nix under git control)
<avn>
Leo`: sure, `transition` is a key word here ;)
estewei has joined #nixos
<avn>
Leo`: defining project requirements in shell.nix, and commiting with project is a step forward ;)
<Leo`>
:)
<Leo`>
It's annoying to have to use bash with nix-shell though...
<Leo`>
I tried --run zsh but that must mess-up the environment somehow since ocamlfind doesn't work anymore with it.
redmq has joined #nixos
ilja_kuklic has joined #nixos
jsgrant has quit [(Quit: Konversation terminated!)]
ilja_kuklic has quit [(Ping timeout: 260 seconds)]
Shoue has quit [(Ping timeout: 258 seconds)]
glines has quit [(Ping timeout: 260 seconds)]
thc202 has quit [(Ping timeout: 256 seconds)]
sdothum has quit [(Quit: ZNC - 1.6.0 - http://znc.in)]
ericsagn1 has quit [(Ping timeout: 240 seconds)]
glines has joined #nixos
jsgrant has joined #nixos
sdothum has joined #nixos
ericsagnes has joined #nixos
Fare has quit [(Ping timeout: 268 seconds)]
Fare has joined #nixos
jsgrant_ has joined #nixos
zimbatm has quit [(Remote host closed the connection)]
pSub has quit [(Write error: Broken pipe)]
lverns has joined #nixos
jsgrant has quit [(Write error: Broken pipe)]
pSub has joined #nixos
pSub has quit [(Changing host)]
pSub has joined #nixos
nmattia_ has joined #nixos
rrrr_ has joined #nixos
aszlig_ has joined #nixos
zimbatm has joined #nixos
BlessJah_ has joined #nixos
ben__ has joined #nixos
tumm_ has joined #nixos
page_ has joined #nixos
<spacekitteh>
good morning angels
siel_ has joined #nixos
earldoug1as has joined #nixos
smartin_ has joined #nixos
rrrr has quit [(Write error: Broken pipe)]
WilliButz has quit [(Write error: Broken pipe)]
tumm has quit [(Write error: Broken pipe)]
ben has quit [(Write error: Broken pipe)]
page has quit [(Write error: Broken pipe)]
smartin has quit [(Write error: Broken pipe)]
siel has quit [(Write error: Broken pipe)]
aszlig has quit [(Write error: Broken pipe)]
andreas303 has quit [(Write error: Broken pipe)]
nmattia has quit [(Write error: Broken pipe)]
earldouglas has quit [(Remote host closed the connection)]
BlessJah has quit [(Remote host closed the connection)]
mjacob has quit [(Remote host closed the connection)]
mjacob has joined #nixos
siel_ is now known as siel
PragCyph1 has joined #nixos
WilliButz has joined #nixos
PragCyphr has quit [(Ping timeout: 250 seconds)]
andreas303 has joined #nixos
angerman has quit [(Quit: Bye)]
cpennington has joined #nixos
angerman has joined #nixos
filterfish has joined #nixos
filterfish has quit [(Remote host closed the connection)]
filterfish has joined #nixos
jsgrant_ has quit [(Ping timeout: 256 seconds)]
sdothum has quit [(Quit: ZNC - 1.6.0 - http://znc.in)]
hexagoxel has quit [(Ping timeout: 264 seconds)]
hexagoxel has joined #nixos
redmq has joined #nixos
mguentner has quit [(Quit: WeeChat 1.6)]
Rotaerk has joined #nixos
mguentner has joined #nixos
sdothum has joined #nixos
nh2_ has quit [(Ping timeout: 250 seconds)]
filterfish has quit [(Read error: Connection reset by peer)]
filterfish has joined #nixos
Fare has quit [(Ping timeout: 250 seconds)]
cpennington has quit [(Remote host closed the connection)]
ashkitten has joined #nixos
jsgrant_ has joined #nixos
<ashkitten>
Hello, I'm new to NixOS (coming from Arch). I'm having issues installing packages, when I do nix-env -i discord, it outputs "warning: name collision in input Nix expressions, skipping ‘/home/ash"
<ashkitten>
warning: name collision in input Nix expressions, skipping ‘/home/ash/.nix-defexpr/channels_root/nixpkgs’
<NixOS_GitHub>
[nixpkgs] matthewbauer closed pull request #20658: GIMP on Darwin (use Quartz) (master...gimp-darwin) https://git.io/v1vfe
<NixOS_GitHub>
[nixpkgs] matthewbauer closed pull request #16192: [wip] wxWidgets: move wxGTK-* to one wxWidgets folder (master...wxwidgets-refactor) https://git.io/vonAR
<clever>
ashkitten: you need to remove the nixpkgs channel in nix-channel --list
<clever>
ashkitten: channels are per-user, and you generaly only want a channel on root
<ashkitten>
oh okay
<ashkitten>
now it's saying "discord matches no derivations"
<clever>
nix-env -iA nixos.discord
<c74d>
also I hear the nixpkgs channel shouldn't be used on NixOS
<c74d>
(it's for non-NixOS usage of Nix)
<ashkitten>
ah
lverns has quit [(Quit: Konversation terminated!)]
<ashkitten>
I need to set allowUnfree = true, and I did that and did nixos-rebuild switch but it's still giving me the error
<clever>
newNixos: i have 2 installs of nixos on armv6, and i am working on a v7 build
nekroze has joined #nixos
mguentner has quit [(Ping timeout: 240 seconds)]
<nekroze>
I am having some trouble setting up a bridge network to allow containers to have IP's on my LAN. Does anyone have any examples of this kind of thing?
<newNixos>
clever: do you think I could get running easily
<newNixos>
or I could bootstrap from my current arch install
<clever>
newNixos: the simplest way to start, is to install nix on another distro like the debian variants for the beaglebone, and then use nix-build to compile the rest up and nixos-install to an SD card
<newNixos>
ok
<clever>
newNixos: i have a couple links that will speed it up as well
<clever>
and the config.platform option has recently been "broken", it no longer accepts an attrset, it instead wants a function from system to attrset
<newNixos>
what I really want is some riscv ... so tired of thinkering with hw
<clever>
so platform = system: pkgs.platforms.raspberrypi2; would work on the newer version
<newNixos>
ill write this down in case
nekroze has quit [(Ping timeout: 268 seconds)]
<clever>
i'm also experimenting with a custom OS that can act as a build slave
<clever>
with a single nix-build command, i get a kernel, initrd, and a ~40mb squashfs
<clever>
and thats enough to boot a raspberry pi 3 over the network, and make it into a nix build slave
AllanEspinosa has quit [(Read error: Connection reset by peer)]
AllanEspinosa has joined #nixos
jsgrant_ has quit [(Quit: Konversation terminated!)]
AllanEspinosa has quit [(Ping timeout: 250 seconds)]
Piece_Maker has joined #nixos
Acou_Bass has quit [(Ping timeout: 260 seconds)]
Piece_Maker is now known as Acou_Bass
ebzzry has joined #nixos
<newNixos>
does anyone know if nixos-infect works on linode? This manual bootstrap isnt working
<clever>
newNixos: there is a new thing that ive been testing out on virtualbox
<clever>
newNixos: if you insert your own ssh public key into configuration.nix, and build it correctly, you will get a tar file, upload it to the server and run it as shown in session.md
nekroze has joined #nixos
<clever>
newNixos: and you will get nixos running from a ramdisk, similar to if you had put an install ISO in the machine
<clever>
then you are free to format the hdd and nixos-install like normal
<nekroze>
I am trying to use declarative nixos containers with a networking bridge so they can sit on my lan and be directly accessable to other machines on the network without a reverse proxy for example. It seems to be linking up the network devices correctly to the veth but I cant get networking in or out of the container. Has anyone successfully used containers with bridged networking?
<clever>
nekroze: i havent tried containers with a bridge yet, but you will need to at least do networking.bridges.br0.interfaces = [ "enp0s20u1" ]; with the LAN interface
<clever>
nekroze: and then somehow add the containers to br0
<clever>
nekroze: i think you need to add the veth interfaces to br0
<newNixos>
clever: thats really clever ;)
<newNixos>
ill try
<nekroze>
clever: I have the bridge configured, I have also tried assigning and IP and prefixLength to the br0 interface that is on the lan as I have had to do that when I used openvswitch bridges with kvm
<clever>
nekroze: under nixos, it will just run dhcpcd on br0 automaticaly, if it lacks an IP
<clever>
and if a lan interface has been added, it will get a reply from the lan
<nekroze>
the bridge gets added to the container (from what I can see this is correct) via the containers.<name>.hostBridge = "br0";
<clever>
nekroze: ah, hadnt seen that option, looks perfect
<nekroze>
I have tried both with an IP assigned to the bridge and without. I do have a single physical nic linked to the bridge that gets an ip via DHCP if its not on the bridge
<nekroze>
when all this didn't work I even tried using a full openvswitch bridge, but it couldnt link the veth up to that. From in the container it just says network is unreachable whenever I try to ping out or do anything
<clever>
nekroze: let me finish cleaning out my laptop, then i can experiment with things on this end
<clever>
hmmm, or qemu would work
<nekroze>
indeed, as I am just working on using pure containers on bare metal, it can be tested pretty easy in vm's
<clever>
hmmm, the nat of qemu may complicate it, but we will see
<nekroze>
clever: interestingly, if I don't assign the static ip to the br0 interface it doesn't get an IP
<clever>
nekroze: did you change any dhcpcd related settings?
<nekroze>
nope
<nekroze>
oh sorry I tell a lie, my live nixops deployment did globally disable dhcp
<clever>
i think setting it to anything will also disable it on all other interfaces
BlessJah_ is now known as BlessJah
<clever>
so if its null, it will just auto-dhcp everyting
<nekroze>
clever: yeah, it is null so it should work. also it doesn't have a .enable switch
<clever>
how is nixops turning dhcp off?
<nekroze>
clever: never mind that part, I have an if statement in my nixops deployment to change a few options if the deployment was declared a test environment
<nekroze>
the physical box needs a static ip for its primary nic
matthewbauer has quit [(Ping timeout: 260 seconds)]
<clever>
nekroze: you will want to move that static ip to br0
<nekroze>
clever: sorry the bridge is connected to one of the other nics, they are all on the same lan but I wan't the containers to have a dedicated nic for their traffic compared to the host that runs the containers
<nekroze>
that nic has no specific configurations
<clever>
ah
<clever>
id give it a second static ip then
<clever>
also, the hostBridge option behaves differently then i was expecting
<nekroze>
but now that I have networking.useDHCP = true the bridge does get a dhcp address on its own if I don't make it static. but I still can't route out of or into the container
<nekroze>
clever: Oh?
<nekroze>
different how?
<clever>
uploading
<clever>
nekroze: http://imgur.com/a/BDhKz with a minimal config, it wound up with the exact same interfaces as the host
<clever>
letme try again with private networking enabled
<nekroze>
clever: thats interesting as my container does not have the same interfaces as the host
<clever>
booting it again with privateNetwork = true;
<clever>
now it has its own network interface
<clever>
which lacks an IP
<nekroze>
clever: ah, thats the configuration my container has, privateNetwork = true. So that sounds like you have hit the mark
<clever>
building it with some more debug utils
<nekroze>
Now I can set the ip by using the container.<name>.localAddress but it still doesnt route
<clever>
nekroze: aha, i ran tcpdump on br0 while booting a container, it never tried to do dhcp!
<clever>
The IPv4 address assigned to the interface in the container. If a hostBridge is used, this should be given with netmask to access the whole network. Otherwise the default netmask is /32 and routing is set up from localAddress to hostAddress and back.
<clever>
the docs for localAddress
<nekroze>
ahh. a netmask on the localAddress i did not probide
<nekroze>
ahh. a netmask on the localAddress i did not provide
<clever>
that will at least give it access to the LAN, but probably not the WAN
<nekroze>
apologies for the double send
<nekroze>
clever: that should be a decent enough start. and I should be able to get wan access by setting the networking.defaultGateway in the container as I believe that configures a route that should do the job
<clever>
yeah
<nekroze>
let me try this
<clever>
nekroze: also, nixos is unable to restart declarative containers when they change
<clever>
or just reboot
<clever>
you need to manualy systemctl restart container@test1
<nekroze>
indeed lan, wan, and dns all working now
<clever>
uh, i didnt set defaultGateway, and i also have wan, lol
<nekroze>
clever: I did notice that it doesn't restart. that is most unfortunate... I'll have to look into why
<clever>
it must be using the routing tables of the host
jacob_ has joined #nixos
jacob_ is now known as Guest93244
<nekroze>
yeah that would be my guess if it didn't pick it up from dhcp
<clever>
nekroze: the issue, is that nixos-rebuild checks to see if the /etc/systemd/system/container@.service symlink has changed or not
<nekroze>
I think I will leave the setting to be safe
<clever>
and restarts when it does
<clever>
but container@test1 reads /etc/containers/test1.conf, which is a bit of an impurity
<clever>
that nixos-rebuild isnt aware of
<clever>
it affects all services using @
<nekroze>
clever: huh, yeah I can see, at first glance, why the issue is extant then
<clever>
its also unable to restart services within a container as seamlessly as it does on the host
<newNixos>
we should open a vps that offers only nixos
<clever>
it instead shuts down the entire container, then boots it back up with the new config
<newNixos>
nix ocean
<newNixos>
and do ipv6 proprely
<nekroze>
clever: that is suboptimal. I guess I will set up some 2 host loadbalancing for the important things then and manually reboot them one at a time via script for a while until we can find a way to make it smoother
<clever>
nekroze: is there a reason you need to use containers?
<nekroze>
clever: My "server hardware" is essentially repurposed gaming hardware of a decent calibre and I was starting to hit resource sharing issues with VM's
<clever>
nekroze: is there anything stopping you from just running all of the services right on the host?
<nekroze>
clever: plus I have been meaning to tear it all down and switch to pure nixos as previously I was running xenserver and a bunch of different distro's
<clever>
ah, similar mess on one of my servers, xen+ubuntu on the host, and a random collection of domU's
<nekroze>
clever: I want some seperation between my home services and my public ones. While down at the moment I run my own little website and git/ci/qa servers along with some media stuff for the internal network and adblocking dns for the house
<clever>
and xen has a memory leak in its qemu, so it needs 64bit of swap and bi-yearly reboots of the guests
<clever>
ah
<clever>
ive got most of those private things on local boxes, but i need to merge things
<nekroze>
so containers seem like they will give me enough isolation for my threat model while also sharing resources a bit better
<clever>
i have ~4 servers in my house, lol
<nekroze>
Yeah I wish I had the money to spring for a little personal rack and a few 2u units but I just have the one beasty box at the moment.
<clever>
the only personal thing running outside of the house is a windows vm for some stuff i wrote in 2004 and am heavily neglecting
<clever>
it originaly ran on a win95 laptop, lol
<clever>
but over the years, various hardware failures have pushed it to newer and newer versions of windows
<nekroze>
haha sounds like our networks have gone through similar phases
<clever>
the last mobo failure killed off the xp server, and now its running on win7 in a domu
<clever>
i'm not sure if anybody is even using that service now
<clever>
but i still keep it up, lol
<clever>
its been running for over a decade now...
<nekroze>
I ditched my windows network and domain for the xenserver like 4 years ago because I had had enough
<nekroze>
wow
<clever>
some of that code pre-dates my exposure to linux
<clever>
one of the servers i need to retire did my personal website, but ive already migrated 90% of the stuff out, i need to see whats even reaching it still
<clever>
the bios battery in that box has also died, so it needs an f1 every time it comes back from a power outage
<clever>
hmmm, cacti still goes thru it
<clever>
but it doesnt even host cacti, it proxy's it onward
<nekroze>
lol I had a similar box. but the power where I live isn't always stable so it got too annoying to reconfigure the bios every boot for the XMP memory and such
<nekroze>
clever: well now that you have taken that roadblock away from me I will continue my little weekend sprint to migrate my home server to nixos. I am really enjoying my network the more it runs nixos over other distro's
<nekroze>
thanks again!
<clever>
i only loose the power once or twice a year
<clever>
but 8 days ago, it died in the middle of a snow storm
<nekroze>
well for the fact there is snow where you live (never seen snow myself) and how stable your power is... you get no sympathy from me on that front
<clever>
lol
nekroze has quit [(Quit: Lost terminal)]
spacekitteh has quit [(Remote host closed the connection)]
matthewbauer has joined #nixos
hotfuzz has quit [(Read error: Connection reset by peer)]
hotfuzz has joined #nixos
<newNixos>
ha! was he upset
matthewbauer has quit [(Ping timeout: 248 seconds)]
<NixOS_GitHub>
[nixpkgs] romildo opened pull request #21037: flat-plat-gtk-theme: remove build dependency on gnome-shell (master...upd.flat-plat-gtk-version) https://git.io/v1rB0
MarcelineVQ has quit [(Ping timeout: 260 seconds)]
MarcelineVQ has joined #nixos
ale-batt has joined #nixos
<ale-batt>
hello
<srhb>
ale-batt: Hello.
<ale-batt>
I am having an issue i can't figure out with nixos, could you help me ?
ilja_kuklic has joined #nixos
angerman has joined #nixos
<joachifm>
ale-batt: please just ask the question :) no need to ask to ask
Kingsquee has joined #nixos
<NixOS_GitHub>
[nixpkgs] abbradar pushed 1 new commit to master: https://git.io/v1rBp
<NixOS_GitHub>
nixpkgs/master edaac2c Nikolay Amiantov: wesnoth: disable input frameworks
<ale-batt>
My trackpad on my thinkpad t410 was strange, it moved as tactile move instead as a mouse, i fixed it with services.xserver.synaptics.enable = true
<ale-batt>
but it's still strange, the cursor "crazy"
<mw>
I have no luck running tor-browser in nixos, it complaiins that the tor executable is missing...
<clever>
ale-batt: i never liked how the synaptics driver behaves, and it may help to turn off the matching kernel driver
justbeingglad has joined #nixos
<ale-batt>
clever: matching kernel driver ?
justbeingglad has left #nixos []
<clever>
ale-batt: there should be a kernel module by the same name, that turns it into a touch screen if you lack the xserver half
<clever>
hmmm, dont see that on my laptop
<ale-batt>
i didn't had the issue when i was on the live boot on my usb ...
<clever>
ive got the same issue on my laptop, but i almost exclusively use an external mouse
<clever>
and it happens to have 2 mice, the joystick in the keyboard is un-affected
<clever>
ale-batt: aha, somebody set CONFIG_MOUSE_PS2_SYNAPTICS=y in the kernel config
<ale-batt>
is it : hardware.trackpoint.enable ?
<clever>
so its stuck on
<joachifm>
mw: please open an issue with details
Guest350 has quit [(Ping timeout: 246 seconds)]
<mw>
joachifm: okidoe, thanks
<clever>
ale-batt: i cant see what has caused that to change though
latk has joined #nixos
latk has quit [(Client Quit)]
Itkovian has joined #nixos
MichaelRaskin has joined #nixos
c0bw3b has joined #nixos
kampfschlaefer has joined #nixos
elurin has joined #nixos
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
<Cinnamon811rc>
Hello! Every day I install mc, gparted, etc., and they disappear on the next day. What is the problem? With PATH? How to write configuration.nix such that all included programs would work any time?
<Cinnamon811rc>
Maybe, somebody knows?
<chris|>
question: why is nix.useSandbox not the default? Is there some downside to using it?
<kmicu>
chris|: ‘performance’ and some minor bugs.
<Cinnamon811rc>
via configuration.nix
<kmicu>
Maybe you’ve booted into a previous generation by choosing older boot entry.
<Cinnamon811rc>
but if I write nixos-rebuild switch it does default system like in config for the next boot, or not?
filterfish has quit [(Remote host closed the connection)]
filterfish has joined #nixos
<clever>
Cinnamon811rc: is /boot mounted correctly?
michalrus has quit [(Quit: leaving)]
<Cinnamon811rc>
am I blocked?
<Cinnamon811rc>
I do not see my messages
<clever>
Cinnamon811rc: ctrl+q, though you may not see it
<Cinnamon811rc>
boot is on a separate fat32 disk
<chris|>
kmicu: kmicu from what I understand, the main overhead is setting up the chroot (bind-mounting etc.) and some minor setup, right?
<clever>
Cinnamon811rc: ah maybe its jut lag
<Cinnamon811rc>
because I could not install grub
<kmicu>
Cinnamon811rc: ‘sudo nixos-rebuild switch’ actives your configuration.nix. So if you have mv, gparted, etc in environment.systemPackages then they should be in PATH after ‘sudo nixos-rebuild switch’. That also requires spawning a new fresh shell.
<clever>
Cinnamon811rc: nixos needs to update the bootloader config every time you do a nixos-rebuild
<Cinnamon811rc>
and I boot from another grub from aptosid
<clever>
Cinnamon811rc: and if you dont sync that config up properly, it will undo all changes upon reboot
<kmicu>
Ahh, so it’s not vanilla NixOS :)
<NixOS_GitHub>
[nixpkgs] mimadrid opened pull request #21041: dstat: python2.72: bad interpreter: No such file (master...fix/dstat-0.7.3) https://git.io/v1rzW
<Cinnamon811rc>
:) and how to sync boot correctly? and what is spawning a new fresh shell?
<clever>
Cinnamon811rc: when i did something similar, i used the source command in grub, to load the nixos grub.conf file from the other grub.conf file
<Cinnamon811rc>
aha good that they are visible
<Cinnamon811rc>
I just put menu entries from nixOS grub to aptosid grub and it worked
<Cinnamon811rc>
and update-grub also
<clever>
but it will break again the next time you do any changes to configuration.nix
<clever>
so you need to source the generated grub.cfg file from the other one
<kmicu>
Cinnamon811rc: Ignore my messages, I’ve assumed your are using a vanilla NixOS.
<Cinnamon811rc>
I repeated it
<Cinnamon811rc>
what does it mean source grub?
<Cinnamon811rc>
is there any vanilla nixos?
<clever>
Cinnamon811rc: there is a command in grub, that will read another config file
<clever>
Cinnamon811rc: so you can do soemthing like source (hd0,0)/boot/grub.cfg to load the nixos grub config into the other one
<Cinnamon811rc>
and also how to include all packages to install everything without listing them in config?
<clever>
but you need to know what the hdd path is, so you may want to play with the grub command line first (hit c at grub)
griff_ has quit [(Quit: griff_)]
<kmicu>
Vanilla NixOS in a sense of a main distro on your system — where NixOS has a full control over GRUB.
<Cinnamon811rc>
it startes from the word source?
<clever>
Cinnamon811rc: yes
<Cinnamon811rc>
I see :)
<clever>
i'm a bit fuzzy on the hdd half of things, but you can test it without editing files
<Cinnamon811rc>
one second please, should I write source (hd0,0)/boot/grub.cfg from aptosid? it is another linux
<clever>
Cinnamon811rc: the simplest test is to hit c at the grub menu, then try typing "source (hd0,0)/boot/grub.cfg"
<clever>
and it does have tab-completion
<clever>
so you can poke around and find it with that
<Cinnamon811rc>
I understand:)
joelmo has joined #nixos
<Cinnamon811rc>
instead of (hd0,0) it should be nixos drive?
<clever>
yeah
<Cinnamon811rc>
aha perfect:) You help me every day
<Cinnamon811rc>
thank you very much
<Cinnamon811rc>
see you and have a good day:)
<clever>
and once you have the exact command, you can add it to the proper grub.cfg
<Cinnamon811rc>
so I test what is exacy command should be
<Cinnamon811rc>
and when I find I will put it where?
<Cinnamon811rc>
to another grub?
<clever>
aptosid's grub config
<clever>
probably one of the files update-grub reads
<Cinnamon811rc>
ok good to grub.40
<Cinnamon811rc>
may I ask you which desktop is better for old not supported ATI AMD video card?
griff_ has joined #nixos
<clever>
i use xfce on all of my systems
<Cinnamon811rc>
is it with mouse?
jacob_ has joined #nixos
<clever>
yes
<Cinnamon811rc>
how to change mouse for something?
jacob_ is now known as Guest24759
<clever>
xfce4-settings-manager
<Cinnamon811rc>
:) ok thank you very much, and if you know how to put all packages into the system it would be perfectly good
<Cinnamon811rc>
without listing them
toogley has joined #nixos
<clever>
all packages?
<toogley>
when will nixos move to wayland as default display server?
<Cinnamon811rc>
yes which do not contradict to each other or maybe from some subgroup
<clever>
Cinnamon811rc: it would be a bit insane to install all packages, there are over 10,000 of them
anelson has quit [(Remote host closed the connection)]
edvorg has quit [(Read error: Connection reset by peer)]
anelson has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
filterfish has joined #nixos
<Cinnamon811rc>
maybe not all from subgroups system administrations, all low level utilities, or even everything
filterfish has quit [(Remote host closed the connection)]
<Cinnamon811rc>
I mean how to write config without explicit listing of packages?
edvorg has joined #nixos
filterfish has joined #nixos
<clever>
you can break the config up over many files, and have one for the config like stuff, and one for packages
filterfish has quit [(Read error: Connection reset by peer)]
filterfish has joined #nixos
<Cinnamon811rc>
do you know how to write by command to include all packages from the server nixos to that file?
griff_ has quit [(Quit: griff_)]
<clever>
you would still need to manualy put in each package you want
<Cinnamon811rc>
I see, but hackers can do it I think
<Cinnamon811rc>
not?:)
<clever>
it would be near imposible to get it to not conflict
<clever>
and it would take up at least 80gig of your hdd
<toogley>
ah, i didn't mean wayland as the default display server, rather as the default windowing system
<Cinnamon811rc>
I mean only names
estewei has quit [(Quit: Quit)]
<Cinnamon811rc>
clever thank you for your great help! see you
Cinnamon811rc has left #nixos ["Leaving"]
m` has joined #nixos
TheAppleMan has joined #nixos
TheAppleMan has quit [(Changing host)]
TheAppleMan has joined #nixos
filterfish has quit [(Remote host closed the connection)]
filterfish has joined #nixos
filterfish has quit [(Remote host closed the connection)]
filterfish has joined #nixos
wkennington has quit [(Read error: Connection reset by peer)]
filterfish has quit [(Remote host closed the connection)]
<clever>
ashkitten: the channels are setup to only update once hydra has built everything
nh2_ has quit [(Ping timeout: 258 seconds)]
<angerman>
over and over I'm confused where to find documentation for commands, and the available commands. E.g. where do I find the `fetchurl` documentation?
<angerman>
clever sure. but how do you go about finding "fetchurl"?
<ashkitten>
clever, i added unstable but it's not working it still tries to build from source
filterfish has quit [(Remote host closed the connection)]
<clever>
angerman: i memorized its location in nixpkgs after reading the a decent chunk of it.....
<clever>
ashkitten: can you pastebin the start of the output when you run nix-env?
filterfish has joined #nixos
<clever>
including the command you ran
<angerman>
clever got any idea how to improve discoverability?
<ashkitten>
clever, no i don't have a web browser :P
<clever>
angerman: http://pastebin.com/aRcuQhyp this tells you where pkgs.fetchurl comes from, reading that line gives the path to fetchurl/default.nix
filterfish has quit [(Remote host closed the connection)]
<ashkitten>
oh i see now
filterfish has joined #nixos
<ashkitten>
clever, it says there's a name collision in the input Nix expressions
<clever>
ashkitten: there is a gist command in nix, or nixpaste can do it purely with stock tools, $ <command> | curl -F 'text=<-' http://nixpaste.lbr.uno
<angerman>
clever ohh, so you can ask for the location of a definition?
<clever>
angerman: yep
filterfish has quit [(Read error: Connection reset by peer)]
filterfish has quit [(Remote host closed the connection)]
filterfish has joined #nixos
systemfault has quit [(Quit: Bye!)]
filterfish has quit [(Read error: Connection reset by peer)]
pi3r has joined #nixos
filterfish has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
m` has quit [(Ping timeout: 250 seconds)]
ale-batt has joined #nixos
<ale-batt>
I add my user in the sudoers file, but after a nixos-rebuild it put the user off ... why and how fix it ?
<clever>
ale-batt: nixos wants to control all files in /etc, users.users.clever.extraGroups = [ "wheel" ]; will add me to the wheel group automaticaly
<ale-batt>
clever: so wheel is the sudo group ?
<clever>
yes
<toogley>
hello. i have installed a new system inside virtualbox (i didn't choose the preconfigured images intentionally), but my keymap is always wrong after boot. although i defined the correct one in configuration.nix and updated it with nixos-rebuild switch. here's my configuration.nix : http://ix.io/1Lpx
ashkitten has joined #nixos
<ashkitten>
When I do sudo su, then try to reset the terminal it tells me reset: unknown terminal type rxvt-unicode-256color
osa1 has joined #nixos
<clever>
ashkitten: i always use sudo -i
<clever>
there is no point in mixing different tools that do the same thing
<osa1>
do I have to install nix under /nix ? my / is in another partition with not much space, I'd like to install it in /home/user/nix , but I can't see in the manual a way to do that.
<clever>
osa1: you can make a symlink from /nix to /home/user/nix
<ashkitten>
clever, whatever, that's not the point of my question
filterfish has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
mudri` has joined #nixos
<osa1>
clever: can I then install packages without sudo?
<clever>
ashkitten: did you install rxvt with nix-env?
<ashkitten>
yes
<clever>
osa1: if you have write to /home/user/nix, sure
proteusguy has joined #nixos
<osa1>
clever: nice, thanks
<clever>
ashkitten: the xrvt stuff is probably in /home/ash/.nix-profile, so root cant find it
<clever>
ashkitten: either nix-env -i it as root, or put it into systemPackages
<ashkitten>
I also tried putting it in systemPackages but it still didn't work
<clever>
ashkitten: does it show up in /run/current-system/sw/share/terminfo/u/ ?
mudri has quit [(Ping timeout: 265 seconds)]
michalrus has quit [(Quit: leaving)]
<ashkitten>
no it doesn't
michalrus has joined #nixos
<clever>
ashkitten: thats probably the issue then, did you nixos-rebuild after adding it to systemPackages?
<ashkitten>
yes
<clever>
ashkitten: and its in /home/ash/.nix-profile/share/terminfo/u/ ?
<osa1>
clever: "error: the path ‘/nix’ is a symlink; this is not allowed for the Nix store and its parent directories"
<clever>
osa1: export NIX_IGNORE_SYMLINK_STORE=1
<osa1>
ugh, it's downloading the binary again
<ashkitten>
er, wait it shows up in /run/current-system/sw/share/terminfo/r/
<Dezgeg>
instead of symlink, use a bind mount
<clever>
ashkitten: oh oops, was thinking of urxvt
<ashkitten>
i'm using rxvt but it's called rxvt-unicode
<ashkitten>
i'm using urxvt but it's called rxvt-unicode
<clever>
ashkitten: yeah, thats what tripped me up and lead to the wrong path
<clever>
Dezgeg: i found a recent bug in nixpkgs, it wants config.platform to be a lambda now, there isnt enough () around the change in top-level/default.nix dealing with platform
<clever>
osa1: sounds more like a network error to me, but you can try the bind-mount Dezgeg sudjested, rm /nix; mkdir /nix; mount --bind /home/user/nix /nix
<osa1>
OK
<NixOS_GitHub>
[nixpkgs] joachifm pushed 2 new commits to master: https://git.io/v1rrD
<NixOS_GitHub>
nixpkgs/master 5008161 romildo: flat-plat-gtk-theme: remove build dependency on gnome-shell
<NixOS_GitHub>
nixpkgs/master ff15333 Joachim F: Merge pull request #21037 from romildo/upd.flat-plat-gtk-version...
vandenoever has joined #nixos
<osa1>
reinstalling after mount --bind
<osa1>
do I have to clear /home/user/nix before reinstalling?
filterfish has joined #nixos
<clever>
osa1: dont think you need to
ale-batt has quit [(Ping timeout: 260 seconds)]
<clever>
osa1: another option is to just install it to /nix normally, then move it over and add the bind mount
<osa1>
OK, I think it worked
<osa1>
I guess I have to manually update $PATH for nix binaries?
<clever>
osa1: the script in ~/.nix-profile/etc/profile.d/nix.sh should do that for you
filterfish has quit [(Remote host closed the connection)]
<ashkitten>
clever, so how do i fix the issue with it not finding the terminfo?
<clever>
ashkitten: no idea, i havent done much work with terminfo stuff
<ashkitten>
hm
<NixOS_GitHub>
[nixpkgs] vcunat pushed 2 new commits to staging: https://git.io/v1rrA
<NixOS_GitHub>
nixpkgs/staging af220f2 Vladimír Čunát: cairo: maintenance 1.14.6 -> 1.14.8
<NixOS_GitHub>
nixpkgs/staging c3d20b2 Vladimír Čunát: gtk3: maintenance 3.22.4 -> 3.22.5
Fare has quit [(Read error: Connection reset by peer)]
Fare has joined #nixos
<osa1>
OK, manually sourcing that file worked.
<osa1>
not sure how .profile is used but restrating my shells didn't do it
<clever>
if [ -e /home/clever/.nix-profile/etc/profile.d/nix.sh ]; then . /home/clever/.nix-profile/etc/profile.d/nix.sh; fi # added by Nix installer
<clever>
this is in my .bashrc on my gentoo machine
<osa1>
is this added by nix installer? because on my system (I use zsh by default) it added that line to .profile
<joachifm>
ashkitten: perhaps you need to add the terminfo output as well
<clever>
osa1: i think i moved it to .bashrc after the installed added it to .profile
<ashkitten>
joachifm, what do you mean by that?
<osa1>
OK makes sense
<joachifm>
ashkitten: rxvt_unicode.terminfo
<joachifm>
ashkitten: or specify extraOutputsToInstall = [ "terminfo" ] or whatever
<Fare>
I'm told Nix has OO inheritance implemented somewhere in it; where is it? Things like prototypes taking self and super as parameters. Or was that just the fact that it's possible, but hasn't been done yet?
<clever>
Fare: there is a way to do package overrides like that, and the haskell overrides have been doing it for a while
<Fare>
I see self: super: at places
<ashkitten>
okay that worked i think
<Fare>
I see self: super: appear in python, too. Cool.
<ashkitten>
Okay one last problem
<Fare>
is there a definition somewhere for tying the knot? Where is the Y combinator / the make-instance function ?
<ashkitten>
sudo -E vim /etc/nixos/configuration.nix is not in 256 colors
<clever>
yours runs the editor as root, mine copies the file to /tmp, runs the editor without root privs, and upon exit, copies the file back
<clever>
so it uses config from the correct home, and wont cause root to own your config files and bork stuff
<ashkitten>
I know about sudoedit, but I like to use sudo vim, and I'm curious why it's broken
markus1189 has joined #nixos
<clever>
its possibly the same ~/.nix-profile stuff, need to install things globaly not per-user
<ashkitten>
install -what- stuff though?
<clever>
whatever you installed as to ash's profile to fix it originally
filterfish has joined #nixos
<ashkitten>
*is confused*
filterfish has quit [(Remote host closed the connection)]
<clever>
ashkitten: you will need to review the list in "nix-env -q" and figure out what is fixing the problem, then add that globally
osa1 has quit [(Ping timeout: 246 seconds)]
osa1_ has joined #nixos
<ashkitten>
I already added rxvt_unicode.terminfo globally and it fixed the problem with sudo -i but there's still the issue with sudo vim
<clever>
ashkitten: maybe compare the output of "sudo -E env", "sudo -i ; env", and just env without any root
osa1_ is now known as osa1
osa1 has quit [(Changing host)]
osa1 has joined #nixos
Itkovian has joined #nixos
roconnor has joined #nixos
p_l has left #nixos []
<ashkitten>
$TERMINFO is the same in all 3
systemfault has joined #nixos
iyzsong has quit [(Quit: night.)]
<clever>
ashkitten: what about TERMINFO_DIRS
<ashkitten>
with sudo -E echo $TERMINFO_DIRS it's the same as echo $TERMINFO_DIRS, but with sudo -i; echo $TERMINFO_DIRS it starts with /root instead of /home/ash
<ashkitten>
okay i get it now
<ashkitten>
it's preserving the $TERMINFO_DIRS with -E but that path is wrong for root user
<clever>
that means it will use what ash has nix-env -i'd
<clever>
rather then what root has
<clever>
which should make the problem less likely
<ashkitten>
so the question is how do i preserve some variables with sudo but not all i guess?
<ashkitten>
because otherwise it'll use root's vimrc and other configuration files when i do a sudo command
<clever>
hence why i always say to use sudoedit
<ashkitten>
alright i guess
filterfish has joined #nixos
<clever>
without that, vim will be editing ash's files as root, and files will randomly become owned by root
<clever>
and then ash cant edit his own config files
<ashkitten>
ash is a girl ^^
<ashkitten>
but yeah okay
<clever>
git does the same thing to its main index file a lot
ashkitten has quit [(Remote host closed the connection)]
m` has joined #nixos
ashkitten has joined #nixos
<ashkitten>
Applications are not reading my .gtkrc-2.0?
filterfish has quit [(Remote host closed the connection)]
<k11`>
I can't upgrade nixos from 16.03 to 16.04 :(
Elena_Morova has joined #nixos
<Elena_Morova>
helloo:3 im from Russia) Does anyone want to chat with me?
<k11`>
Failed to stop -.slice: Operation refused, unit -.slice may be requested by dependency only.
<k11`>
remote system hungs out (
<k11`>
from 16.03 to 16.09 of course
lverns has joined #nixos
<k11`>
because I haven't something like a serial console, there is no way to debug it for me :(
<MichaelRaskin>
Isn't there a kernel change and a huge systemd version jump anyway, so the reboot is needed in any case?
filterfish has joined #nixos
<k11`>
MichaelRaskin: the systems failed to reboot to
filterfish has quit [(Read error: Connection reset by peer)]
<MichaelRaskin>
Pity
Elena_Morova has quit [()]
<k11`>
may be need to generate new hardware-configuration.nix
ashkitten has quit [(Ping timeout: 252 seconds)]
<k11`>
I haven't attys too.. :(
filterfish has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
<k11`>
hmm
michalrus_ has joined #nixos
<k11`>
the machine in infinite loop of umounting my nfs mount points
zagy has joined #nixos
michalrus has quit [(Ping timeout: 260 seconds)]
michalrus_ is now known as michalrus
michalrus has quit [(Remote host closed the connection)]
eacameron has joined #nixos
<gchristensen>
anyone around with permission on Hydra to restart release-16.09 jobs?
zraexy has joined #nixos
michalrus has joined #nixos
anelson has quit [(Remote host closed the connection)]
anelson has joined #nixos
zagy1 has joined #nixos
zagy has quit [(Ping timeout: 256 seconds)]
eacameron has quit [(Ping timeout: 260 seconds)]
bennofs1 has joined #nixos
filterfish has joined #nixos
filterfish has quit [(Remote host closed the connection)]
<k11`>
Can I complete upgrading nixos using chroot?
filterfish has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
<MichaelRaskin>
I think so.
<pi3r>
buildPythonPackage does not exist anymore in nixpkgs-unstable ?
<MichaelRaskin>
I think it lives inside pythonPackages
<vandenoever>
is there a nice way to use rust nightly on nixos?
Fare has quit [(Ping timeout: 245 seconds)]
zagy1 has quit [(Ping timeout: 252 seconds)]
<MichaelRaskin>
We have a rustcNightly something (and a separate binary version); but Nix packages are supposed to have a fixed version, so for true nightly you need to update the expression once a day
<MichaelRaskin>
Overriding just the source part would probably work, though
<vandenoever>
MichaelRaskin: should i see that in pkgs/development/compilers/rust/rustc.nix ? I just see RUSTC_BOOTSTRAP = "1";
<vandenoever>
is rustNightly in a branch?
<MichaelRaskin>
there is separate nightly.nix and nixghtlyBin.nix
<vandenoever>
ah got it
<MichaelRaskin>
They are present in NixPkgs master
filterfish has joined #nixos
<pi3r>
MichaelRaskin: that it, thanks a lot
filterfish has quit [(Read error: Connection reset by peer)]
lverns has quit [(Ping timeout: 240 seconds)]
<vandenoever>
hmm, current master gives me cannot coerce a set to a string, at /home/oever/nixpkgs/pkgs/build-support/trivial-builders.nix:7:14 when i do 'nix-shell -p rustcNightlyBin'
<MichaelRaskin>
Add .rustc
<vandenoever>
MichaelRaskin: that does the trick, thanks
<MichaelRaskin>
It is good enough for nix-build, but nix-shell is more strict
zraexy has quit [(Ping timeout: 256 seconds)]
<MichaelRaskin>
The first step in debugging was nix-instantiate --eval-only '<nixpkgs>' -A rustcNightlyBin
m` has quit [(Ping timeout: 240 seconds)]
<vandenoever>
MichaelRaskin: and that also tells me that there is no cargo in there
<vandenoever>
i just tried building something but cargo cannot find crate for std
<MichaelRaskin>
Maybe feeding this to «makeRustPlatform» would give you proper cargo
cfricke has joined #nixos
Fare has joined #nixos
filterfish has joined #nixos
m` has joined #nixos
filterfish has quit [(Remote host closed the connection)]
latk has joined #nixos
<NixOS_GitHub>
[nixpkgs] fpletz pushed 1 new commit to master: https://git.io/v1riX
<NixOS_GitHub>
nixpkgs/master 9074d98 Franz Pletz: linux: add patch to fix CVE-2016-8655...
Shoue has quit [(Ping timeout: 258 seconds)]
afics has quit [(Quit: WeeChat 1.6)]
<gchristensen>
fpletz: that wasn't patched in the mainline? huh...
<fpletz>
gchristensen: nope
<gchristensen>
that seems very strange
<fpletz>
the patch went into the net tree over a week ago…
<NixOS_GitHub>
[nixpkgs] fpletz pushed 1 new commit to release-16.09: https://git.io/v1riQ
<NixOS_GitHub>
nixpkgs/release-16.09 90f82d2 Franz Pletz: linux: add patch to fix CVE-2016-8655...
<gchristensen>
well as soon as release-16.09 figures out how to build the kernel again (it has been failing due to a race condition) I'll send out an announcement
m` has quit [(Ping timeout: 245 seconds)]
mizu_no_oto has joined #nixos
<joachifm>
the race condition thing is odd, it's supposed to have been fixed upstream a long time ago
filterfish has joined #nixos
<gchristensen>
also, I didn't think hydra ran with --cores set to >1?
filterfish has quit [(Remote host closed the connection)]
<toogley>
i accidentally deleted my /etc/nixos/configuration.nix - what is the easiest way to revert the file?
<fpletz>
I've been super-busy the last few days, now the challenge is to clean everything up and make pull requests… security improvements for nixos-containers (bind-mount only needed store paths), build everything with gcc 6, support for odroid xu4 and refactoring of sd image expressions \o/
m` has joined #nixos
<gchristensen>
wowsa, fpletz, nice!
<gchristensen>
I ... spent a few minutes thinking about how to make my announcement emails better? :|
<gchristensen>
fpletz: do you know if odroid xu4 support also implies c2 support?
yonk has joined #nixos
ebzzry_ has quit [(Ping timeout: 264 seconds)]
<yonk>
Can someone tell me if there is a speficic package I need for section 3 of the man pages (the C library)? I'm missing it on my install
<fpletz>
gchristensen: nope, the architecture of the c2 is arm64 (or aarch64, armv8), I'm currently bootstrapping nixpkgs/nixos for that but compilation takes ages
<yonk>
clever: man-pages did the trick, the other two matched no derivations. thanks a bunch!
<fpletz>
gchristensen: the xu4 is a regular armv7 (32-bit) with 8 cores, so compatible with rpi2
bennofs has joined #nixos
<bennofs>
What would you say is the minimum required disk space for NixOS? I feel like less than 100GB is quite difficult for NixOS on Desktop
<bennofs>
(assuming you install at least a few qt/gtk apps)
<clever>
yonk: all of those are attribute paths, not names
<clever>
yonk: so you need to use -A with them
<fpletz>
we're also building armv7 now on our mf hydra, I'm still in the process of fixing some builds but ideally I wanto to build arm packages on the official hydra
cfricke has quit [(Quit: WeeChat 1.6)]
<fpletz>
gchristensen: my personal release goal for 17.03 is arm as a first-class architecture ;)
<clever>
fpletz: i have a nix expression here, that generates a kernel, initrd, and root.squashfs, which boots a rpi3 into a 40mb ramdisk, that is capable of doing nix builds
bennofs1 has quit [(Ping timeout: 250 seconds)]
Shoue has joined #nixos
<fpletz>
clever: yeah, I checked your code a few weeks, no-os or something? :)
<fpletz>
+ago
<gchristensen>
fpletz: cool: D
<clever>
fpletz: yeah, not-os, ive done some fairly large changes recently that i havent had a chance to clean up and push
proteusguy has quit [(Ping timeout: 250 seconds)]
<NixOS_GitHub>
[nixpkgs] LnL7 closed pull request #20940: mbedtls: fix build for Darwin (master...fix-mbedtls-darwin) https://git.io/v1B1k
<gchristensen>
fpletz: well if you want, I can set you up with console access to my c2 :P
<gchristensen>
(proper serial console, so you're not worried about breaking network)
<clever>
fpletz: mainly, i added ntp to solve some jan 1st issues, made the store writable via overlayfs, and added dhcp to the initrd to tftp the rootfs
osa1_ has joined #nixos
<clever>
fpletz: and then just because i could, i added plymouth to it, lol
osa1_ has quit [(Client Quit)]
filterfish has joined #nixos
filterfish has quit [(Remote host closed the connection)]
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
<k11`>
How can I chroot to nixos system root?
<clever>
k11`: there is a nixos-install --chroot command
<clever>
k11`: but the main thing to keep in mind, is that /nix/var/nix/profiles/system/sw/bin has to be in $PATH
<clever>
and mount everything up like you would for any chroot
<k11`>
clever: ok
<fpletz>
gchristensen: do you have a ssd which you can connect to the c2? building on mmc/sd will destroy them prettry quickly and some gigs of swap are needed
glines has joined #nixos
lverns has joined #nixos
<fpletz>
clever: nice, my priority right now is to get a full nixos running though ;)
Seichi has joined #nixos
<clever>
fpletz: my main thought for the above, was to get a quick&dirty build slave up, that can double the speed at which i build proper nixos (2 rpi's working together)
<gchristensen>
fpletz: hmm maybe. if you're interested, I am happy to work to make it available for you.e
<clever>
gchristensen: another option that ive done on this end, iscsi
<clever>
fpletz: and tgt is a bit confusing, there are 3 parts to it
<nathan7>
ftr kernelAutoModules actually breaks booting for me
<nathan7>
idk why
<Dezgeg>
yeah, I have observed that as well for armv7 when I tried to enable that
<clever>
fpletz: tgtd is the daemon, it doesnt support any config files and comes up blank, tgtadm is a remote-control to alter things in the daemon, it also has no config files
redmq has quit [(Ping timeout: 264 seconds)]
Gaff3 has joined #nixos
<clever>
fpletz: then you have tgt-admin, a perl script that can read targets.conf, run tgtadm, and configures everything
<Dezgeg>
maybe enabling all the possible things even as modules bloats the kernel itself too much
* nathan7
drops that commit
ashkitten has joined #nixos
<nathan7>
perhaps
<nathan7>
I'm booting with UEFI GRUB
<nathan7>
and I have a good 128G of RAM
<nathan7>
so… should work out, I think?
<fpletz>
clever: thanks a lot!
<k11`>
Can installer recover existing system using existing config and nix/store?
<fpletz>
nathan7: out of curiosity, how expensive is that machine? :)
<Dezgeg>
yes, but typically at least in the u-boot world someone hardcodes the kernel to load at address X and e.g. the dtb at address X+32MB
<clever>
k11`: yeah, nixos-install --chroot or just run nixos-install with everything mounted to re-build from configuration.nix
<nathan7>
Dezgeg: I doubt GRUB does that
<joko>
Hello, I am packaging a shell script which depends on wget, what would be the proper way to define wget's path?
<Dezgeg>
I hope not
<nathan7>
joko: I tend to use substituteAll
eacameron has joined #nixos
<nathan7>
joko: so I have @wget@/bin/wget in the script itself
<Gaff3>
Hello, I have a quick noob question if anyone has a second.
<joko>
nathan7: thanks
<nathan7>
fpletz: the single-threaded perf is not amazing, but for things like kernel builds, parallelism swamps the influence of that
<gchristensen>
oh wow
<gchristensen>
we need some nixops-packet support please
<nathan7>
gchristensen: WIP
<Gaff3>
I'm trying to setup a devEnv. If I have code/default.nix, code/projectA/default.nix, code/projectB/default.nix, how can I get nix-shell to run both A and B and drop me into a shell?
* gchristensen
has the vapors
Guest350 has joined #nixos
Ralith has joined #nixos
<Ralith>
pi3r: afaik, yes, you do
Shell has joined #nixos
<Shell>
joko: you can create a wrapper using https://nixos.org/wiki/Nix_Runtime_Environment_Wrapper to wrap the shell script and add stuff to PATH, or otherwise you can patch the shell script to call @wget@/bin/wget and substituteAll it. the latter is normally preferred if reasonably possible.
<Ralith>
pi3r: your current derivation produces an executable; an executable is not a shell environment containing that executable
<nathan7>
gchristensen: like, I have basically everything working at this point
<gchristensen>
nathan7: do you work there?
<nathan7>
gchristensen: no, I run lsq.io, the Packet folks are just good friends of ours
<nathan7>
gchristensen: and we're porting our platform over, which is built entirely on NixOS
<gchristensen>
very cool
<MichaelRaskin>
Gaff3: maybe write an expression that does a buildEnv and use this expression?
<gchristensen>
nathan7: does LSQ want to sponsor a type-2a build box for Hydra? :)
oida has quit [(Quit: WeeChat 1.0.1)]
<nathan7>
gchristensen: right now my account is unbilled
<gchristensen>
so maybe 4-5 type-2a's then, sounds good
<nathan7>
gchristensen: and I'll talk to Zach about providing some boxes to the NixOS foundation or w/e
<gchristensen>
that would be very cool
<nathan7>
gchristensen: I've already mentioned my intention of supporting NixOS officially on Packet to them
eacameron has quit [(Ping timeout: 248 seconds)]
<Gaff3>
MichaelRaskin: thanks, I'll try that
<gchristensen>
well let us know if there is help we can provide ... :)
<fpletz>
nathan7: awesome, thanks for the link and your efforts!
<nathan7>
gchristensen: like, even if literally nobody else in the world cares about this, it'll happen
bennofs1 has quit [(Ping timeout: 260 seconds)]
<gchristensen>
I care
<nathan7>
any public support we can demonstrate just strengthens my argument
<gchristensen>
exactly
<nathan7>
right now, unpatched 4.8 kernels appear to work
<nathan7>
I just have to finish setting up bonding
<nathan7>
I've been trying to get 4.4 to work, but the Ubuntu 4.4 kernels for AArch64 backport a bunch of stuff
<gchristensen>
anything I can do to express support for this idea, to packet directly?
<Dezgeg>
where's the DTB coming from for that platform? or is it really pure UEFI?
<nathan7>
show up in #packethost? idk
<nathan7>
Dezgeg: it's pure UEFI
<Dezgeg>
huh, interesting
filterfish has joined #nixos
<nathan7>
most of the server-class AArch64 boards are UEFI
<nathan7>
like this straight-up has an American Megatrends bootsplash
<Dezgeg>
yeah
filterfish has quit [(Remote host closed the connection)]
<jluttine>
i don't have permissions to mount android with mtpfs. i can mount it as root user. any ideas what group i should add to my user account? also, are there any recommended groups a user account should be added to?
<Dezgeg>
all the consumer stuff will probably still need kernelDTB = true;
<nathan7>
Dezgeg: so I turned that off
<nathan7>
Dezgeg: because there *are no DTBs in the tree*
<MichaelRaskin>
jluttine: do you have access to /dev/bus/usb entries?
<Dezgeg>
hm, that sounds like a bug somewhere
<nathan7>
Dezgeg: and the build contains `cp asdf/dtb/* someplace`
<nathan7>
Dezgeg: which becomes `cp someplace`
<nathan7>
Dezgeg: which obviously errors out
<Dezgeg>
ah right, the directory structure is probably different (because why standardize on anything)
<jluttine>
MichaelRaskin: `ls /dev/bus/usb/001` works
<fpletz>
uefi is indeed very convenient, my next challenge is to generate a sd image for the odroid-c2 with hardkernels binary blobs and get the u-boot signed with their x86-64 only signing binary :/
<nathan7>
fpletz: that's what we have the whole build host thing for, don't we?
<MichaelRaskin>
jluttine: you want ls -l
<nathan7>
fpletz: alternatively, fire up qemu-user q=
<MichaelRaskin>
You want not just be able to list the entries, but I guess you also need write access
<nathan7>
Dezgeg: possible, I haven't looked too closely
<Dezgeg>
yeah, I can tackle that one
Gaff3 has quit [(Ping timeout: 260 seconds)]
<jluttine>
MichaelRaskin: ok. i'm not sure how to test that.. `ls -l /dev/bus/usb/001` works as well, but i guess that's not relevant..
<MichaelRaskin>
It should show yout the owning group and permissions
glines has quit [(Ping timeout: 264 seconds)]
Guest350 has quit [(Ping timeout: 246 seconds)]
<fpletz>
nathan7: yeah, sure! I'm just complaining that we need to support so many weird boot setups in the first place because arm doesn't have a standardized boot process
<MichaelRaskin>
If the entries inside 00?/ are group-writeable and the group is usb or something like that, you can just add you there (and relogin)
<fpletz>
but well, first world problems :)
<Dezgeg>
well, uefi won't solve the problem of vendor-specific boot blobs either
<nathan7>
Dezgeg: how so?
<nathan7>
oh, on that note
<jluttine>
MichaelRaskin: root:root, no other groups
<MichaelRaskin>
Hm.
<nathan7>
our GRUB UEFI shit is super broken on non-x86
<nathan7>
GRUB will autodetect that UEFI is the only possible target on AArch64
<nathan7>
so it works
<MichaelRaskin>
I actually wrote a sudo-enabled script to chmod/chgrp/chown some of this when needed, but you can also add extra udev rules (which is the right way to use in general)
<Dezgeg>
because the UEFI implementation itself will probably be loaded from a vendor-specific location from a SD card or something
m` has quit [(Ping timeout: 258 seconds)]
<nathan7>
but if I actually turn UEFI on in the NixOS config, it blows up
<NixOS_GitHub>
[nixpkgs] xvapx opened pull request #21047: steam: added documentation to nixpkgs manual (master...fix/steam-doc) https://git.io/v1r1n
bennofs1 has joined #nixos
<nathan7>
Dezgeg: I'm fairly certain this is onboard flash
<nathan7>
Dezgeg: but these boards are custom, so I don't actually know
<Dezgeg>
well sure on the expensive fancy devices :)
<nathan7>
I primarily concern myself with server-class systems, so I'm a little biased
<nathan7>
though my initial impetus for playing with this was bringing my PINE64 to life
<clever>
in theory, proper UEFI support could be added to it
<nathan7>
oh god, the pi's bootstrap sequence
<nathan7>
that shit is straight from hell
<k11`>
clever: unfortunately executing nixos-install from installer ISO didn't help me to recover my system
<nathan7>
not going aaaanywhere near it
Shoue has quit [(Ping timeout: 260 seconds)]
ashkitten has quit [(Ping timeout: 258 seconds)]
<clever>
nathan7: ? lol
<nathan7>
clever: the goddamn Broadcom graphics chip is what actually bootstraps the CPU
<nathan7>
iirc
<nathan7>
making an undocumented proprietary piece of wtf your root of trust
<k11`>
clever: It seems that problem is unique for me...
<clever>
nathan7: yep, a boot rom on the VC4 dual-proc chip is what bootstraps the system
bennofs2 has joined #nixos
<clever>
nathan7: but the above github is an open-source implementation of everything after the bootrom
<jluttine>
MichaelRaskin: ok.. i've never had to modify any udev rules in other distros i've used. so i'm wondering am i just missing something trivial which would make this work automatically
bennofs1 has quit [(Ping timeout: 258 seconds)]
glines has joined #nixos
ambro718 has joined #nixos
filterfish has joined #nixos
<clever>
nathan7: there are also some interesting features on the rpi from a security point of view
filterfish has quit [(Remote host closed the connection)]
<clever>
nathan7: there is a 2nd MMU layer, between what the arm thinks is physical, and the real ram
<nathan7>
…under the control of what, exactly
<clever>
nathan7: this can allow the gpu firmware to lock out sections of ram and control registers, so you can protect keys from the arm cpu
<nathan7>
because this sounds like the setup for subvirt-style attacks
<clever>
nathan7: its just a normal memory mapped IO register, that all cores (gpu and arm) have access to
<clever>
nathan7: but if used correctly, the gpu firmware will block the arm from having access
<MichaelRaskin>
jluttine: grepping NixOS source for USB, it looks like for some reason nobody cared enough to add such a thing
<nathan7>
clever: I'm kind of more excited about having TrustZone and stuff on server-class systems q=
<k11`>
Hmm, now I see warnings in boot log on those lines:
<clever>
nathan7: i believe it has no effect on the VC4 cores, so an open source gpu firmware can just turn such a thing off, it to setup an open-source hypervisor on another core
<jluttine>
MichaelRaskin: ok, thanks! :/
eacameron has joined #nixos
<clever>
k11`: are the things in question symlinks or directories?
ashkitten has joined #nixos
<MichaelRaskin>
jluttine: I think you want to add this to services.udev.extraRules
<k11`>
It seems like permissings problem, but I don't know the actual paths
<ashkitten>
How can i enable grub on efi?
<k11`>
permissions*
<nathan7>
clever: "open source GPU firmware" is veeeery far up the complexity scale
<clever>
k11`: ah, nixos is capable of fully recreating /etc on bootup, so try just renaming /etc from the installer and reboot
stepcut has quit [(Remote host closed the connection)]
<nathan7>
I wish we'd separate state more carefully
<clever>
nathan7: it doesnt have to actualy implement gpu things, you can just ignore all the gpu features and treat it as a dedicated hypervisor core
stepcut has joined #nixos
<nathan7>
so that I'd have a dir with my SSH keys and all
<nathan7>
clever: I suppose, yeah
<nathan7>
clever: but GPU cores aren't exactly regular CPU cores, and they also tend to come with a tonne of errata
<clever>
nathan7: the rpi3 has 4 arm cores (64bit capable), 2 VC4 cores (no mmu, normaly runs an rtos called threadx), and 12 V3D cores (the shaders)
<nathan7>
clever: mhm
<MichaelRaskin>
jluttine: the minimal is probably SUBSYSTEM="usb", GROUP="usb"
<NixOS_GitHub>
[nixpkgs] zimbatm pushed 1 new commit to master: https://git.io/v1rML
<clever>
nathan7: and nixos doesnt obey that file, so networking doesnt come up
eacameron has quit [(Remote host closed the connection)]
<nathan7>
clever: use my stuff, it's less shit than the nixos-in-place script that is super fucking broken
<nathan7>
clever: (but the authors don't give a fuck, so I gave up on getting something better merged)
eacameron has joined #nixos
edvorg has quit [(Ping timeout: 250 seconds)]
<clever>
nathan7: ah, that would work for the main os, but not for the tar idea, the tar'd nixos will have to mount a 2nd rootfs (the real one) and steal the file out of the host
<nathan7>
clever: no.
<nathan7>
clever: this does not rely on that
<clever>
or is it doing something funny to get the ip's?
<nathan7>
clever: there is a metadata service on a link-local IP
<clever>
ah, nice
<clever>
AWS has something similar, after you get an ip via dhcp
<nathan7>
clever: I read the metadata (with a Go program, not a bloody bash script), and produce a systemd-networkd config file
<NixOS_GitHub>
[nixpkgs] zimbatm pushed 1 new commit to master: https://git.io/v1rMN
<nathan7>
clever: the nixos-in-place code messes up the subnet, which really threw me for a loop
<nathan7>
clever: because I couldn't reach certain parts of the internet
<clever>
nathan7: i'm sort of responsible for nixos-in-place, it was written after i told somebody about how i mutated my gentoo netbook into nixos, lol
<k11`>
Hmm
<nathan7>
clever: my only missing feature is setting up the "anchor IP"
eacameron has quit [(Ping timeout: 240 seconds)]
<nathan7>
clever: which is necessary for floating IPs to work out
<clever>
nathan7: but i feel that the tarball idea is much cleaner, nixos isrunning from a ramdisk, so you are free to format the hdd and do a clean install
<nathan7>
clever: yeah
<clever>
nathan7: floating ip's?
<nathan7>
clever: similar to EC2's elastic IPs
<k11`>
systemd/system-generators symlinked to authorized_keys file not a directory :(
<clever>
ah
<gchristensen>
clever: it is dangerous showing people scare stuff like that :P
<gchristensen>
you never know what evil they'll do
<clever>
nathan7: and that reminds me, ec2 doesnt actualy put the public ip on the box, i think they always get a private ip, and something else sort of does 1 to 1 NAT
<nathan7>
clever: yeah
<nathan7>
clever: I do similar things in my personal networks (border routers do 1:1 NAT)
<clever>
so when you change the elastic ip, the box's ip doesnt have to change
<toogley>
c0bw3b: ah, thanks
osa1 has joined #nixos
filterfish has joined #nixos
<clever>
nathan7: this also reminds me of a datacenter i once used in costa rica, one day the server just went down, but i could still reach it over the vpn
filterfish has quit [(Read error: Connection reset by peer)]
<osa1>
I'm following the "hello" package tutorial in nix manual but I'm confused. do I have to edit a global package db to be able to build the package? I've created .nix and .sh (builder) files.
<clever>
nathan7: and further investigating, revealed it had a 172.16.0.0/20 ip address
filterfish has joined #nixos
<clever>
nathan7: and with some more proding via another server in the same datacenter, i discovered that it shared a broadcast domain with things that had public IP's
<nathan7>
clever: heh
<fpletz>
clever: lol
<clever>
nathan7: so the dhcp server derped and gave my server a private ip, which somehow worked (they must have configured it to allow both)
<clever>
nathan7: and the help line wasnt able to figure out what was wrong
<clever>
i fixed it by having dhcpcd renew the leasse, and it got a public ip
filterfish has quit [(Read error: Connection reset by peer)]
<clever>
nathan7: i have a feeling that i could probably arp poison the switches in that datacenter...
<nathan7>
clever: my personal infra has static IPs assigned at the hypervisor level (so the VMs get it by DHCP / other autoconf)
mudri` has quit [(Ping timeout: 258 seconds)]
<nathan7>
clever: and then 1:1 NAT for public IPs
<clever>
heh, i dont really have any hypervisors active on my LAN
k11` has left #nixos []
isHavvy has joined #nixos
Havvy has quit [(Ping timeout: 268 seconds)]
<nathan7>
clever: my personal infra is largely hosted on a pair of SmartOS boxes running a mixture of zones and actual hardware VMs
<clever>
nathan7: ah, mine is scattered over a couple consumer desktops, and 1 old server grade box
eacameron has joined #nixos
<nathan7>
clever: I'm in the rare situation of having a friendly colo provider (coloclue.nl) with BGP peering and other neat goodies
<clever>
heh
<clever>
the most i have here is fiber internet
<clever>
my router is a dual-processor server with 3gig of ecc ram, it runs hydra and does NAT for the house
<clever>
it came out of an xray machine
<clever>
everything else is just regular consumer desktops, mostly old crap
jedai42 has quit [(Read error: Connection reset by peer)]
<clever>
nathan7: you ever get to play with multicast?
eacameron has quit [(Remote host closed the connection)]
<clever>
nathan7: heh, in my case its over a controled private network
<nathan7>
clever: I'm not sure what I'd even multicast tbh
<clever>
nathan7: my ISP does phone+tv+internet over the same fiber network
<nathan7>
clever: same here for at home
<clever>
nathan7: every tv channel is its own multicast group
<nathan7>
I haven't actually bothered to get rid of my ISP's router because the factory FritzBox firmware isn't all that bad
<clever>
nathan7: the fiber modem spits out 802.1q tagged packets, 1 vlan for the internet, 1 vlan for the tv service (where the multicast happens)
<nathan7>
yeah, same thing here
<NixOS_GitHub>
[nixpkgs] LnL7 opened pull request #21048: wxmac: fix build with (staging...darwin-wxmac) https://git.io/v1rDD
<clever>
nathan7: and my isp router is horid, there is no way to turn the wifi<->wired isolation off
<clever>
so nfs and ssh are entirely useless from a laptop
<nathan7>
PPPoE over one VLAN, idkwtf over the other
<clever>
my isp has bare dhcp and ipv4 over the internet vlan
<clever>
and i believe bare dhcp+ipv4 (in the 10.0.0.0) + multicast for the tv vlan
filterfish has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
<clever>
i did manage to configure nixos to handle both vlans, and i reverse engineered the routing tables in the ISP router (i put a linux box on both the wan and lan ports, and just brute-forced it)
<clever>
but i have never gotten nixos to route the multicast traffic from WAN->LAN correctly
xwvvvvwx has joined #nixos
<clever>
and to make debuging more fun, the STB's cheat for the first 30 seconds, they stream over tcp, to make it tune faster, then switch to multicast as the igmp kicks into gear
<xwvvvvwx>
Hey so I want to make a little fix on vscode
<xwvvvvwx>
So I made a nix-shell with the build dependencies
<xwvvvvwx>
modified some code
<xwvvvvwx>
and then followed the steps to build a version of vscode from source
<xwvvvvwx>
everything runs fine, no errors, except for at the very end when I try to run the executable I get an error "No such file or directory"
ale-batt has joined #nixos
<xwvvvvwx>
I can see the file
<nathan7>
yes
<nathan7>
that means it's missing dynamic libraries or the loader
<nathan7>
vscode is Electron-based, no?
<ale-batt>
How can i install make ? i add in my configuration.nix but it's not exist ...
<nathan7>
and I'm guessing you're using a prebuilt Electron binary because building that takes an incredibly long time
<nathan7>
ale-batt: gnumake
<xwvvvvwx>
yes
<ale-batt>
thx !
<xwvvvvwx>
this is the output from file:
<xwvvvvwx>
code-oss: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.26, BuildID[sha1]=551a54ca2de5272e95f06d3912e628175c46fd3b, stripped
<nathan7>
xwvvvvwx: you can try patchelf-ing it into submission
<clever>
ale-batt: you generaly never install tools like make on nixos, you use nix-shell
<xwvvvvwx>
ok
<xwvvvvwx>
that was my question how to do that
<nathan7>
I thiiink we might have some electron stuff in nixpkgs
<xwvvvvwx>
in the vscode derivation they are doing some patchelf stuff
<nathan7>
xwvvvvwx: you can just .overrideDerivation (drv: { patches = (drv.patches or []) ++ [ ./xwvvvvwx-fixes-stuff.patch ]; })
<LnL>
xwvvvvwx: the fetchelf --set-interpreter should fix that
<clever>
xwvvvvwx: the fixupPhase on line 61-68 handles the patchelf'ing
<xwvvvvwx>
ok so thats not entirely clear to me
<xwvvvvwx>
I have a binary that I need to patch so that it can find it's dynamic libs right?
<nathan7>
yeah
<xwvvvvwx>
I need to write a new nix expression to do that?
<nathan7>
no
<nathan7>
xwvvvvwx: you could generate a patch file for your modifications to the source code
<xwvvvvwx>
ok what is a patch file and how do I make one?
<nathan7>
xwvvvvwx: .overrideDerivation lets you modify a derivation
<nathan7>
diff -ru
<nathan7>
so like
<nathan7>
generally what I do is extract the source to a thing (something like tar xf $(nix-build '<nixpkgs>' -A vscode.src))
<nathan7>
ending up with a dir like vscode-1.2.3 or something
<nathan7>
I rename that to vscode-1.2.3-orig, and extract it again
<nathan7>
and modify the non-orig one, then run diff -ru vscode-1.2.3{-orig,} > my-modifications.patch
<xwvvvvwx>
ok that feels a bit cumbersome to run every time I change a few lines of javascript
eacameron has quit [(Ping timeout: 268 seconds)]
<xwvvvvwx>
Is there a quicker way?
<nathan7>
oh, no
<nathan7>
this is what I do when I'm done messing around and I want to have an actual package I can integrate into my confi
<nathan7>
g
ale-batt has quit [(Ping timeout: 260 seconds)]
<nathan7>
though it's generally quick enough that I do end up using the same process to build test versions
<xwvvvvwx>
ok
eacameron has joined #nixos
<xwvvvvwx>
I guess I could wrap it up in a shell script and it's not the worst thing in the world
<xwvvvvwx>
isn't there just some patchelf command I can run ?
<nathan7>
literally the one in there
<xwvvvvwx>
yes but this line doesn't work: --set-rpath "${atomEnv.libPath}:$out/lib/vscode"
<xwvvvvwx>
what should I replace the variables with?
civodul has joined #nixos
<nathan7>
nixpkgs.atomEnv.libPath
<NixOS_GitHub>
[nixpkgs] dezgeg pushed 1 new commit to master: https://git.io/v1ryy
<NixOS_GitHub>
nixpkgs/master bdab6fe Tuomas Tynkkynen: kernel: Use built-in dtbs_install target instead of rolling our own...
<xwvvvvwx>
hah
mudri` has joined #nixos
<xwvvvvwx>
ok easy
<nathan7>
nix-instantiate '<nixpkgs>' -A atomEnv.libPath --eval-only
<nathan7>
will give you the value of that
<xwvvvvwx>
nice, and the ${out} can just be the directory of the binary?
eacameron has quit [(Remote host closed the connection)]
<nathan7>
wherever the library dir is, I guess?
<nathan7>
ah, it's the same dir, yeah
<xwvvvvwx>
thanks!
<xwvvvvwx>
ok
<xwvvvvwx>
so I ran `patchelf --set-interpreter "$(cat $NIX_CC/nix-support/dynamic-linker)" --set-rpath "${nix-instantiate '<nixpkgs>' -A atomEnv.libPath --eval-only }:." ./code-oss`
<xwvvvvwx>
and it did something
<xwvvvvwx>
but now when I try to launch vscode I get ./code-oss: error while loading shared libraries: libgtk-x11-2.0.so.0: cannot open shared object file: No such file or directory
osa1 has quit [(Ping timeout: 264 seconds)]
<xwvvvvwx>
Am I just missing something from my build environment?
filterfish has joined #nixos
lverns has quit [(Ping timeout: 245 seconds)]
kuznero has joined #nixos
<kuznero>
Hi All!
eacameron has joined #nixos
<kuznero>
I am almost complete noob with NixOS. Though I've got quite a bit of concepts. Is there pulished Xfce configuration available somewhere? I mean as good as the one e.g. provided with Manjaro?
<ashkitten>
On Arch I use xf86-input-evdev-debounce because my mouse's middle button clicks several times in a few miliseconds. How can I use this patch on NixOS?
slack1256 has joined #nixos
<Dezgeg>
nathan7: I think with this change https://git.io/v1ryy the kernelDTB = true; should work
eacameron has quit [(Ping timeout: 265 seconds)]
<nathan7>
Dezgeg: nice, will cherry-pick that in a bit
<kuznero>
Or maybe there is a way to "clone" Manjaro settings from ~/.config/xfce4 somehow so that it is reproducible with NixOS?
hotfuzz has quit [(Read error: No route to host)]
hotfuzz has joined #nixos
slack1256 has quit [(Remote host closed the connection)]
<kuznero>
Ok, maybe you can guide me about where I can find detailed documentation on how I can configure xfce in configuration.nix?
filterfish has quit [(Remote host closed the connection)]
<c0bw3b>
other than that the xfce-specific config file from your Manjaro can probably be duplicated to your nixos xfce
eacameron has joined #nixos
<kuznero>
c0bw3b: thanks!
eacameron has quit [(Ping timeout: 245 seconds)]
filterfish has joined #nixos
eacameron has joined #nixos
eacameron has quit [(Ping timeout: 264 seconds)]
lsix has joined #nixos
eacameron has joined #nixos
Itkovian has joined #nixos
<kuznero>
c0bw3b: there are only 4 options available... is there a github repository where I can find code itself such that I can just inspect what to experiment with?
eacameron has quit [(Ping timeout: 258 seconds)]
<kuznero>
Is there any way to configure startup apps in xfce through nix config?
zraexy has quit [(Ping timeout: 260 seconds)]
filterfish has quit [(Remote host closed the connection)]
<NixOS_GitHub>
[nixpkgs] LnL7 opened pull request #21049: db: disable dbm support by default (staging...db-dbmsupport) https://git.io/v1rQJ
Rotaerk has quit [(Quit: Leaving)]
<NixOS_GitHub>
[nixpkgs] joachifm pushed 5 new commits to master: https://git.io/v1rQl
<NixOS_GitHub>
nixpkgs/master 19b9617 Joachim Fasting: couchdb service: fix test in preStart...
<NixOS_GitHub>
nixpkgs/master fafb665 Joachim Fasting: syslogd service: assert conflict with rsyslogd...
<NixOS_GitHub>
[nixpkgs] offlinehacker opened pull request #21050: chromium module: add support for chromium policies as nixos module (master...nixos/programs/chromium/add) https://git.io/v1rQ8
kuznero has quit [(Quit: leaving)]
alx741 has joined #nixos
redmq has quit [(Ping timeout: 264 seconds)]
oida has joined #nixos
filterfish has joined #nixos
eacameron has joined #nixos
<gchristensen>
clever: ping
<clever>
gchristensen: pong
<gchristensen>
clever: I'm ready to try the take-over ramdisk
<rotaerk>
kmicu, well I get that nixpkgs is periodically built/tested and these updates are published on channels but.. I don't understand how these channels relate to the nixpkgs-channels github
<rotaerk>
it's got source, not built packages
<rotaerk>
is it just that nixpkgs-channels are snapshots of the successfully-built versions of nixpkgs?
<LnL>
clever: or if you decide you want to use zfs you can just just kexec, partition, nixos-rebuild and done :D
<rotaerk>
but the channels revisions are versions of nixpkgs aren't they?
<clever>
LnL: yep, but the GPU breaks when you kexec
<clever>
LnL: so you need lan or serial console
<gchristensen>
sounds like rotaerk neneds an explanation of our hybrid source/binary release model
<rotaerk>
hmm I'm aware that you can download binaries as a substitute for building from source locally
<rotaerk>
and that hydra builds these binaries
<clever>
rotaerk: the nixos channels are versions of nixpkgs that have passed all nixos tests, and probably wont brick a nixos machine
<clever>
rotaerk: the -small channels are ones that have only passed testing, the non-small channels have had hydra attempt to build every single package in nixpkgs
<kmicu>
‘nixpkgs-channels’ repository is a ‘silly’ repository that allows you to find out a revision hash of each channel.
filterfish has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
eacameron has quit [(Ping timeout: 258 seconds)]
<rotaerk>
so is nixpkgs-channels a subset of nixpkgs then, a sampling of its history, but not the full history (only the moments in history that passed tests)
<gchristensen>
kmicu: silly?
<gchristensen>
clever: this takes ages to build
<clever>
gchristensen: i would have just made it branches on the main nixpkgs
<kmicu>
So I know that e.g. ‘nixos-unstable’ channel is on ‘Latest commit 1c50bdd 3 days ago ‘.
<clever>
gchristensen: it has to rebuild kexectools to fix a bug caused by hardening, and systemd depends on kexec, and util-linux depends on systemd, and half the os depends on util-linux
<clever>
gchristensen: i need to file that PR and get hydra to help out
<kmicu>
gchristensen: why do we need a *separate* repository to track that? Is it not silly?
<gchristensen>
ah, I see kmicu
<LnL>
permissions perhaps?
<kmicu>
Any reason against not having ‘nixos-unstable’ branch in NixOS/Nixpkgs?
<LnL>
I think it's only hydra that pushes to the channels
<rotaerk>
oh I think I see... nixpkgs-channels just houses branches of nixpkgs for each channel
<clever>
LnL: i think its because the release-16.09 and release-16.09 branches collide (one is to push security to 16.09 and let hydra test, the other is the real channel)
m0rphism has joined #nixos
<LnL>
or that
<clever>
but that poor naming leads to people using the wrong release-16.09 anyways
<clever>
because it says release, it must be the stable one!
<gchristensen>
holy hell this builds a lot of expensive-to-build stuff. please do put in that PR :P
<clever>
gchristensen: i'm a bit iffy about just blanket disabling the hardening in gcc, but its a tool that only root will ever use, and it just does nothing without root
<sphalerite>
isn't the channel one called nixos-16.09?
<gchristensen>
also it is pulling in cups :|
<clever>
kmicu, LnL: what are your opinions on hardeningDisable = [ "all" ]; in kexectools?
<gchristensen>
fpletz, too, maybe
<LnL>
probably not much you can do about
<sphalerite>
is there a safe way to specify a user's password declaratively?
<clever>
sphalerite: initialPasswordHash
<clever>
the hash will be world readable, but it should be hard to brute-force, so you basicaly only loose /etc/shadow
<vandenoever>
MichaelRaskin: the src version of nightly does work
<kmicu>
There is users.extraUsers.foo.hashedPassword if you have a common threat model.
<gchristensen>
clever: this is too much.
<gchristensen>
clever: is there a more minimal option?
lverns has joined #nixos
<clever>
gchristensen: if the host already has kexec in $PATH, you can use that
<clever>
gchristensen: now my script uses the fixed kexec, but systemd used the broken kexec (and wont mass-rebuild)
<sphalerite>
is it possible to have multiple X sessions open at once with the default display manager?
<clever>
sphalerite: not currently
<gchristensen>
clever: there we are, this way we can focus my time on waiting for the xfer than the build :P
alx741_ has joined #nixos
alx741_ has quit [(Client Quit)]
alx741 has quit [(Ping timeout: 258 seconds)]
k11` has joined #nixos
<k11`>
It seems my system completely broken
<clever>
kmicu: what error happens on bootup?
<k11`>
I can boot root before upgrade but nixos-rebuild hangs it down
<sphalerite>
clever: ok. It would be possible with gdm though right?
<clever>
k11`: did you try "nixos-rebuild boot" ?
<clever>
sphalerite: the current nixos modules set everything to :0, so it will likely get upset when things start to run on :1
<sphalerite>
aw ok. Thanks for the info
<k11`>
clever: when I doing something with nixos-rebuild it breaks the system
<clever>
k11`: "something" ?
<k11`>
clever: test or switch
<gchristensen>
clever: alpine linux is just not the thing, I guess: "kexec_load failed: Function not implemented"
<clever>
k11`: there are some changes that cant be done online, and test/switch do them, boot is sometimes safer
<clever>
gchristensen: does the hosting provider offer a rescue shell, or other distros?
<fpletz>
gchristensen: clever: regarding disabling hardening for kexectools, should be safe and shouldn't break any dependent packages… should we just push to staging?
alx741 has joined #nixos
<gchristensen>
fpletz: SGTM
xwvvvvwx has quit [(Quit: WeeChat 1.6)]
<clever>
fpletz: sounds good here too
<sphalerite>
My config for i3 currently still lives in my home directory, but I want to move it to the system config. I can do the i3wm part of that easily using the services.xserver.windowManager.i3.configFile setting. However, that config file contains the reference to the status bar config file, for which there is no relevant option. Is there a neat way to get that handled by nix as well?
schneefux has quit [(Quit: ZNC 1.6.3 - http://znc.in)]
<sphalerite>
I could modify the package so that what I want is the default config, but that seems like overkill
<k11`>
clever: I tried nixos-rebuild boot, but on rebooting I have same problem with broken paths
<clever>
k11`: what paths is it saying are broken?, can you pastebin all of the console output?
schneefux has joined #nixos
<k11`>
clever: systemd says: Failed to open directory /nix/store/<HASH>-etc/shells: Not a directory
<k11`>
clever: I can only take screenshot
Itkovian has joined #nixos
yonk has quit [(Ping timeout: 258 seconds)]
<clever>
k11`: upload to imgur and i can take a look
<LnL>
clever: changing that file is a mass rebuild right?
<clever>
LnL: yeah, everything uses setup.sh
<clever>
including the compiler
filterfish has joined #nixos
filterfish has quit [(Remote host closed the connection)]
osa1 has joined #nixos
osa1 has quit [(Changing host)]
osa1 has joined #nixos
<clever>
k11`: it looks like that generation of nixos is tottally corrupt, what was the outcome of nix-store --verify --check-contents --repair ?
Fare has quit [(Ping timeout: 258 seconds)]
<k11`>
clever: I can't understand how it had happened. Are there the right way to fix this? Or I need reinstall nixos.
<clever>
k11`: it should be possible to recover without a reinstall, can you pastebin the output of `nix-store --verify --check-contents`
<k11`>
clever: I already tried this too.
<k11`>
clever: one minute
mudri` has quit [(Quit: WeeChat 1.6)]
Fare has joined #nixos
hotfuzz has quit [(Read error: No route to host)]
hotfuzz has joined #nixos
osa1 has quit [(Ping timeout: 260 seconds)]
<LnL>
clever: I tried that before, bash gets stuck when I use substituteInPlace apparently
<LnL>
nix-build works fine tho
<Mic92>
can somebody download something from cache.nixos.org?
<Mic92>
ok it is just me
alx741_ has joined #nixos
alx741 has quit [(Ping timeout: 252 seconds)]
hiratara has quit [(Ping timeout: 240 seconds)]
filterfish has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
TimePath[m] is now known as TimePath
hiratara has joined #nixos
<yonk>
can someone give me a hint how I would go about installing broadcom-sta by hand (not via extraModulePackages in the configuration.nix)? I need to modify it because the hash in the current channel is broken and I don't want to build the whole system on my laptop
<k11`>
clever: The result is same.... no corruptions found.
Fare has quit [(Ping timeout: 265 seconds)]
<clever>
k11`: its possible that the system was malfunctioning during the nixos-rebuild, so that generation is toasted and the hashes say its fine
<clever>
k11`: what is the output of "realpath /nix/var/nix/profiles/system-181-link/etc" ?
offlinehacker[m] has joined #nixos
<offlinehacker[m]>
in which output should one place bash/zsh/fish completion files? bin/lib/doc/man or should i introduce new output, and how should it be named?
<clever>
k11`: what generation does the /nix/var/nix/profiles/system symlink point to, and are you still going thru /mnt, a chroot, or booted into another generation?
<k11`>
clever: I boot livecd, mount to /mnt and chroot to it
<clever>
k11`: and which generation is system pointed to?
<clever>
yeah, thats a bit strange, but should be safe to ignore
<clever>
is 179 a generation that can boot?
<k11`>
clever: No I need to rolling back yet
<clever>
k11`: oh, how much free disk space is there?
Fare has joined #nixos
<k11`>
clever: ~3.5G
<clever>
k11`: might cut it a bit close, but i have an idea on how to rescue it with less data loss, what livecd are you using?
<k11`>
clever: It's nixos 16.09 install cd.
<clever>
k11`: exit out of the chroot, and rename /mnt/nix to /mnt/old-nix, then just run nixos-install and it will recreate all of /nix from the configuration.nix you already have
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
filterfish has quit [(Remote host closed the connection)]
<k11`>
clever: Ok, but it seems I have too less free space
<sphalerite>
oh wow, the steam package pulls in almost a whole ubuntu system x)
<clever>
k11`: was there anything of value in things like "nix-env -q"?
<k11`>
clever: Do I need to call nixos-generate-config?
<clever>
you can if you think the hardware config needs to be updated
Itkovian has quit [(Client Quit)]
mudri has joined #nixos
Itkovian has joined #nixos
<Sirolu>
Hello.
Itkovian has quit [(Client Quit)]
<Sirolu>
I was wondering if it is possible to do something specific, and if so, how: is it possible to create a Nix derivation which has a local file or folder as its source code?
<clever>
Sirolu: src = ./.;
<Sirolu>
And how do I fetch that information?
<clever>
every time you eval the nix expression, it will re-import that directory, and if it has changed, it will rebuild things
<Sirolu>
Ah, that makes sense. Though I am still not sure how to access it in the expression.