<MichaelRaskin>
Illegal instruction? You have an old CPU that lacks some of the instructions used in the build in the binary cache?
<MichaelRaskin>
Not sure, just looks like that
<xcmw>
MichaelRaskin: I have a MacBook Pro (Retina, 15-inch, Mid 2015)
<xcmw>
So I think that is unlikely
<lambdamu>
xcmw: i doubt this is nix specific
<spacekitteh>
xcmw: is it the same error location every time?
<greymalkin>
Bork! neither services.mysql with package=mysql57 or package=mariadb succeed to start.
<spacekitteh>
it could be a ram failure
<xcmw>
Same thing everytime
digitus has quit [(Quit: digitus)]
nathan7 is now known as edef
seanz has quit [(Quit: Leaving.)]
<lambdamu>
xcmw: can you enter nix-shell?
<xcmw>
lambdamu: No
<lambdamu>
same error?
<xcmw>
Yes
<xcmw>
builder for ‘/nix/store/isbryz447ai922ziw37x49q6pdpwz53k-cabal2nixResult.drv’ failed due to signal 4 (Illegal instruction: 4)
<clever>
xcmw: does osx have anything like dmesg, which may say which process exactly is failing?
<xcmw>
clever: No idea what dmesg is
takle has quit [(Remote host closed the connection)]
<Acou_Bass>
eey guys, i asked about this earlier but i had to shoot, the joys of having a job -_- im trying to upgrade the linux kernel in nixos using boot.kernelPackages = pkgs.linuxPackages_latest - for whatever reason the nvidia driver refuses to build its modules to this kernel... has anyone gotten this to work?
<greymalkin>
Okay, nm. It was my "initialScript" that failed to run... though I don't know why that would be.
<clever>
xcmw: is dmesg a valid command, what happens when you try to run it?
<xcmw>
clever: A bunch of random crap is printed
<clever>
xcmw: is there anything near the end relating to illegal instructions?
<xcmw>
clever: No
calvertvl has joined #nixos
<pikajude>
clever, xcmw: it would be in console.app
<xcmw>
Nothing there
markus1189 has joined #nixos
<gchristensen>
Acou_Bass: currently, in stable, the latest nvidia driver is not compatible with the latest linux kernel
markus1199 has joined #nixos
<Acou_Bass>
ahhh ok
calvertvl has quit [(Quit: Page closed)]
<gchristensen>
and IIRC the nvidia driver in unstable isn't really stable yet? not sure
<Acou_Bass>
ill cope with it for now :P my wifi card doesnt like 4.4 for some reason (works fine on 4.9 if i disable nvidia) but ive got a spare i can use for now hehe
calvertvl has joined #nixos
markus1219 has quit [(Ping timeout: 260 seconds)]
markus1209 has quit [(Ping timeout: 260 seconds)]
systemfault has joined #nixos
<LnL>
xcmw: it seems they are using a custom version of nixpkgs, perhaps they don't have some of the stuff we introduced to support sierra
<peterhoeg>
Acou_Bass: you could also try nouveau instead although it may not work for your use-case(s)
<Acou_Bass>
nah i play gamez :P
<Acou_Bass>
like i said its no problem for now, i can use my spare wifi card for now
<Acou_Bass>
chances are at the very most all ill have to do is wait for the next nixos release, or possibly just keep my eye on the nvidia nixpkg to see if it does start building correctly against latest linux
takle has joined #nixos
<gchristensen>
Acou_Bass: next nixos release is in about 1 month :)
<Acou_Bass>
does nixos have a similar release cycle to ubuntu then, ie. every ~6 months?
xcmw has quit [(Quit: Lost terminal)]
<gchristensen>
yeah
seanz has joined #nixos
<Acou_Bass>
ive only actually been using nixos since 16.03 so havent really paid attention to things like this
<gchristensen>
March(.03) and September(.09)
<Acou_Bass>
hehe so a month 'behind' ubuntu :D though similar 'major' numbers i suppose
takle has quit [(Ping timeout: 260 seconds)]
<gchristensen>
"We've discovered (and purged) cached pages that contain private messages from well-known services, PII from major sites that use cloudflare, and even plaintext API requests from a popular password manager that were sent over https (!!).
<Acou_Bass>
a few of my slightly more 'tinfoil hat' oriented friends have described cloudflare as an NSA honeypot to catch tor users... im starting to think they are wrong, its just a mess from top to bottom
AllanEspinosa has quit [(Ping timeout: 260 seconds)]
takle has quit [(Ping timeout: 260 seconds)]
erasmas has quit [(Quit: leaving)]
seanz has quit [(Quit: Leaving.)]
jsgrant- has joined #nixos
mkoenig has quit [(Ping timeout: 240 seconds)]
AllanEspinosa has joined #nixos
mudri has quit [(Quit: WeeChat 1.7)]
systemfault has quit [(Quit: Bye!)]
ibrahims has quit [(Ping timeout: 255 seconds)]
AllanEspinosa has quit [(Ping timeout: 240 seconds)]
<greymalkin>
Is there any nix-level support for composer packages?
<greymalkin>
(php)
dbmikus has quit [(Ping timeout: 268 seconds)]
<gchristensen>
I don't know of any, greymalkin, but I would like that to be changed
mkoenig has joined #nixos
<greymalkin>
Acou_Bass: Have you seen US government websites? They're all a mess. I wouldn't use that as evidence that it's not government run...
<Acou_Bass>
XD true
AllanEspinosa has joined #nixos
<gchristensen>
globin: if I rewrote the configuration doc, would I be at risk of it not merging? :P
<gchristensen>
configuration.xml
Rotaerk has joined #nixos
ibrahims has joined #nixos
<clever>
greymalkin: the main website for a datacenter i use, has an ssl cert that expires 130 days ago
<clever>
greymalkin: and judging by what i saw in tcpdump when debuging network issues, i suspect i can mitm any box in the datacenter, potentialy including the main website
<clever>
and if it already has ssl warnings, is anybody going to notice a self-signed cert?......
mkoenig has quit [(Ping timeout: 255 seconds)]
mkoenig has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
AllanEspinosa has quit [(Ping timeout: 260 seconds)]
mkoenig has quit [(Ping timeout: 240 seconds)]
<lambdamu>
The lack of a navigation/toc sidebar makes the nix/nixpkgs docs unwieldy to read
<lambdamu>
or is it just me?
georges-duperon has quit [(Ping timeout: 255 seconds)]
mkoenig has joined #nixos
hopio3151 has quit [(Remote host closed the connection)]
<Acou_Bass>
it never really bothered me but i do see what you mean
ibrahims has joined #nixos
<Acou_Bass>
im a bit scatter-brained anyway so it works for me
<gchristensen>
there is a ToC
ibrahims has quit [(Ping timeout: 240 seconds)]
<lambdamu>
at the top?
<lambdamu>
i thought of a sidebar toc
<greymalkin>
gchristensen: I think lambdamu means placing the ToC in a sidebar so that it is visible, even when you're in the depths of NixOS Appendix A (configuration options)
<lambdamu>
yep
<gchristensen>
I see
<gchristensen>
that is probably doable, we use docbook
<greymalkin>
I will have to agree... my browser has about 20 nix-related windows open, because I ctrl-click on the entries so that I don't lose my place... which means I can't find the right tab so I go back to the ToC tab and ctrl-click again.
<greymalkin>
s/windows/tabs/
<lambdamu>
yes i saw a sphinx rendering somewhere which had a sidebar
<gchristensen>
I'm not sure sphinx and docbook are the same?
<gchristensen>
or how they relate to each other
<lambdamu>
i'm not sure either but i was under the impression that sphinx is not coupled with a specfic markup lang
<lambdamu>
hm nah im wrong
<lambdamu>
the mention restructured text on the website
davidak has quit [(Quit: Leaving.)]
ibrahims has joined #nixos
masia[m] has left #nixos ["User left"]
AllanEspinosa has joined #nixos
<greymalkin>
clever: Re SSL warnings, I think I'm about the only person I know who checks *why* a cert is invalid before adding a temporary exception.
<clever>
greymalkin: i also check it, but my browser doesnt remember the exception for that long, and i sometimes skip checking the next time it comes up
<clever>
but i do try avoiding entering anything secret on the site anyways
ibrahims has quit [(Ping timeout: 268 seconds)]
mkoenig has quit [(Ping timeout: 260 seconds)]
mkoenig has joined #nixos
<gchristensen>
clever: how do I build nixpkgs' docs again?
<nh2>
hi, does nixos have a patch for CVE-2017-6074? It seems like the upstream fix is only in Linux 4.10 so far
takle has quit [(Ping timeout: 240 seconds)]
<clever>
2017-02-23 19:45:13 -NixOS_GitHub:#nixos- nixpkgs/master 1d68edb Graham Christensen: linux kernels: patch against DCCP double free (CVE-2017-6074)
<clever>
2017-02-23 19:45:41 < gchristensen> *sigh* I did this wrong.
<clever>
2017-02-23 19:49:09 < gchristensen> back in a bit, clearing my head, then going for try 2
ibrahims has joined #nixos
<clever>
gchristensen: how has try#2 gone?
<clever>
nh2: i believe you can also use boot.blacklistedKernelModules as a short-term fix
<clever>
nh2: if the kernel cant load the module, then the buggy code will never be in ram
<clever>
you simply loose dccp support, though ive never seen anything that used it
<gchristensen>
nh2: ah, I'm running build tests for DCCP right now
<gchristensen>
nh2: patches will be released within the next few hours, sorry for the delay
<nh2>
gchristensen: thank you!
lambdamu_ has joined #nixos
<gchristensen>
you're welcome :) glad to hear there are some concerned users about it!
<gchristensen>
clever: if you can write up an email about it, I would like that. I will forward it to the security announcement email list.
<nh2>
gchristensen: CVEs are what makes me do lots of stuff at 3 am :D
<clever>
i'm not that good with formal things like that
ibrahims has quit [(Ping timeout: 260 seconds)]
<gchristensen>
clever: if you can write instructions, I'll write the formal stuff :)
<gchristensen>
nh2: you're telling me :P
lambdamu has quit [(Ping timeout: 268 seconds)]
<NixOS_GitHub>
[nixpkgs] benley opened pull request #23129: qemu-vm: Make it possible to adjust RAM allocation at runtime (master...qemu-vm) https://git.io/vyfuJ
<gchristensen>
whoa^
<clever>
nh2: gchristensen: hmmm, boot.blacklistedKernelModules did nothing to stop a python script from loading dccp_ipv4 and dccp
<unlmtd[m]>
I have libinput enabled. which on other distros gives me the trackpoint scrolling, but with nixos not
<nh2>
clever: I'm also testing it at the moment
<clever>
nh2: acording to the man page, blacklist makes the kernel ignore the special pci:123 aliases on a module
<clever>
but does nothing to actualy stop loading
<clever>
so it just stops auto-loading, because a driver claimed support at compile-time
<nh2>
clever: I can confirm, it doesn't prevent a modprobe
<clever>
and now that i have made a single dccp socket, the module is in-use, and i cant unload it
digitalmentat has quit [(Ping timeout: 268 seconds)]
<nh2>
However, there is a workaround for this behaviour; the install command instructs modprobe to run a custom command instead of inserting the module in the kernel as normal, so you can force the module to always fail loading with:
<NixOS_GitHub>
[nixpkgs] peterhoeg pushed 1 new commit to master: https://git.io/vyf27
<NixOS_GitHub>
nixpkgs/master 8e3d0b8 Peter Hoeg: awless: 0.0.13 -> 0.0.14
ibrahims has quit [(Ping timeout: 240 seconds)]
batdog has quit [(Ping timeout: 240 seconds)]
mkoenig has joined #nixos
sellout- has joined #nixos
batdog has joined #nixos
<nh2>
clever: works for me, thank you (I rebooted though)
<nh2>
gchristensen clever: I appreciate the quick work on security updates a lot. It's been my main concern in the past whether nixos can handle it swiftly, and it seems that it can. That's great
<nh2>
although with the CloudFlare exploit just now there may be even bigger things to worry about ;)
ibrahims has joined #nixos
<gchristensen>
nh2: you're welcome! I'm glad to hear it. May I ask where you're using NixOS? also, I must invite you to participate in the next week's bug h,nt by commenting here and I'll ping you:https://github.com/NixOS/nixpkgs/issues/23072
<gchristensen>
oohh ...
<gchristensen>
clever: using `builtins.pathExists` I can make a allowUnfreePredicate function which allows unfree software as long as it is already installed
<clever>
gchristensen: neat
<gchristensen>
what an abuse :P
ibrahims has quit [(Ping timeout: 260 seconds)]
roconnor has quit [(Ping timeout: 260 seconds)]
takle has joined #nixos
slack1256 has joined #nixos
ibrahims has joined #nixos
takle has quit [(Ping timeout: 240 seconds)]
takle has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
<NixOS_GitHub>
[nixpkgs] grahamc opened pull request #23130: nixpkgs: allow packages to be marked insecure (master...insecure-packages-with-docs) https://git.io/vyfVB
<gchristensen>
ok I'm way past my bedtime, but I got curl done, kernel done, that email sent, docs done...
<gchristensen>
good night :)
MP2E has joined #nixos
takle has quit [(Ping timeout: 240 seconds)]
<spacekitteh>
nite
sellout- has quit [(Ping timeout: 240 seconds)]
<nh2>
gchristensen: I'm using it on the server for some projects I run. I commented there to be CC'd
eacameron has quit [(Remote host closed the connection)]
<nh2>
gchristensen: what would be nice to have is some tool that can intersect the software I've installed on my nixos with the software that has security vulns in release-16.09, and can tell me accordingly
nh2 has left #nixos []
mguentner has quit [(Quit: WeeChat 1.7)]
mkoenig has quit [(Ping timeout: 240 seconds)]
nh2 has joined #nixos
sellout- has joined #nixos
lambdamu_ has quit [(Ping timeout: 260 seconds)]
lambdamu has joined #nixos
justan0theruser has quit [(Ping timeout: 255 seconds)]
<copumpkin>
maybe it's in the activation script managed by the systemd module...
<copumpkin>
but I see nothing there either
<spacekitteh>
wtf, $700 to preorder the new ryzen 1800x chip
ibrahims has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
<clever>
heading off to bed now
Drakonis_ has joined #nixos
<copumpkin>
alright, thanks for the help :) I'll continue staring
slack1256 has quit [(Ping timeout: 240 seconds)]
athan has joined #nixos
mizu_no_oto has quit [(Quit: ["Textual IRC Client: www.textualapp.com"])]
mkoenig has quit [(Ping timeout: 260 seconds)]
alx741 has quit [(Ping timeout: 260 seconds)]
AllanEspinosa has quit [(Ping timeout: 240 seconds)]
sellout- has quit [(Quit: Leaving.)]
roconnor has joined #nixos
ibrahims has joined #nixos
ibrahims has quit [(Ping timeout: 240 seconds)]
<eacameron>
Is there a way to run nix expressions on the command line and get results printed to stdout? I have some very simple config nix files and I'd like to run something like "nix-eval '(import ./config.nix).name" and have the value of that attr printed to stdout.
ibrahims has joined #nixos
jsgrant has joined #nixos
<lambdamu>
eacameron: there is nix-instantiate
<lambdamu>
eacameron: also i really like nix-repl for these things
<eacameron>
lambdamu: Hmm...I've never used that directly
<eacameron>
lambdamu: How would nix-repl help?
<eacameron>
I get that I can do it in the repl
<eacameron>
But I want to use this in a script
phreedom has quit [(Ping timeout: 240 seconds)]
<lambdamu>
eacameron: ah ok in that case that doesn't help
ibrahims has quit [(Ping timeout: 255 seconds)]
<lambdamu>
nix-instantiate should do, though
edef is now known as nathan7
nathan7 is now known as edef
<lambdamu>
you need the --eval flag if you don't want to build a derivation
<lambdamu>
nix-instantiate --eval --strict -E 'let foo = 1; bar = 2; in { result = foo + bar; }'
<eacameron>
lambdamu: Aha! That looks exactly what I need!
phreedom has joined #nixos
<eacameron>
lambdamu: Yes that works great. Thanks so much!
jsgrant has quit [(Quit: Peace Peeps. o/ If you need me asap, message me at msg@jsgrant.io & I'll try to get back to you within 36 hours.)]
hexagoxel has quit [(Ping timeout: 240 seconds)]
<lambdamu>
eacameron: you're welcome
ibrahims has joined #nixos
hexagoxel has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
layus__ has joined #nixos
layus_ has quit [(Read error: Connection reset by peer)]
systemfault has quit [(Remote host closed the connection)]
eacameron has joined #nixos
<sphalerite>
I still don't really know what the best way to write a project-specific default.nix is. What I currently do is { pkgs ? import <nixpkgs> {} }: let inherit (pkgs) stdenv dep1 dep2...; in stdenv.mkDerivation {...} — but is there a better way?
takle has joined #nixos
<copumpkin>
why not just with pkgs?
<copumpkin>
I usually do something like that, except version-locked
derjohn_mob has quit [(Remote host closed the connection)]
<LnL>
or callPackage, if you want something that looks more like the definitions in nixpkgs
<sphalerite>
copumpkin: so it's clear exactly which deps I'm using — because buildDeps isn't the only place that other packages can be referred to
eacameron has quit [(Ping timeout: 260 seconds)]
<copumpkin>
fair enough
takle has quit [(Ping timeout: 240 seconds)]
<LnL>
something like { pkgs ? ... }: let drv = { stdenv, hello }: stdenv.mkDerivation ...; in pkgs.callPackage drv {}
<sphalerite>
Ah yes, that's nice
<sphalerite>
I think I'll move to that, thanks
<LnL>
and for more complex things I sometimes move the drv part to a separate file
eacameron has joined #nixos
ibrahims has joined #nixos
pi3r has joined #nixos
<sphalerite>
also, why does nix-shell not allow both -p and -E?
eacameron has quit [(Ping timeout: 268 seconds)]
<sphalerite>
or both -p and using shell.nix
<sphalerite>
(use case: I want a mostly-pure shell, with inotify-tools so I can use a script that watches files for changes and rebuilds)
phreedom has quit [(Ping timeout: 260 seconds)]
<sphalerite>
ooh, I could use nix-shell --run in the watch script instead
<sphalerite>
that also means any new deps will get pulled in as necessary
takle has joined #nixos
Drakonis_ has quit [(Quit: Leaving)]
thc202 has joined #nixos
bennofs1 has joined #nixos
goibhniu has joined #nixos
jluttine has joined #nixos
romildo has joined #nixos
<copumpkin>
sphalerite: -p can behave a bit like -E
<copumpkin>
in that you can put an arbitrary expression inside its parameter
<copumpkin>
nix-shell -p "(import ./path/to/expression).foo" for example
phreedom has joined #nixos
ericsagnes has quit [(Ping timeout: 255 seconds)]
ambro718 has joined #nixos
mk-fg has joined #nixos
mk-fg has quit [(Changing host)]
mk-fg has joined #nixos
indi_ has joined #nixos
zraexy has quit [(Quit: Leaving.)]
indi_ has quit [(Ping timeout: 255 seconds)]
indi_ has joined #nixos
wangoe has joined #nixos
bennofs1 has quit [(Ping timeout: 255 seconds)]
DutchWolfie has joined #nixos
indi_ has quit [(Remote host closed the connection)]
<Jnsaff>
good morning, is there an easy way of finding out and visualizing the "tree" of packages that went into building/installing a specific package?
eacameron has quit [(Ping timeout: 260 seconds)]
<Jnsaff>
I see a list of stuff in /nix/store but I don't really know who requires what
<goibhniu>
hi Jnsaff, you can see it on hydra and you can also query nix-store locally in various ways
<goibhniu>
Jnsaff: check out nix-store --query, in the man page
<Jnsaff>
cool, thank you
<goibhniu>
you're welcome
<obadz>
copumpkin: sorry to hear :-D
<obadz>
copumpkin: it makes sense though, don't you think?
<romildo>
sphalerite, I have learned that software development should be done with nix-shell. Just adding compilation tools and libraries to environment.systemPackages is not enough, because some hooks are not run.
mudri has joined #nixos
<sphalerite>
I see, good to know :) it's not the only way, as I indicated on the issue (I'm lheckemann), but yeah it does seem the most sensible way
<romildo>
sphalerite, therefore I solved my issue with cmake not finding ncurses with the nix-shell:
<romildo>
In this case I was using the CLion IDE for development.
<sphalerite>
for greater simplicity and to encode your dependencies in a file, you may also want to add a shell.nix or default.nix, then you can just use nix-shell --run clion (or --command clion)
gilligan_ has joined #nixos
<gilligan_>
hi
griff_ has joined #nixos
<romildo>
sphalerite, thank you for the tips you gave in the issue tracker.
ilyaigpetrov has joined #nixos
<sphalerite>
np
ibrahims has quit [(Ping timeout: 260 seconds)]
m0rphism has quit [(Quit: WeeChat 1.4)]
m0rphism has joined #nixos
derjohn_mob has quit [(Ping timeout: 255 seconds)]
<romildo>
The packages included in the nix-shell depends on the project. Is it possible to define different set of dependencies in shell.nix or default.nix to use with nix-shell for the different projects I work on?
mkoenig has quit [(Ping timeout: 260 seconds)]
<sphalerite>
yes, nix-shell will look for shell.nix and then default.nix in the current working directory
pi3r2 has quit [(Ping timeout: 260 seconds)]
<gchristensen>
globin: ping
<sphalerite>
the contents you'd want would be something like with import <nixpkgs> {}; stdenv.mkDerivation {name = "projectName"; src = lib.cleanSource ./. ; buildInputs = [cmake ncurses ...]; }
ibrahims has joined #nixos
<romildo>
sphalerite, what is the purpose of src = lib.cleanSource ./. in your example?
<sphalerite>
romildo: mkDerivation requires a src attribute. lib.cleanSource filters out git repos and such
ibrahims has quit [(Ping timeout: 260 seconds)]
derjohn_mob has joined #nixos
nadley has quit [(Remote host closed the connection)]
FRidh has joined #nixos
m0rphism has quit [(Quit: WeeChat 1.4)]
m0rphism has joined #nixos
xadi has quit [(Quit: Leaving.)]
m0rphism has quit [(Quit: WeeChat 1.4)]
m0rphism has joined #nixos
ibrahims has joined #nixos
nh2 has joined #nixos
m0rphism has quit [(Client Quit)]
MoreTea has joined #nixos
m0rphism has joined #nixos
<MoreTea>
hiya
<MoreTea>
builtins.derivation and builtins.toFile are obviously having side effects. Are there other builtins that have side effects?
mudri has quit [(Ping timeout: 268 seconds)]
ibrahims has quit [(Ping timeout: 240 seconds)]
derjohn_mob has quit [(Ping timeout: 255 seconds)]
<nh2>
bendlas: I meant at the nixos package level. E.g. when the nixos team identifies nixos packages with security issues, not only write it into the security-announce email, but also provide those package names in a parseable format so I can ask my nixos machine which of those packages it uses; then I can set up a notification (e.g. email) to ping me when I use an affected package
<srhb>
When deploying an example with nixops to libvirtd, I get stuck on: could not connect to ‘root@192.168.122.208’ -- anyone know the solution?
<gchristensen>
MoreTea: fetchTarball maybe?
m0rphism has quit [(Quit: WeeChat 1.4)]
<gchristensen>
MoreTea: builtins.fetchurl
m0rphism has joined #nixos
<srhb>
Looks like there's an unmerged fix for that...
<srhb>
It's so much easier to do actual deployments with NixOps on gce etc., the virtual backends are obnoxiously broken at time. :-P
<gchristensen>
I agree :(
<gchristensen>
that is tough, I can't help with that PR
<srhb>
No problem, thanks anyway. :)
<srhb>
I'll just switch to virtualbox instead for now.
<MoreTea>
@srhb, be aware that running vbox is not possible if you're using KVM at the same time (which you probably are if you're trying to work with libvirt)
m0rphism has quit [(Client Quit)]
<srhb>
MoreTea: Thanks! I wasn't aware.
<srhb>
Hmm, tricky.
ericsagnes has quit [(Ping timeout: 268 seconds)]
m0rphism has joined #nixos
m0rphism has quit [(Client Quit)]
m0rphism has joined #nixos
ibrahims has joined #nixos
<srhb>
Protip: Remember to disable the gui when you deploy 10+ vms
<gchristensen>
lol
ibrahims has quit [(Ping timeout: 260 seconds)]
<LnL>
srhb: yeah, and don't run your vbox commands to quickly
obadz has quit [(Ping timeout: 255 seconds)]
<gchristensen>
LnL: can you review some docs I wrote?
obadz has joined #nixos
<LnL>
sure, but probably in a couple of hours after work
<gchristensen>
cool :)
<srhb>
Isn't chromium supposed to install its chromium.desktop somewhere when installed system wide?
<srhb>
Or well, symlink it into a profile
m0rphism has quit [(Quit: WeeChat 1.4)]
m0rphism has joined #nixos
<LnL>
srhb: where you talking about compiling without gui support or just not running the ui at the same time?
<srhb>
LnL: deploying. I'm all over the place now though. :-)
<srhb>
It's fine, just needed to set headless.
<LnL>
oh you mean the vms themselves?
ibrahims has joined #nixos
<srhb>
Yes
<NixOS_GitHub>
[nixpkgs] edolstra pushed 1 new commit to master: https://git.io/vyJm7
<greymalkin>
Why is nixops suddenly getting authentication errors on a new virtualbox instance? I've destroyed and deployed three times now and can't make it work again, even though the network and "hardware" configurations haven't changed.
<greymalkin>
Actually, they have ... I changed a shared folder from readOnly to rw... I'm testing to see if that caused it.
<greymalkin>
Nope.
<greymalkin>
Still getting authentication errors out of the blue.
<Baughn>
Anyone got an up-to-date, not overly clever XMonad config I could borrow? I'm a bit unsure how to make it all work with NixOS today.
mkoenig has joined #nixos
phreedom has quit [(Quit: No Ping reply in 180 seconds.)]
phreedom has joined #nixos
ixxie has quit [(Ping timeout: 240 seconds)]
globin_[m] has joined #nixos
c0bw3b has joined #nixos
mizu_no_oto has joined #nixos
romildo has quit [(Quit: Leaving)]
ibrahims has quit [(Ping timeout: 260 seconds)]
fadenb has quit [(Quit: WeeChat 1.7)]
<MoreTea>
greymalkin, did you delete your deployment? that might help
mizu_no_oto has quit [(Quit: Computer has gone to sleep.)]
sellout- has joined #nixos
<NixOS_GitHub>
[nixpkgs] mogria opened pull request #23149: findutils: updatedb now uses writable database outside of nix-store by default (master...updatedb-standalone) https://git.io/vyJ8O
<viric>
What is worst? No use of gnupg signing at all, or constant use with a key not-super-safe? gpg-agent running all the time, in a server used through ssh.
<viric>
I tend to switch between paranoid key safety and no usage at all
<viric>
as if someone expected from me to be really careful about my key
dramforever has joined #nixos
<viric>
or I'm kicked out of the tinfoil-hat club
mudri has quit [(Ping timeout: 260 seconds)]
<viric>
I once created an always-offline gpg master key, to have short-lived (few months) signing subkeys.
georges-duperon has quit [(Ping timeout: 255 seconds)]
<viric>
at first renewal of subkey I noticed I forgot the password of the master key
<viric>
And now... vlock -an on sleep. Can be done?
shlevy has joined #nixos
<viric>
Mic92: that should be in nixos as a boolean
<Mic92>
viric: the same repo contains an example for slock
<Mic92>
viric: you can make that happen, I guess
<viric>
I will spend some time on it later
<viric>
that slock looks like single-session lock
ibrahims has joined #nixos
<viric>
better than nothing, and maybe that can call vlock -an
<viric>
what is that of calling sudo from services? I hadn't seen that before
<Mic92>
viric: it should be not done like that, systemd services support a User parameter.
<goibhniu>
how do you start gpg agent? (it used to work for me when it was an option in configuration.nix)
<Mic92>
Is there an ubuntu package for multiuser nix?
ibrahims has quit [(Ping timeout: 240 seconds)]
<Mic92>
goibhniu: the release changelogs states, that this should be done in your shell configuration (bashrc etc)
<Mic92>
gpg-agent now uses a static socket path
IITaudio has joined #nixos
<IITaudio>
Hi, what package on nixos provides the linux header module.h?
<Mic92>
IITaudio: linuxHeaders maybe?
<IITaudio>
Mic92: i installed it but find /nix/store -name "module.h" does not return anything
<goibhniu>
Mic92: thanks, I tried to follow that, but failed
ibrahims has joined #nixos
<IITaudio>
Mic92: excpet some android things or other non-kernel module.h files
xadi has joined #nixos
<contrapumpkin>
any switch-to-configuration.pl gurus around? I'm trying to understand which code path starts completely new systemd services
<arrais>
wow, @MoreTea! ruby is exactly what I'm interested in
<Mic92>
IITaudio: that one maybe linuxPackages.kernel.dev ?
<Mic92>
this what out-of-tree modules uses
<Mic92>
*this is
<arrais>
thanks, LnL! (sorry for the delay)
<IITaudio>
Mic92: yep! How have you found it?
pi3r has joined #nixos
<Mic92>
IITaudio: I followed the kernel parameter which was passed to the wireguard package
ibrahims has quit [(Ping timeout: 240 seconds)]
Sonarpulse has joined #nixos
<Mic92>
IITaudio: linuxPackages points to the default linux version in nixos, so you have to select a different package if you select a different kernel.
<IITaudio>
Mic92: yes I'm using the default, thanks!
cpennington has quit [(Ping timeout: 240 seconds)]
<contrapumpkin>
is the binary-caches nix configuration option a prioritized list? as in it'll try the first one, then the second one, then the third one, then build from source?
<contrapumpkin>
or does it try them in parallel
ibrahims has joined #nixos
<ikwildrpepper>
contrapumpkin: I always assumed it was
MercurialAlchemi has quit [(Ping timeout: 260 seconds)]
<contrapumpkin>
yeah it looks like the nix source code traverses them in order
<gchristensen>
might like having extra-binary-caches prioritized first, actually
<gchristensen>
since those are likely more specific to the operation
Dezgeg has quit [(Ping timeout: 268 seconds)]
ibrahims has quit [(Ping timeout: 255 seconds)]
georges-duperon has joined #nixos
ibrahims has joined #nixos
danharaj has joined #nixos
xadi has quit [(Ping timeout: 268 seconds)]
cpennington has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
erasmas has joined #nixos
mkoenig has joined #nixos
ibrahims has joined #nixos
danharaj has quit [(Ping timeout: 260 seconds)]
xadi has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
dmi3y has quit [(Quit: dmi3y)]
xadi has quit [(Ping timeout: 240 seconds)]
<NixOS_GitHub>
[nixpkgs] mogria closed pull request #23149: [WIP]findutils: updatedb now uses writable database outside of nix-store by default (staging...updatedb-standalone) https://git.io/vyJ8O
mizu_no_oto has quit [(Quit: Computer has gone to sleep.)]
taeradan has quit [(Remote host closed the connection)]
ibrahims has joined #nixos
dramforever has quit [(Quit: Leaving)]
dmi3y has joined #nixos
xadi has joined #nixos
DutchWolfie has quit [(Quit: Konversation terminated!)]
<contrapumpkin>
niksnut: if you're around, can you point me at the snippet of code that starts up brand new units in switch-to-configuration.pl? I'm kind of stumped at how it happens in a normal system, mostly because I think I understand the flow of the code and still don't see where it would happen :)
ibrahims has quit [(Ping timeout: 240 seconds)]
<avn>
Folks, does this `du: cannot read directory '/nix/store/x8jp6f32nj2l7hv96gdlaw75lviwjpp1-firefox-unwrapped-51.0.drv.chroot': Permission denied` a bug?
<contrapumpkin>
gchristensen: that's inside a loop that runs over all units that existed prior to the new configuration being activated, so a brand new unit wouldn't get covered by that (my understanding and confusion outlined here: https://github.com/NixOS/nixpkgs/issues/23121#issuecomment-282224026)
<contrapumpkin>
(see the `activePrev` that the loop is running over)
<gchristensen>
ahh
<NixOS_GitHub>
[nixpkgs] Profpatsch opened pull request #23151: git-dit: init at 0.1.0 (master...git-dit-0.1) https://git.io/vyJoR
<niksnut>
contrapumpkin: iirc, we just do "systemctl start multi-user.target", which will start new units pulled in by multi-user.target
<niksnut>
"This should start most changed units we stop here as well as any new dependencies"
<contrapumpkin>
the line I linked to is only if multi-user.target changes
<niksnut>
right
<contrapumpkin>
I see
<contrapumpkin>
okay, that gives me something to test, thank you
<contrapumpkin>
really puzzling bug (not sure if you caught the issue with my new EC2 userdata thing, but it completely broke EC2 service start)
xadi has quit [(Quit: Leaving.)]
<contrapumpkin>
but I'll track it down and fix before 17.03 branch-off :P
jmiven has quit [(Quit: co'o)]
<gchristensen>
domenkozar: I caved and ordered the xps 15 ... it should arrived today
ibrahims has quit [(Ping timeout: 260 seconds)]
jmiven has joined #nixos
fadenb has joined #nixos
fadenb has quit [(Client Quit)]
dmi3y has quit [(Quit: dmi3y)]
ibrahims has joined #nixos
Ivanych has quit [(Ping timeout: 240 seconds)]
pi3r has quit [(Ping timeout: 260 seconds)]
fadenb has joined #nixos
fadenb has quit [(Client Quit)]
fadenb has joined #nixos
fadenb has quit [(Client Quit)]
reinzelmann has quit [(Quit: Leaving)]
fadenb has joined #nixos
estewei has joined #nixos
<NixOS_GitHub>
[nixpkgs] mogria opened pull request #23152: findutils: updatedb now uses writable database outside of /nix/store by default (staging...updatedb-standalone) https://git.io/vyJiv
pi3r has joined #nixos
wangoe has quit [(Ping timeout: 255 seconds)]
<NixOS_GitHub>
[nixpkgs] peterhoeg pushed 1 new commit to master: https://git.io/vyJXZ
<NixOS_GitHub>
nixpkgs/master eae1866 Peter Hoeg: sensu: 0.17.1 -> 0.28.0
<lambdamu>
so anyone running the ck-patched kernel? SCHED_AUTOGROUP and RT_GROUP_SCHED need to be made optional in common-config.nix in order to make it work, would there be interest to make that work out of the box?
<viric>
gchristensen: What's the purpose, if everything in a repository is Signed-off-by?
<contrapumpkin>
niksnut: omg very excited, will look more carefully in a few
<gchristensen>
niksnut: you have some of the _best_ commit messages (and some of the worst, see: "Grr" ;) )
phreedom has joined #nixos
<contrapumpkin>
I loved the Grr one
<contrapumpkin>
I even complimented him on it
<contrapumpkin>
niksnut: very cool. I'll need to think through how this building block applies to #520 but it definitely seems helpful
tg has joined #nixos
<pikajude>
emoji-only commit messages from now on
<contrapumpkin>
niksnut: main thing I'm curious about is how you plan on using this for builtins.fetchurl and builtins.fetchTarball. It seems very close to being feasible to do what I want, but those two would violate "pure derivations can't depend on impure ones" rule today
<contrapumpkin>
I think the main difference between what you do and what I want is just that you say "pure can't depend on impure" and I say "pure can depend on impure, but do so very loudly in a way you can keep an eye on"
<contrapumpkin>
niksnut: re: the caching mechanism, that's where I was suggesting a UUID (which is a UX nightmare but I don't have a better idea)
<contrapumpkin>
I'll write questions/points on the commit :)
<contrapumpkin>
easier to track probably
sellout- has quit [(Quit: Leaving.)]
tg has quit [(Quit: Leaving)]
adfaure has joined #nixos
<adfaure>
Hello, I decided to use nixos with plasma 5. It works well, but can someone help me to deactivate kwallet please ? It pops up each time I need a password...
<sphalerite>
contrapumpkin: fixed-output derivations are an exception according ot the commit message
<contrapumpkin>
sphalerite: that's the other direction
<contrapumpkin>
that says FO can depend on impure
<contrapumpkin>
I'm wondering whether impure can depend on everything
digitalmentat has joined #nixos
<sphalerite>
oh whoops, I misread
<sphalerite>
why wouldn't it?
<contrapumpkin>
I hope it can!
<contrapumpkin>
that'd be amazeballs
<contrapumpkin>
but it complicates evaluation/build I significantly I think
FRidh has joined #nixos
<contrapumpkin>
since normally we can spit out all .drv files at once and then build them
<contrapumpkin>
whereas this would force some to be deferred
<contrapumpkin>
I think?
<contrapumpkin>
oh never mind
<FRidh>
adfaure: have you checked the settings with kwalletmanager5 ?
<contrapumpkin>
due to the other restriction we can still do that
tg has joined #nixos
<adfaure>
Hi FRidh, no I dont have this command, it seems not installed
<adfaure>
FRidh: okay, I guess i'll give up plasma
<niksnut>
contrapumpkin: well, there already is a UUID, namely the "virtual" output path
<contrapumpkin>
niksnut: the issue is for it to be a persistent handle for humans to ask for it to be blown away/expired. Say we replace channels with fetchTarball and I don't always want to update at tarball-ttl frequency, then I need some sort of consistent handle to be able to identify the tarball-fetching instance to ask Nix to expire it and grab me another one. And of course "lockfiles" would get identified by the same thing...
<contrapumpkin>
or am I missing something?
<jophish_>
copumpkin: I think the ticket you're referring to is #666
* contrapumpkin
shrieks
<contrapumpkin>
the devil!
<contrapumpkin>
(but thanks :))
<jophish_>
no problem :) I'd quite like to get #666 closed for hercules
<jophish_>
although the NIX_BUILD_HOOK hackery seems to be working for now
dingen has joined #nixos
adfaure has quit [(Quit: Page closed)]
dingen is now known as dreamer
dreamer has quit [(Changing host)]
dreamer has joined #nixos
mudri has joined #nixos
dreamer has left #nixos []
Ivanych has quit [(Ping timeout: 240 seconds)]
Ivanych has joined #nixos
<NixOS_GitHub>
[nixpkgs] vbgl pushed 3 new commits to master: https://git.io/vyU3j
<NixOS_GitHub>
nixpkgs/master 71739f3 Vincent Laporte: Revert "ocamlPackages.spacetime: only build with ocaml-4.04"...
<NixOS_GitHub>
nixpkgs/master 9493ced Vincent Laporte: ocamlPackages.{ojquery,tuntap,uri}: adds ocamlbuild as dependency
<NixOS_GitHub>
nixpkgs/master 690968d Vincent Laporte: ocamlPackages.utop: 1.19.2 -> 1.19.3
<NixOS_GitHub>
[nixpkgs] benley closed pull request #23129: qemu-vm: Make it possible to adjust RAM allocation at runtime (master...qemu-vm) https://git.io/vyfuJ
bennofs1 has joined #nixos
fre has quit [(Quit: WeeChat 1.7)]
adfaure has joined #nixos
xwvvvvwx_ has joined #nixos
<adfaure>
Hello I have a lot of trouble desactivating kwallet for plasma 5 with nixos
<adfaure>
FRidh: I was able to launch kwalletmanager, but i cant save the configuration from my user and I cant launch kwalletmanager as root
ibrahims has quit [(Ping timeout: 255 seconds)]
IITaudio has quit [(Ping timeout: 260 seconds)]
takle has quit [(Remote host closed the connection)]
<contrapumpkin>
niksnut: I think something wonky is happening with startup filesystem resizing, probably unrelated to my ec2 startup commit
<contrapumpkin>
I reverted my commit there and my instance starts up with a tiny filesystem. journalctl -k shows that resize2fs wants me to e2fsck the filesystem and refuses to touch it, despite tune2fs setting check interval/count to never
chakerbenhamed has joined #nixos
ibrahims has joined #nixos
<chakerbenhamed>
thanks lambdamu I will check it out ;)
<lambdamu>
adfaure: did you try "security.pam.services.kdm.enableKwallet = false;"
pi3r has joined #nixos
<lambdamu>
adfaure: I do not use kde, so this is a guess, also assuming you have kdm as display manager
<adfaure>
Hi, will it works if I put that in my configuration.nix ?
<lambdamu>
yes that is were you can change the setting
<adfaure>
(lambdamu: yeah I am totally lost)
<lambdamu>
i do not know if it has the intended effect
<FRidh>
adfaure: how come you cannot save?
<FRidh>
what happens
<adfaure>
FRidh: from the settings panel of kwaletmanager5 it says `KDE Wallet Configuration You are not allowed to save the configuration`
<niksnut>
contrapumpkin: maybe the filesystem was not shutdown cleanly during AMI generation?
chakerbenhamed has quit [(Quit: WeeChat 0.4.2)]
sellout- has joined #nixos
danharaj has joined #nixos
takle has joined #nixos
johnw has quit [(Ping timeout: 276 seconds)]
takle has quit [(Ping timeout: 240 seconds)]
<FRidh>
adfaure: That's mean, and odd.
xwvvvvwx_ has quit [(Quit: WeeChat 1.7)]
<FRidh>
Could you open an issue?
<contrapumpkin>
niksnut: tune2fs on the image right before I take my EBS snapshot says this: Filesystem state: clean
<contrapumpkin>
is that the field I'm looking for?
<FRidh>
Did you upgrade from kde4? If so, could you remove your whole configurattion (yes, not nice, but it tends to make a big difference with kde)
<contrapumpkin>
niksnut: that's the state right before the snapshot (it's unmounted, that's the result of reading the raw device), and when it boots up, it complains that I need to e2fsck
<FRidh>
adfaure: great, thanks
<NixOS_GitHub>
[nixpkgs] benley pushed 2 new commits to master: https://git.io/vyU8E
<contrapumpkin>
it just says don't impose a time limit on it
<contrapumpkin>
and dont impose a mount limit
<contrapumpkin>
we could use tune2fs to lie, but I'm more puzzled why this would have changed recently
<contrapumpkin>
the only thing I can think of is that we used to mount the root fs earlier, and that triggered the "online" codepath rather than this unmounted codepath
<contrapumpkin>
even though frankly I'd expect the online codepath to be even more paranoid
<contrapumpkin>
(but maybe it expects that whoever mounted it already checked it?)
<clever>
only change i see is one i helped with back in dec, it was fudging the number before it umounted
<contrapumpkin>
ah yes
<contrapumpkin>
is that enough though?
<contrapumpkin>
given the check, it doesn't seem like the mount count is the issue
<jophish_>
dmj`: hi
<jophish_>
I should change my irc nick probably
<dmj`>
jophish_: hello!
<jophish_>
thanks FRidh :)
<jophish_>
dmj`: What's up? :)
<contrapumpkin>
I suppose I could add the fsck back
<contrapumpkin>
that's probably the difference...
<contrapumpkin>
not sure why it would work in the VM test and not in practice
<dmj`>
jophish_: Not much :) I’ve noticed the great changes you’ve made to hydra in your branch (per that reddit post). I’m trying to incorporate them onto a nixos machine, but get errors about a missing JSONObject class in hydra-eval-jobs.cc. It seems like a newer nix dep. is needed (since I believe that’s where it’s installed).
<jophish_>
oh cool. If by "great changes" you mean "quirky hacks" I understand :D
<dmj`>
all the same to the end user :)
<dmj`>
cloning private repos is a nice touch, iirc that was added (and submodule fetching)
xwvvvvwx has joined #nixos
<jophish_>
dmj`: What I'm doing is running a branch called fixes-and-features on a nixops managed machine. When I set this up Hydra was very difficult to compile (not sure if that's changed), so it probably only compiles with a particular (old) version of nixpkgs
hotfuzz_ is now known as hotfuzz
<jophish_>
I believe that version is b341de8
johnw has joined #nixos
<jophish_>
my philosophy is "try not to break anything" and wait for hercules to land"
<jophish_>
on the other hand, this setup has been ticking over as an internal hydra server for the better part of a year without any major peoblems
xwvvvvwx_ has joined #nixos
xwvvvvwx_ has quit [(Client Quit)]
<dmj`>
jophish_: ah, I see. I will try that commit then. Have been keeping an eye on hercules as well. Very exciting work being done there.
Dezgeg has joined #nixos
<jophish_>
It's a shame that they didn't get merged when they still applied cleanly, in fact one got closed today, I'm sure the other PRs will follow :(
xwvvvvwx_ has joined #nixos
xwvvvvwx has quit [(Ping timeout: 240 seconds)]
<jophish_>
dmj`: I think that's from June last year, and hydra was of a similar vintage when I forked fixes-and-features
<jophish_>
I'm sorry that there's not a nicer answer to your needs
indi_ has joined #nixos
xwvvvvwx_ has quit [(Client Quit)]
<dmj`>
jophish_: yes, was sad to see all those commits go to the wayside. Not that they couldn’t be salvaged. Allowing hydra to fetch private github repos w/o leaking credentials is def. ideal. Current plan of attack is to make a server that receives webhooks and updates a local mirror on the hydra box itself.
<jophish_>
dmj`: I'd invite you to contribute to Hercules. Although it's not usable at the moment
<jophish_>
so Hydra + hacks is probably best for getting something going quickly
<pi3r>
Does someone know by any chance what is the name of the `ghc` package (not the compiler ?)
indi_ has quit [(Ping timeout: 260 seconds)]
<dmj`>
jophish_: I accept :) being hydra backwards compatible is a smart move in the interim. Hydra + hacks will do for now :) will let you know if I can get fixes-and-features working w/ a more current nixpkgs.
<jophish_>
yeah, that'd be great thanks!
<jophish_>
although I wouldn't try too hard if I were you :)
oida has quit [(Read error: Connection reset by peer)]
hiratara has quit [(Read error: Connection reset by peer)]
<dmj`>
heh :] we’ll see what happens
hiratara has joined #nixos
tokudan has joined #nixos
oida has joined #nixos
oida has quit [(Read error: Connection reset by peer)]
<jophish_>
I've got to go now, dmj`. Keep me posted!
<dmj`>
jophish_: Will do! and thank you.
oida has joined #nixos
phreedom has quit [(Ping timeout: 260 seconds)]
oida has quit [(Read error: Connection reset by peer)]
mizu_no_oto has joined #nixos
<gilligan_>
hey folks - I could use some suggestion here. I'm trying to write a derivation that takes as `src` another derivation and just needs to overwrite the content of one file. At first I thought I could just `echo ${data} > $src/some/file` but my src is in the /nix/store which is the first thing I am confused about right now x__X - I am expecting it to be copied to somewhere
<viric>
This facebook compressor, zstd, looks quite good
adfaure has quit [(Ping timeout: 260 seconds)]
<contrapumpkin>
gilligan_: $sourceRoot
<contrapumpkin>
gilligan_: you might want to look at overrideDerivation though...
Vorpal has quit [(Ping timeout: 240 seconds)]
oida has joined #nixos
<gilligan_>
contrapumpkin, well no.. overrideDerivation does not do what I want/need
<NixOS_GitHub>
[nixpkgs] fpletz pushed 1 new commit to master: https://git.io/vyUuv
<NixOS_GitHub>
nixpkgs/master 7eb7126 Franz Pletz: Revert "lxc: fix build for glibc-2.25"...
<gilligan_>
contrapumpkin, oh
<gilligan_>
contrapumpkin, ok that is not something I thought of. I was actually *hoping* that there would be a better way than my rather dull approach ;)
<gilligan_>
contrapumpkin, thanks!
<contrapumpkin>
my main problem with those prePatch approaches is that they work fine until the inner derivation also defines prePatch
<contrapumpkin>
as in, the patchiness needs to compose more nicely
<contrapumpkin>
but oh well
<contrapumpkin>
it's usually fine :)
<gilligan_>
contrapumpkin, in this case it is a small project in which i have everything under my control so it'll be fine
<gilligan_>
the thing is that i have a scenario in which building derivation A is quite time consuming and I need to later build a derivation A' that just has a minor change in one of the files
Vorpal has joined #nixos
Vorpal has quit [(Changing host)]
Vorpal has joined #nixos
<gilligan_>
so actually now that I think about it using any kind of overrideDerivation does not make sense because then I would have evaluate (build) the whole thing again
mizu_no_oto has quit [(Quit: Computer has gone to sleep.)]
<LnL>
I thought overrideDerivation required the derivation to be evaluated not built, does it?
<gilligan_>
well i'll just give this a try now then
rmrfroot has quit [(Remote host closed the connection)]
takle has quit [(Remote host closed the connection)]
<gilligan_>
LnL, actually let me make this more concrete - I have a derivation `app` which builds a nodejs app (and the build is bloody slow). The result will be used for some testing and later on is the input to creating a docker image - however I need a modified app as input to docker. I need to alter some version file (all a bit complicated to explain ;) - But I doesn't seem to me that overrideDerivation would do the job for me in that
<gilligan_>
case
<contrapumpkin>
gchristensen: yeah I see that in a few places, not sure how it happens
<LnL>
gilligan_: ah, can't you make a new derivation for the app that just copies it and applies some patches
<LnL>
or do you need to go through the build again?
<clever>
contrapumpkin: i think tune2fs -C 0 would be equal to the fsck
<contrapumpkin>
yeah
<contrapumpkin>
little c sets the max
<contrapumpkin>
I might switch to that
<clever>
and capitol C sets the current count
<gilligan_>
LnL, yeah no i defnitely do not want to build anything again ;) i'm creating a simple derivation that just modifies the file during the buildPhase now
<clever>
contrapumpkin: and if C > c, it needs a check
<contrapumpkin>
yeah
<copumpkin>
okay I'll do that in my commit instead of an actual fsck, since the fsck seems kinda unnecessary in practice
<contrapumpkin>
whoa there's two of me
<copumpkin>
contrapumpkin: go away
contrapumpkin has quit [(Quit: Page closed)]
<clever>
lol
johnw has joined #nixos
systemfault has joined #nixos
fiddlerwoaroof has quit [(Ping timeout: 260 seconds)]
pi3r has quit [(Ping timeout: 255 seconds)]
eacameron has quit [(Remote host closed the connection)]
<viric>
gchristensen: I had troubles with the wifi card on tails, in that xps 15.
<gchristensen>
whoa, the "bios" has a "Peak Shift" functionality which stops using A/C power as much as possible during "peak hours"
<viric>
maybe a newer tails fixes that
<gchristensen>
viric: this is also the 9560 which just came out
<viric>
gchristensen: peak hours?
hiratara has joined #nixos
<gchristensen>
like my utility charges more for electricity from 9am to 3pm
<viric>
o_O
<viric>
how much power does it drain?
<gchristensen>
I don't know
<viric>
do you also have different price depending on the hour?
jsgrant- has quit [(Quit: Peace Peeps. o/ If you need me asap, message me at msg@jsgrant.io & I'll try to get back to you within 36 hours.)]
<viric>
some have that here too... but you can only know the price-per-hour of the next day
<viric>
not further
jsgrant- has joined #nixos
<lambdamu>
i'm running on xps 13 with broadcom wifi, has no ethernet connective and nixos install cd requires internet, that was fun for a whole night :D
<gchristensen>
it is just 2 different costs, the peak cost and the night cost
<viric>
ahh
<viric>
bearable
<pi3r>
pkgs.haskellPackages seems to come with a bunch of available packages (mtl,lens,text,vector,...).
<pi3r>
How do I know when one package is not there and need to use ghcWithPackages ?
danharaj has quit [(Ping timeout: 240 seconds)]
sellout- has quit [(Ping timeout: 260 seconds)]
<lambdamu>
pi3r: the haskell.packages.ghc* come with all sensible packages in scope
ttuegel has joined #nixos
phreedom has quit [(Ping timeout: 260 seconds)]
<pi3r>
lambdamu: do you know where I can find the list ? I guess it is somewhere in the source but I can't find it
Sonarpulse has quit [(Ping timeout: 240 seconds)]
gilligan_ has quit [(Ping timeout: 255 seconds)]
<lambdamu>
pi3r: wait i think what i said makes no sense, what are you trying to do?
FRidh has quit [(Remote host closed the connection)]
<lambdamu>
you can load a ghc with nix-shell and ask for the available packages via ghc --show-packages
<lambdamu>
but per default there are only the packages that come with ghc
<lambdamu>
for everything else you have to use ghcWithPackages
<pi3r>
In my list for instnace there is no lens but I am using it and it does not fail ...
takle has joined #nixos
mizu_no_oto has joined #nixos
<pi3r>
but if I remove turtle then it is going to fail
JagaJaga has quit [(Ping timeout: 240 seconds)]
BramD_ has quit [(Ping timeout: 240 seconds)]
romildo has joined #nixos
seanparsons has quit [(Ping timeout: 245 seconds)]
<pi3r>
I can also remove text and vector without problem
<lambdamu>
the linked file makes sense to me
phreedom has joined #nixos
BramD_ has joined #nixos
<pi3r>
And something even more puzzling if I want to use the package 'ghc' (not the compiler) I can put it there in the ghcWithPackages list but it won't find it
<lambdamu>
ghc gets 4 additional packages one is processes adhoc with callCabal2nix which i can only guess what it does and the resulting ghc is passed as buildinput
seanparsons has joined #nixos
romildo has quit [(Client Quit)]
<lambdamu>
the ghc that is in scope in the builder knows about those 4 packages then
romildo has joined #nixos
<pi3r>
I did try that with nix-shell --pure to be sure I don't use a global ghc which would know about these packages
<lambdamu>
so regarding ghc that is a package that is always known to the compiler
<lambdamu>
it is in the boot list i linked earlier
derjohn_mob has joined #nixos
<pi3r>
well whenever I import (Util) from the ghc package it does complain when I try a script with runghc
ambro718 has quit [(Ping timeout: 260 seconds)]
<pi3r>
telling me that ghc-8.0.2 is an hidden package ...
MoreTea has quit [(Ping timeout: 260 seconds)]
<lambdamu>
hm ok in that case it should probably be in the list passed to ghcWithPackages