gchristensen changed the topic of #nixos to: NixCon live stream: https://www.youtube.com/c/NixCon/live https://nixcon2018.org/ || NixOS 18.09 released https://discourse.nixos.org/t/1076 || https://nixos.org || Latest NixOS: https://nixos.org/nixos/download.html || Latest Nix: https://nixos.org/nix/download.html || Logs: https://logs.nix.samueldr.com/nixos/ || #nixos-dev, #nix-darwin, #nixos-aarch64, #nixos-chat, #nixcon
<clever> andi-: probably, i havent tested deploy yet
<clever> andi-: and i need to redo that entire nixos config layer, to fetch what is pre-generated on the machine
<{^_^}> [nixpkgs] @zimbatm merged pull request #49379 → nixos/dockerPreloader: preload docker images → https://git.io/fxQMu
<{^_^}> [nixpkgs] @zimbatm pushed commit from @nlewo to master « nixos/dockerPreloader: preload docker images (#49379) »: https://git.io/fxp9h
<andi-> clever: no worries, just curious as mic linked it to me... Have you thought about spot instances yet? Haven't seen that yet while briefly looking at it
kreisys has quit [Read error: Connection reset by peer]
<clever> andi-: i was using spot instances for all testing, and its currently hard-coded to only do spot, i plan to make it a nix option in the deployment file
<andi-> nice
<andi-> clever++
<{^_^}> clever's karma got increased to 33
<Mic92> clever++
<{^_^}> clever's karma got increased to 34
<Mic92> some more points until 42.
<hyperfekt> oh you can do that?
<hyperfekt> once for today and once for the first time you helped me :)
<hyperfekt> clever++
<{^_^}> clever's karma got increased to 35
<hyperfekt> clever++
<{^_^}> clever's karma got increased to 36
jcrben has quit [Quit: Ping timeout (120 seconds)]
jcrben has joined #nixos
<azazel> clever: You were right... again!
<azazel> thanks
<drakonis> clever++
<{^_^}> clever's karma got increased to 37
<ottidmes> clever++
<{^_^}> clever's karma got increased to 38
<drakonis> clever girl
brodul has quit [Ping timeout: 272 seconds]
<rotaerk> so how does this work? can he spend this karma and do something evil?
<clever> Drakonis: i'm a dude :P
<azazel> :-D
<drakonis> never watched jurassic park?
<azazel> drakonis: which episode?
Laalf has quit [Quit: No Ping reply in 180 seconds.]
jbgi has quit [Ping timeout: 260 seconds]
<drakonis> this scene
<aleph-> Yo Drakonis
<clever> Drakonis: not sure, but i do remember that line from somewhere
<aleph-> clever: It's Jurrasic Park
<drakonis> sup aleph-
<aleph-> Drakonis: Studying for a triple byte interview.
<{^_^}> [nixpkgs] @dtzWill merged pull request #48672 → w3m: 20161120 -> 20180125 → https://git.io/fxVis
<{^_^}> [nixpkgs] @dtzWill pushed 2 commits to staging: https://git.io/fxpHk
<aleph-> Figure I'll look for a non ops position too.
<aleph-> Cursing that mopidy is a pile of excrement
<ottidmes> when I run this, I expect it to fail on the random sha256 I have provided. Or is this normally checked elsewhere? with import <nixpkgs> {}; toString (fetchurl { url = https://raw.githubusercontent.com/dani-garcia/bitwarden_rs/f1b10006002414b861a2c8930ff3e60619a20059/Cargo.lock; sha256 = "0aca8gks7wl7yi821bcca16c94zcc41agin5j0vfz500i0sngzzx"; })
<aleph-> And angers me
hakujin has joined #nixos
jperras has joined #nixos
<aleph-> clever: Anyway to over ride the python version certain services use?
<aleph-> Want to run mopidy with py3, curious if it'll fix any of my issues.
markuskowa has joined #nixos
Taneb has quit [Quit: I seem to have stopped.]
hamishmack has joined #nixos
<{^_^}> [nixpkgs] @matthewbauer pushed to master « Add myself to a few CODEOWNERS paths »: https://git.io/fxpHY
<clever> aleph-: in that case, it just runs the "binary" inside mopidyEnv
<clever> aleph-: which is a wrapper around pkgs.mopidy that mutates PYTHONPATH
<aleph-> Gotcha...
<clever> aleph-: so you need to make an overlay that overwrites mopidy
<clever> 17909 mopidy = callPackage ../applications/audio/mopidy { };
<{^_^}> [nixpkgs] @dywedir opened pull request #49661 → bat: fix build on darwin → https://git.io/fxpH8
<clever> aleph-: and it uses pythonPackages as an arg, so you can mopidy = super.mopidy.override { pythonPackages = super.python3Packages; };
thc202 has quit [Ping timeout: 252 seconds]
<aleph-> Actually before I even do this, so reason I want to is from mopidy-scan failing to index files. Failing on a large number like such, "Failed local:track:30%20Seconds%20to%20Mars/30%20Seconds%20To%20Mars/01%20Capricorn%20%28A%20Brand%20New%20Name%29.mp3: gst-resource-error-quark: Could not open file "/media/Data/Audio/Music/>"
<aleph-> Which I figured could possible be an issue with unicode in file names from what I've been reading.
<aleph-> Not sure otherwise since it works fine with mpd and on an ubuntu machine,
<clever> aleph-: i have a different python program, which runs a function to recursively list a directory, and it tries to parse every single filename as utf8
<clever> and if any filename fails, it throws an exception, and doesnt tell you what file caused the problem
<clever> so you must then spend hours looking over strace output to find out why, lol
<aleph-> Huh, well it's failing on a lotttttt of files.
<aleph-> Like 1000's
<aleph-> Curious if py3 which fixed unicode a wee bit might help
<aleph-> Since there's not much out there for nixos
<clever> in my case, the exception isnt caught, and it unravels the stack hard enough to break my connection to the server
<clever> and upon reconnecting, it tries to list the same dir, and fails again
xourt has quit [Quit: WeeChat 1.4]
<azazel> maybe monkey patching builtins' open maybe a way to trace the file opening
ilikeheaps has quit [Ping timeout: 245 seconds]
iyzsong has joined #nixos
<azazel> clever: also, if you can run the program by hand, entering post-mortem debugger when the exception its raised maybe the quickest way to trace which file gives you issues
<azazel> s/its/is/
<clever> azazel: syncplay tends to catch the exception though, and then resume things, so that may not work
ilikeheaps has joined #nixos
<aleph-> This wouldn't even be a problem if I din't want to use my google music from ncmpcpp
<azazel> ncmpcpp?
<azazel> hell of an acronym:-)
<Mic92> worst name ever
<aleph-> Yeahhhhh
<aleph-> azazel: Terminal mpd client.
<azazel> oh
<Mic92> and the thing is ncmpc is also written in c++
<Mic92> so the name does not even make sense anymore.
<azazel> I'm still using SlimServer here... It seems the only one around to do a decent indexing of classical/jazz music
<aleph-> Okay might just be a timeout issue... fucking kidding me?
<{^_^}> [nixpkgs] @Mic92 merged pull request #49661 → bat: fix build on darwin → https://git.io/fxpH8
<{^_^}> [nixpkgs] @Mic92 pushed 2 commits to master: https://git.io/fxpQG
<aleph-> Apologies on the cursing.
iyzsong has quit [Ping timeout: 240 seconds]
<aleph-> Yeah it's totes a unicode issue I think.
<azazel> aleph-: it seems strange though, maybe the filesystem has a mount option that will help you fix these errors?
<aleph-> Maybe? Just on default ext4
<azazel> ah, copied from a remote or win fs?
Lears has joined #nixos
<aleph-> Nah, external hd
<aleph-> This is my laptop
<{^_^}> [nixpkgs] @leenaars opened pull request #49662 → pythonPackages.xapian: init -> 1.4.8 → https://git.io/fxpQ7
alex```` has quit [Ping timeout: 268 seconds]
<azazel> oh, reiserfs is still in mount's man page
<azazel> who would have thought that the fs would have such fate...
<aleph-> Man this is annoying.
<aleph-> Missing like 5000+ tracks
ilikeheaps has quit [Ping timeout: 240 seconds]
ilikeheaps has joined #nixos
<hyperfekt> i did it. bcachefs encrypted root works on nixos now :D as always with nixos, the solution was much simpler than i had initially anticipated
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/3fb4eb1c432 (from 45 minutes ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
<azazel> hyperfekt: bcachefs! I've read something but never used it
<jasongrossman> hyperfekt: That's good news!
<{^_^}> [nixpkgs] @flokli merged pull request #49658 → gitlab: refactor and fix test → https://git.io/fxp6z
<{^_^}> [nixpkgs] @flokli pushed 2 commits to master: https://git.io/fxp7w
markuskowa has quit [Quit: Page closed]
<hyperfekt> azazel: it's awesome because not only do you get native caching, you also get native encryption >with authentication<. and i think erasure coding is being finished up right now
<azazel> nice
<hyperfekt> (dm-crypt doesn't give authentication, so this (disregarding the lack of vetting) is basically the best full disk encryption in existence right now
<azazel> what about activating encryption directly supported by the drive? never tried it
<ottidmes> azazel: that is what I use
<ottidmes> azazel: with sedutil
mayhewluke has quit [Ping timeout: 272 seconds]
<azazel> ottidmes: ah, is sedutil available in nixos? Plus, did the "resume" issues that plagued such encryption have been resolved?
zack_moe_ has quit [Ping timeout: 252 seconds]
<ottidmes> azazel: I will check, if not, I really should make a pull request
mayhewluke has joined #nixos
<azazel> ottidmes: it is it seems, sedutil is in nixpkgs
<ottidmes> azazel: no, that is a fundamental issue with how it works I believe
<azazel> ottidmes: you are referring to the suspend/resume issues?
<azazel> it seem quite a big issue to me, I couldn't live without suspend and resume
<ottidmes> azazel: yes
<ottidmes> azazel: ah now I remember, I believe only hibernate (full power off) is not working, because then your whole disk has to be reencrypted
<ottidmes> azazel: you can workaround that also, but I am not sure what the status on that is
<hyperfekt> azazel: Unless the drive uses a really weird sector size, i highly doubt that'll give you authentication. for that to work you need to store a bit of data (i think it's called an HMAC) with every piece of data you want to be able to individually decrypt (like a sector)
<ottidmes> azazel: but I think I should still do a pull request at some point, because my sedutil package has a bit more, and I also have a bunch of useful sedutil scripts that make it a lot easier to work with
<azazel> ottidmes: oh, you secret stash
<{^_^}> [nixpkgs] @flokli opened pull request #49663 → gitlab: 11.4.3 -> 11.4.4 → https://git.io/fxp7p
<ottidmes> azazel: basically sedutil alone is quite low level, while my scripts do everything you are likely to do with sedutil without having to check the wiki every time :P
<azazel> ottidmes: but from what I read, you just only need to use sedutil during the setup or if you want to change/remove the password
hakujin has quit [Quit: WeeChat 2.3]
<hyperfekt> older versions of nixos suggested what packages provide a binary when it wasnt found, anyone know how to reenable that?
<ottidmes> azazel: to lock/unlock the drive, to enable/disable drive encryption, to set the PBA (Pre Boot Authorization, basically a small partition that is there when it is locked, to allow you to unlock) and to set/change the password
<azazel> hyperfekt: mine still does that
aristid has quit [Ping timeout: 252 seconds]
<azazel> hyperfekt: it uses the command-not-found package (from what my shell) says.. realy?
aristid has joined #nixos
<azazel> mmm... it's not in nixpkgs...
<ottidmes> hyperfekt: also very useful: https://github.com/bennofs/nix-index
<hyperfekt> hmm, i get an error 'unable to open database' when i run command-not-found. maybe it's broken
<clever> hyperfekt: are you using nix-channel and one of the proper channels?
<hyperfekt> no, i'm on unstable. from your question i assume that's a requirement for it to work?
jperras has quit [Ping timeout: 268 seconds]
<ottidmes> hyperfekt: it depends on how you have setup your NixOS, if you use channels, then it should work, but just use my own nixpkgs repo, so I too get:
<ottidmes> cannot open database `/nix/var/nix/profiles/per-user/root/channels/nixos/programs.sqlite' at /run/current-system/sw/bin/command-not-found line 13.
<clever> hyperfekt: command-not-found uses a database in the channel
<clever> ottidmes: programs.sqlite is the older command-not-found system
<hyperfekt> i mean, i made an installer from an unstable fork. i'll switch to the channel and check if it works
<clever> ah, the new one is also using programs.sqlite
<hyperfekt> clever: it's uncanny how you know the answer to every problem ^-^'
<ottidmes> clever: I just ran command-not-found in my shell, so internally it now uses something else?
<clever> -r--r--r-- 1 root root 3.3M Dec 31 1969 /nix/var/nix/profiles/per-user/root/channels/nixos/programs.sqlite
<hyperfekt> clever++
<{^_^}> clever's karma got increased to 39
<clever> ottidmes: you need a channel called nixos, on the root user, that came from nixos.org
hamishmack has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<ottidmes> clever: I know, that is why it is not working for me, but I have other tools to help if I dont know which package has some missing binary
fendor_ has quit [Remote host closed the connection]
<clever> ottidmes: you can also use nix-index and nix-locate directly to update your own db
<ottidmes> how does one use rustPlatform from nightly, I tried just adding latest.rustChannels.nightly.rust to my systemPackages, but it does not seem to have effect, how do I use the nightly rustPlatform?
fendor_ has joined #nixos
<ottidmes> clever: yep, those are the other tools I meant
<ottidmes> clever: but you mean I can learn command-not-found to use my own db?
<ottidmes> or whatever mechanism is used now
<clever> last i looked, command-not-found has that path hard-coded
<clever> 11 my $dbPath = "/nix/var/nix/profiles/per-user/root/channels/nixos/programs.sqlite";
<clever> still is
jperras has joined #nixos
<ottidmes> clever: but if nix-index can build a valid programs.sqlite, I could just create said path and put it there if need be, or patch command-not-found
init_6 has joined #nixos
<clever> ottidmes: i had heard that this was improved to use nix-index, but the source says otherwise
<clever> this script is responsible for generating the tars that nix-channel downloads
<clever> and it generates the programs.sqlite db
<{^_^}> [nixpkgs] @matthewbauer merged pull request #49417 → Disallow references to nativeBuildInputs → https://git.io/fx7X1
<{^_^}> [nixpkgs] @matthewbauer pushed 19 commits to staging: https://git.io/fxp5r
<hyperfekt> fixed it :) i'll remember for me it's neither ideal nor necessary to use a fork at any point, now that i know of disabledModules.
jperras has quit [Ping timeout: 252 seconds]
ilikeheaps has quit [Ping timeout: 252 seconds]
<{^_^}> [nixpkgs] @dtzWill opened pull request #49664 → bup: 0.29.1 -> 0.29.2 → https://git.io/fxp5P
<ottidmes> clever: if it still just calls command-not-found, it might be easier to just overlay it with one that uses nix-locate
<clever> ottidmes: yeah, thats always an option
<clever> just call nix-locate bin/$1
Mr_Keyser_Soze has quit [Ping timeout: 268 seconds]
Ariakenom has quit [Quit: Leaving]
erictapen has quit [Ping timeout: 272 seconds]
ilikeheaps has joined #nixos
nD5Xjz has quit [Ping timeout: 252 seconds]
teto has quit [Ping timeout: 246 seconds]
mvnetbiz has quit [Ping timeout: 252 seconds]
Mr_Keyser_Soze has joined #nixos
hamishmack has joined #nixos
nD5Xjz has joined #nixos
<ottidmes> clever: (hiPrio (pkgs.writeShBin "command-not-found" ''exec ${pkgs.nix-index}/bin/nix-locate /bin/"$1"'')) this works, but I am not sure when it should be run, because I just get "command not found" whenever I try something
<clever> ottidmes: is that in nix-env or systemPackages?
<__red__> So, I'm a little confused by something in the nix manual.
ilikeheaps has quit [Ping timeout: 268 seconds]
<ottidmes> clever: systemPackages, and you might need writeScriptBin instead of my writeShBin
<__red__> If I want to edit nix.conf to change the number of cores, /etc/nix/nix.conf is a symlink to /etc/static/nix/nix.conf
<__red__> which is linked into the nix store
<clever> ottidmes: double-check what is in /run/current-system/sw/bin/command-not-found, and try lib.mkBefore, rather then hiPrio
<clever> ottidmes: i think hiPrio only works on nix-env, and has no impact on systemPackages
<{^_^}> [nixpkgs] @matthewbauer pushed to master « gemrb: add libiconv »: https://git.io/fxp5A
<ottidmes> clever: it sure has an impact
<clever> __red__: https://nixos.org/nixos/options.html#nix.buildcores
<__red__> which means... I probably shouldn't be editing that directly
<clever> __red__: line 1-3 should also explain that
<__red__> looking
<ottidmes> clever: I have been using it to fix conflicts
<clever> 1-3 of the existing nix.conf
<clever> ottidmes: what is the contents of the current /run/current-system/sw/bin/command-not-found ?
<ottidmes> clever: /nix/store/1za07b36f2vi0c315fcq1xmh6r30f9yq-command-not-found/bin/command-not-found
<__red__> clever: so it's built from configuration.nix? I should have seen that coming
<clever> ottidmes: and if you just cat it?
<clever> __red__: yep
<{^_^}> [nixpkgs] @matthewbauer pushed to master « gsasl: disable check on darwin »: https://git.io/fxp5h
<ottidmes> clever: exec /nix/store/gir1i0lim2594lx4ykc6ydy4wiknw9vh-nix-index-0.1.1/bin/nix-locate /bin/"$1"
<clever> ottidmes: the #! appears to be missing
<ottidmes> clever: no its there, I just did not copy it
<ottidmes> clever: #!/bin/sh
<clever> ottidmes: and if you manually run nix-locate /bin/hello, it finds something?
<ottidmes> clever: yes
<ottidmes> clever: if I manually run command-not-found hello it works too
<clever> ottidmes: aha, `type command_not_found_handle`, run that
<ottidmes> clever: that is why I was wondering, what is triggering it, I thought maybe because I was using zsh it did not work, but in bash I dont see anything either
ilikeheaps has joined #nixos
<ottidmes> clever: thats it
<clever> ottidmes: the real thing doing the job, is the bash function command_not_found_handle, which has a full storepath baked into it
<ottidmes> clever: hardcoded path to the old one
<clever> /home/clever/nixpkgs/nixos/modules/programs/command-not-found/command-not-found.nix: command_not_found_handle() {
init_6 has quit []
<clever> ottidmes: it refers to a variable in a let block, so there is no way to override it
fendor_ has quit [Remote host closed the connection]
<clever> youll just need to re-define that bash function again
<clever> and perhaps programs.command-not-found.enable = false; to make the old one get out of the way
<clever> then you dont have to force priority
fendor_ has joined #nixos
<clever> oh!
<ottidmes> clever: good case of disabledModules :)
<clever> programs.command-not-found.dbPath
<samueldr> (with a custom command-not-found which wraps around nix-locate)
<clever> the nixos module already lets you change the db path
<samueldr> updated gist with the wrapper script (uses ruby)
<samueldr> though almost none of that is "well integrated" in my systen :/
<clever> one moment, grabbing something else of use
<clever> all of the nixos options involving shell init, and what file each winds up in
<clever> environment.shellInit, programs.bash.shellInit, environment.loginShellInit all wind up naked in /etc/profile
<clever> environment.interactiveShellInit winds up in /etc/bashrc
<ottidmes> samueldr: thanks, I am now going to write my own little module to replace the existing one for command-not-found
<clever> environment.extraInit winds up in a special set-environment file, that is sourced by /etc/profile and a few other things
<samueldr> users.users.clever.isNormalUser = true; # <- understatement :)
<clever> samueldr: lol :D
<ottidmes> clever: thats useful indeed, had to figure that out initially, but now that I have quite a lot of config, I tend to just look at my previous uses so I can infer it from that (which particular shell attr to use)
<samueldr> my command_not_found_handle is defined in my definitely-not-nix-managed bash profile, and the command-not-found command is `nix-env --install`ed from a private repo with WIP script :/ though the handle alias should be safe to use as long as `command-not-found` is eventually available in PATH
<clever> samueldr: heh, need to switch over to something like environment.interactiveShellInit!
<__red__> weird
<__red__> the installPhase fails, but if I use make install it succeeds
<samueldr> slowly moving towards something like that, where everything will be in the system... it's hard to break the mold from >10y of use
<__red__> guess I have more reading up to do on the installPhase
<clever> __red__: the default installPhase calls `make install` witl a lot more params
<__red__> I guess it's one of the params that's causing the issue
<clever> line 1085 will print those extra flags before running `make install`
<__red__> yup, I turned verbose on and knowthe flags its giving
<__red__> so I should run the three combinations and find out which makes it fail
<{^_^}> Channel nixpkgs-18.03-darwin advanced to https://github.com/NixOS/nixpkgs/commit/0e614d02923 (from 23 hours ago, history: https://channels.nix.gsc.io/nixpkgs-18.03-darwin)
<__red__> yup, it doesn't like the INSTALL= part
<__red__> odd
<clever> that will set the ${INSTALL} var inside the makefile
<__red__> which is what I expected
<__red__> oh, I see the problem
<__red__> its set INSTALL=$(out)
<__red__> whereas the makefile thinks that INSTALL should be the install binary (-c)
<clever> heh
<clever> that var is usually used to make debian packages for ex
<clever> basically, you set both INSTALL (to a dummy root dir) and PREFIX (to /usr/ for ex)
<__red__> right
<clever> then it will install to /dummy/usr/ but expect files to be in /usr/ at runtime
<clever> and now its not mixed in with other packages, so you can tar it up and make a package
<__red__> so - I guess I need to either manually run make install by making that my build [hase
<__red__> or find a way to unset that variable
<clever> also, i read setup.sh far too often
<clever> i just found 2 open tabs, both to installPhase, lol
lassulus_ has joined #nixos
<ottidmes> samueldr: why the parentheses in your command-not-found exists check?
pie_ has quit [Ping timeout: 268 seconds]
<samueldr> pretty sure they're not necessary, probably an artifact from a previous variation where the subshell had a command first
<clever> __red__: i cant see any obvious source for that INSTALL=, so i would just `installPhase = "make install";` and your done
<samueldr> after all, within a bash `if` you can write any commands, including subshells
<samueldr> (in nore naïve sh compatible shells, `[` itself is a command)
<samueldr> which '[' # -> /run/current-system/sw/bin/[
<samueldr> "/run/current-system/sw/bin/[" --version # -> [ (GNU coreutils) 8.29
<{^_^}> [nixpkgs] @dtzWill opened pull request #49665 → tootle: 0.1.5 -> 0.2.0 → https://git.io/fxpdi
georges_ has quit [Ping timeout: 264 seconds]
<samueldr> (though bash has the '[' builtin it will prefer, `type '['` will show that)
lassulus has quit [Ping timeout: 268 seconds]
lassulus_ is now known as lassulus
<ottidmes> samueldr: ok, that matched my thoughts, hence my suprise
<ottidmes> samueldr: I now use: if ! type command-not-found > /dev/null 2>&1; then
<{^_^}> [nixpkgs] @averelld opened pull request #49666 → r-ps: patchShebangs in ./configure to fix build → https://git.io/fxpdD
<clever> samueldr: you can also `command [ --version`
<__red__> clever: done - and thanks :-)
<clever> samueldr: or not, ack!
<clever> samueldr: `command` usually forces bash to search PATH and ignore internals
<samueldr> :D
<clever> but [ seems to have very special parsing, lol
<samueldr> `command`'s help says it suppresses shell function lookups... so I'm not sure what it does with builtins
<ottidmes> other than compatiblity, is there nowadays any reason to use [ of [[
<clever> $ command time --version
<clever> time (GNU Time) 1.9
<clever> time is a shell keyword
<clever> samueldr: it is able to supress a keyword at least
<samueldr> `command printf --version` `$(which printf) --version` <- doesn't suppress builtins
<ottidmes> to use [ over [[, I just use [[ everywhere except in those that need to be /bin/sh compatible
<samueldr> ottidmes: haven't validated, but doesn't `[[` not handle file checks?
<samueldr> hm no, it should (according to help)
<samueldr> guess there's not much reasons, except to not use bashism
<__red__> btw, the mac osx builds that are integrated into github - are those also done on a hydra server or is there a different way to test it
<__red__> or cross-compiled even?
<ottidmes> samueldr: as soon as your scripts starts to become somewhat involved, you will not be compatible anymore anyway, so I try to just stick with bashim when I can, and clearly state I am using bash
<clever> __red__: there arent really any good cross-compile options for targeting darwin right now
<samueldr> ottidmes: exactly, as long as you make it "clear" that it's not posix sh compliant, it isn't an issue imho :)
<ottidmes> samueldr: So I use /bin/sh style only at places like when you only have /bin/ash (some pre boot script), /usr/bin/env bash everywhere else, and zsh as my goto shell :P (bash is almost the default when searching about scripts)
<__red__> since I'm compiling up a cross-compiler for another platform trying to cross-compile it to a third platform sounds a little inception-esque for me ;-)
silver has quit [Read error: Connection reset by peer]
<{^_^}> [nixpkgs] @matthewbauer pushed 3 commits to master: https://git.io/fxpFC
Supersonic has quit [Disconnected by services]
Supersonic112 has joined #nixos
Supersonic112 is now known as Supersonic
Synthetica has quit [Quit: Connection closed for inactivity]
<{^_^}> [nixpkgs] @Mic92 merged pull request #49629 → pyre: 0.0.14 -> 0.0.17 → https://git.io/fxxsK
<{^_^}> [nixpkgs] @Mic92 pushed 4 commits to master: https://git.io/fxpF2
<feep> yo
<feep> what's the package that has libsqlite3.so?
<ottidmes> feep: use nix-locate from nix-index: https://github.com/bennofs/nix-index
<ottidmes> feep: wineStaging, sqliteInteractive, sqlite, and sage
dbmikus has joined #nixos
<ottidmes> feep: so I would go with sqlite
annadane has joined #nixos
<feep> nm worked it out
<feep> thanks
<annadane> hi, about how long is the stable release cycle?
dbmikus has quit [Client Quit]
dbmikus has joined #nixos
<hyperfekt> is there any way as part of nix or the stdlib to condition multiple (but not all) definitions on one mkIf?
<ottidmes> annadane: every half year a stable release is made, most safe things will be backported to stable though
<infinisil> hyperfekt: Like?
<ottidmes> hyperfekt: you mean config = mkIf ... { ... } ?
<annadane> ottidmes, does it follow a "when it's ready" approach or set date, judging from a quick glance through the manual it looks so far to be set date
<infinisil> hyperfekt: Can do it like this too: `{ lib, ... }: lib.mkIf (...) { ... }`
<infinisil> Or if you need to have unconditioned things too:
<infinisil> { lib, ... }: with lib; mkMerge [ { ... } (mkIf (...) { ... }) ]
<ottidmes> annadane: set date for when it it supposed to be ready, but its not a hard deadline, they try to make the release as good as possible, so they rather wait a bit longer for a release and fix remaining problems, than holding to the date
<hyperfekt> infinisil: That's exactly what I was looking for, thanks :)
annadane has left #nixos ["Leaving"]
<ottidmes> annadane: in general its best to wait a month or so after the set date to update to the next stable, than its likely everything has updated to point to that version (documentation and prebuilt images and such), but that has just been my experience
<drakonis> annadane left hmm
<drakonis> maybe you should direct your messages to them
<ottidmes> Drakonis: good point :P hate it when that happens, had it happen to me a few time after making a package for someone...
<drakonis> ha
<drakonis> annadane is still on the network
<drakonis> wait no
<drakonis> sorry about that
<ottidmes> Drakonis: ah well, maybe someone googles this particular questions, and finds this log, so it might still be useful. I have been able to answer some of my questions that way in the past
<drakonis> they'll be back and i'll refer to the messages
<drakonis> whenevver they're on oftc again
<infinisil> ,tell
<{^_^}> infinisil: Use `,tell john Remember to do the laundry` to send this to john next time he's talking in this channel
<hyperfekt> infinisil++
<{^_^}> infinisil's karma got increased to 32
<drakonis> infinisil++
<{^_^}> infinisil's karma got increased to 33
<drakonis> love that command
<infinisil> tell? It's not too useful because it can't send the message when the person joins, it has to receive a message
<infinisil> (yet)
<ottidmes> anyone know how to use rustPlatform of rust nightly? I have the nixpkgs mozilla overlay installed
dbmikus has quit [Read error: Connection reset by peer]
sigmundv has quit [Ping timeout: 246 seconds]
georges has joined #nixos
<{^_^}> [nixpkgs] @benley opened pull request #49668 → stern: Include bash & zsh completion support → https://git.io/fxpbY
dbmikus has joined #nixos
Mr_Keyser_Soze has quit [Ping timeout: 240 seconds]
Mr_Keyser_Soze has joined #nixos
dramforever has joined #nixos
hellrazor has joined #nixos
dbmikus has quit [Quit: WeeChat 2.3]
dbmikus has joined #nixos
hellrazo1 has quit [Ping timeout: 268 seconds]
ottidmes has left #nixos ["WeeChat 2.2"]
nD5Xjz has quit [Ping timeout: 240 seconds]
dramforever has quit [Ping timeout: 256 seconds]
Mr_Keyser_Soze has quit [Ping timeout: 272 seconds]
fendor_ has quit [Read error: Connection reset by peer]
mayhewluke has quit [Ping timeout: 252 seconds]
mayhewluke has joined #nixos
ilikeheaps has quit [Read error: Connection reset by peer]
ilikeheaps has joined #nixos
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/5ea21ad32ad (from 2 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
dbmikus has quit [Quit: WeeChat 1.9.1]
MichaelRaskin has quit [Quit: MichaelRaskin]
drakonis has quit [Read error: Connection reset by peer]
<{^_^}> [nixpkgs] @kalbasit opened pull request #49669 → Provide all the binaries needed by the vim-go plugin → https://git.io/fxpAO
<yl[m]> Mic92: the above PR is the vim-go fix we discussed a few days ago here and in #49498
<{^_^}> https://github.com/NixOS/nixpkgs/pull/49498 (by kalbasit, 3 days ago, closed): neovim: take withExtraPackages and add them to the PATH
rhitakorrr has quit [Quit: WeeChat 2.2]
slack1256 has joined #nixos
ThatDocsLady has joined #nixos
<{^_^}> [nixpkgs] @matthewbauer pushed to master « Revert "stage.nix: pkgsi686Linux only works on x86 family" »: https://git.io/fxpxT
endformationage has quit [Quit: WeeChat 1.9.1]
ThatDocsLady has quit [Ping timeout: 264 seconds]
slack1256 has quit [Remote host closed the connection]
slack1256 has joined #nixos
<{^_^}> [nixpkgs] @matthewbauer opened pull request #49670 → Fix pkgsi686Linux assertion → https://git.io/fxpx2
<{^_^}> [nixpkgs] @hyperfekt opened pull request #49671 → bcachefs root support → https://git.io/fxpxp
jsgrant- has joined #nixos
jD91mZM2 has joined #nixos
zack_moe has joined #nixos
<hyperfekt> here we go :D it's rather minimal, but should suffice for the most common scenario. this is the first 'real' PR that i've ever made :)
jbgi has joined #nixos
<hyper_ch> bcachefs has integrated encryption?
jbgi has quit [Ping timeout: 252 seconds]
<hyperfekt> hyper_ch: yep. it's pretty cool because you get authentication, which schemes like dm_crypt can't do
<hyper_ch> hyperfekt: I use native root encryption on zfs
<hyperfekt> does that use a block layer? since zfs uses blocks instead of extents i don't think they have something like bcachefs, right?
<hyper_ch> no idea where the difference is
aleph- has quit [Ping timeout: 245 seconds]
ddellacosta has quit [Remote host closed the connection]
<hyperfekt> hyper_ch: basically with your encryption a sufficiently advanced adversary could shuffle parts of your filesystem around, or do chosen-ciphertext attacks
<hyperfekt> it's not great, but so far it was the best we had
<hyper_ch> hyperfekt: https://xkcd.com/538/
* colemickens is trying to read up on authentication in encrpytion
<colemickens> Are either of you running a / on bcachefs?
<hyper_ch> couldn't I just shuffle tings around with DD anyway?
<hyperfekt> okay, it doesn't have plausible deniability yet :b
<hyperfekt> well, you could still shuffle them around, but bcachefs would notice if you replaced ls with a payload in that untrusted pdf you made me download
<hyperfekt> colemickens: yes, I am :)
<colemickens> I am trying to figure out if I can get my root partition moved over w/o getting out of bed :)
<hyper_ch> zfs would also notice that things got shuffled around
<hyper_ch> colemickens: hire 10 indian programmers to move it over
<elvishjerricco> I'm pretty sure that ZFS even ensures that identical blocks are unidentifiable to an attacker (unless you have dedup on, in which case that info is leaked elsewhere).
<colemickens> ha, I used LVM so I might have a shot
<elvishjerricco> (which is dumb. ZFS's native encryption is unforunately leaky)
<hyper_ch> I never liked lvm much
<hyperfekt> hyper_ch: not if i change the checksums along with it ;) it's definitely hard and more of a theoretical property rather than something available in exploit kits (today), but the cryptographers aren't making this stuff up because they're bored
<colemickens> I didn't either until I needed to do what I'm doing now :)
dontdieych has quit [Quit: ZNC 1.7.1 - https://znc.in]
<hyperfekt> oh, i was assuming a block layer. i don't know much about zsf' encryption
<elvishjerricco> Yea I don't know the implementation details, but the acronym MAC comes to mind.
<hyper_ch> elvishjerricco: how is zfs's encryption leaky?
<hyper_ch> colemickens: what are you doing now?
<colemickens> er, for some reason I thought LVM helped with online resize, but I guess it doesn't change the fact that ext4 isn't going to allow it.
<colemickens> In retrospect, I don't remember why I used LVM here :S
<hyper_ch> ext4 can be resized
<elvishjerricco> hyper_ch: It leaks 100% of the pool structure, which includes dataset hierarchy and names, dedupe tables, and file sizes.
<hyper_ch> elvishjerricco: ah ok
<elvishjerricco> This is because they wanted to support using encryption on only some of your datasets, and encrypting the sub-pool-structure would have required a massive re-engineering of the pool structure
<hyperfekt> I looked it up, looks like ZFS uses a similar scheme as bcachefs, that does authentication as well. Nice!
<elvishjerricco> hyperfekt: There was a talk given about it somewhere... Like an hour of detailed info that went over my head :P
<hyper_ch> you probably mean tom caputi's talk
<elvishjerricco> Still, I'm not a fan because of the leaky pool structure. I'm sticking to ZFS over LUKS, since Grub can read that stuff and /boot can sit in my ZFS pool.
<hyper_ch> I love native zfs encryption
<jasongrossman> elvishjerricco: ZFS leaks file sizes?! That's the first thing I've heard that's worrying from the point of view of real-world compromises.
<elvishjerricco> jasongrossman: I'm not sure how precise it is. The manual doesn't specify. Could be exact file sizes or just the number of blocks used.
<elvishjerricco> Still bad, but slightly less bad if it's just number of blocks
<jasongrossman> elvishjerricco: Hm. Number of blocks is not so obviously worrying. Right.
<hyper_ch> and what if you have compression enabled?
<hyper_ch> what is reported then?
<elvishjerricco> The dedupe table thing is the part that frustrates me the most. Like that's an actually enormous leak. And I know dedupe is discouraged for most people anyway, but still
<elvishjerricco> hyper_ch: No clue how ZFS's compression works TBH :P
<hyper_ch> it works by adding compression=xxx when creating a dataset or inheriting it from a parent dataset :)
<elvishjerricco> Heh yea I've got 2x compression ratio on all my machines. Freakin miracle stuff.
<hyper_ch> lz4?
<jasongrossman> hyper_ch: Made me laugh. :-) (Which is a sign that this should all be on nixos-chat.)
<elvishjerricco> I just used the default. Don't remember which that was
<hyper_ch> I think setting it to on will use lz4
<elvishjerricco> I switched my desktop over to ZFS tonight with compression enabled, and followed up with `nix-store --optimise` for the first time ever. Lowered my disk usage by 65%
<hyper_ch> tankSubi/encZFS/Nixos compressratio 1.46x -
<hyper_ch> especially on SSDs its nice to free up space
a_ has joined #nixos
<hyper_ch> actuall with switch from ext4 to zfs it freed up double because of the default 5% reservation for root
<hyperfekt> compression is really cool. i think bcachefs can even use different algorithms on your cache and your main device, recompressing more strongly in the background. perfect with the decryption speeds lz4 offers
<{^_^}> [nixpkgs] @aszlig pushed to master « autoPatchelfHook: Correctly detect PIE binaries »: https://git.io/fxpp1
<elvishjerricco> hyperfekt: I thought bcachefs didn't actually support getting more space out of its compression yet? Like I know it has a feature called "compression," but the homepage says you won't actually get to store more data with it.
<emily> good feature
<hyperfekt> True, the accounting isn't quite there yet, because it's somewhat complicated to do in a way that doesn't just work, but is optimal.
<hyper_ch> "Not quite finished - it's safe to enable, but there's some work left related to copy GC before we can enable free space accounting based on compressed size: right now, enabling compression won't actually let you store any more data in your filesystem than if the data was uncompressed"
<hyperfekt> If you want the technical details, they're here https://www.patreon.com/posts/bcachefs-extents-20740671
a_ has quit [Client Quit]
dontdieych has joined #nixos
drakonis_ has quit [Ping timeout: 268 seconds]
drakonis_ has joined #nixos
Theuni2 has joined #nixos
<{^_^}> [nixpkgs] @gnidorah opened pull request #49672 → maxx: fix urls → https://git.io/fxphr
alex```` has joined #nixos
<{^_^}> [nixpkgs] @danieldk opened pull request #49673 → cargo-asm: fix build on macOS Mojave → https://git.io/fxphi
alex```` has quit [Ping timeout: 244 seconds]
mayhewluke has quit [Ping timeout: 252 seconds]
mayhewluke has joined #nixos
<chris|> lassulus: any interesting findings? :)
<{^_^}> [nixpkgs] @dywedir merged pull request #49673 → cargo-asm: fix build on macOS Mojave → https://git.io/fxphi
<{^_^}> [nixpkgs] @dywedir pushed 2 commits to master: https://git.io/fxpjs
<{^_^}> [nixpkgs] @typetetris opened pull request #49674 → haskellPackages.sdl2: test-suite needs x server, so deactivate it → https://git.io/fxpj1
agander has joined #nixos
srhb has quit [Remote host closed the connection]
ddellacosta has joined #nixos
orivej has quit [Ping timeout: 245 seconds]
<{^_^}> [nixpkgs] @clefru closed pull request #49515 → androidsdk: Re-add emulator → https://git.io/fxbeR
ddellacosta has quit [Ping timeout: 268 seconds]
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/c64624b8435 (from 2 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
<colemickens> hyperfekt: how are bcachefs snapshots these days?
<colemickens> Last update from the dev was in March...
<{^_^}> [nixpkgs] @asymmetric opened pull request #49675 → pythonPackages.prettytable: 0.7.1 -> 0.7.2 → https://git.io/fxhes
srhb has joined #nixos
jsgrant- has quit [Remote host closed the connection]
jbgi has joined #nixos
<hyperfekt> colemickens: Nonexistent, and it's gonna take a considerable while and a diskformat change until we get them. My only real gripe :/
<colemickens> Okay. I figure I don't have them anyway, so I can't really resist for that reason.
<hyperfekt> Kent is working on that filesystem alone, so you can't expect any miracles. I hope that changes soon.
<colemickens> Yeah, I read through some docs and the Patreon page. Sometimes it's surprising to me how few people it takes to move the needle on important things even like an FS for Linux.
<colemickens> No doubt that much of that is due to Kent's talent, but still.
<colemickens> Wish Facebook would throw some money at him, like they did the BTRFS author(s)
erictapen has joined #nixos
<hyperfekt> Yeah, I cannot grasp how there is almost no investment on the most basic parts of the most important software ecosystem in the world.
Ariakenom has joined #nixos
<{^_^}> [nixpkgs] @jethrokuan opened pull request #49676 → gin-config: init → https://git.io/fxhe9
nschoe has joined #nixos
slack1256 has quit [Ping timeout: 244 seconds]
<nschoe> 'morning everyone
FRidh has joined #nixos
<nschoe> I have been reading a lot of nix,nixOS-related stuff these last couple of weeks, and I just wanted to say that the nix-pills helped _a lot_ to clarify some shadowy spots. So thanks again fro writing them, it's awesome :)
jbgi has quit [Ping timeout: 250 seconds]
srhb has quit [Remote host closed the connection]
crmlt_ has joined #nixos
crmlt_ has left #nixos [#nixos]
<{^_^}> [nixpkgs] @peti merged pull request #49674 → haskellPackages.sdl2: test-suite needs x server, so deactivate it → https://git.io/fxpj1
<{^_^}> [nixpkgs] @peti pushed 2 commits to master: https://git.io/fxhvk
<{^_^}> [nixpkgs] @peti merged pull request #49666 → r-ps: patchShebangs in ./configure to fix build → https://git.io/fxpdD
<{^_^}> [nixpkgs] @peti pushed 2 commits to master: https://git.io/fxhvY
<{^_^}> [nixpkgs] @c0bw3b merged pull request #49467 → cmusfm: init at 2018-10-11 → https://git.io/fxd3u
<{^_^}> [nixpkgs] @c0bw3b pushed 2 commits to master: https://git.io/fxhvs
<{^_^}> [nixpkgs] @taku0 opened pull request #49677 → thunderbird, thunderbird-bin: 60.2.1 -> 60.3.0 [Critical security fixes] → https://git.io/fxhvc
philippD has joined #nixos
<{^_^}> [nixpkgs] @peti pushed to haskell-updates « hackage-packages.nix: automatic Haskell package set update »: https://git.io/fxhv0
erictapen has quit [Ping timeout: 246 seconds]
Itkovian has joined #nixos
Peetz0r has quit [Ping timeout: 252 seconds]
Peetz0r has joined #nixos
jtojnar has joined #nixos
<{^_^}> [nixpkgs] @jtojnar closed pull request #49431 → regextester: 0.1.7 -> 1.0.1 → https://git.io/fx5OS
c0bw3b_ has joined #nixos
Peetz0r has quit [Ping timeout: 260 seconds]
drakonis_ has quit [Ping timeout: 246 seconds]
Peetz0r has joined #nixos
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<{^_^}> [nixpkgs] @symphorien opened pull request #49680 → nix-du: 0.2.0 -> 0.3.0 → https://git.io/fxhfu
Thra11 has joined #nixos
drakonis_ has joined #nixos
vidbina has joined #nixos
jasongrossman has quit [Ping timeout: 240 seconds]
Itkovian has joined #nixos
drakonis_ has quit [Ping timeout: 272 seconds]
drakonis_ has joined #nixos
electrocat has joined #nixos
<electrocat> how do i install glibc-dev? i want glibc headers in my .nix-profile
<symphorien> what for ?
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<electrocat> symphorien: autocompletion stuff for vim
<symphorien> I imagine nix-env -iA nixos.glibc.dev should work
<electrocat> doesn't seem to work
<electrocat> just installs ordinary glibc
<symphorien> hum anyway I don't know if nix-env puts $out/include in the profile
<electrocat> i can manually install it by passing the store path
<electrocat> rather unsatisfactory
periklis has joined #nixos
<{^_^}> [nixpkgs] @Profpatsch merged pull request #49649 → jq: 1.5 -> 1.6 (!!!!) → https://git.io/fxpZD
<{^_^}> [nixpkgs] @Profpatsch pushed commit from @dtzWill to master « jq: 1.5 -> 1.6 »: https://git.io/fxhJt
<steveeJ> is there a way to configure nix channels declaratively?
<electrocat> steveeJ: you can edit .nix-channels
grenade has quit [Ping timeout: 276 seconds]
<{^_^}> [nixpkgs] @gebner merged pull request #49680 → nix-du: 0.2.0 -> 0.3.0 → https://git.io/fxhfu
<{^_^}> [nixpkgs] @gebner pushed 2 commits to master: https://git.io/fxhJG
Mateon2 has joined #nixos
Mateon1 has quit [Ping timeout: 268 seconds]
Mateon2 is now known as Mateon1
<{^_^}> [nixpkgs] @Fuuzetsu merged pull request #49585 → libid3tag: patch CVE-2017-11550 and CVE-2017-11551 → https://git.io/fxN5s
<{^_^}> [nixpkgs] @Fuuzetsu pushed 2 commits to master: https://git.io/fxhJc
grenade has joined #nixos
agander has quit [Ping timeout: 246 seconds]
drakonis_ has quit [Ping timeout: 246 seconds]
<steveeJ> electrocat: in the user's directory you mean?
drakonis_ has joined #nixos
<electrocat> steveeJ: yes
vidbina has quit [Ping timeout: 268 seconds]
feep has quit [Disconnected by services]
nschoe has quit [Quit: Program. Terminated.]
nschoe has joined #nixos
<steveeJ> electrocat: will those be available in the configuration.nix?
<{^_^}> [nixpkgs] @flokli merged pull request #49663 → gitlab: 11.4.3 -> 11.4.4 → https://git.io/fxp7p
<{^_^}> [nixpkgs] @flokli pushed 2 commits to master: https://git.io/fxhUk
johanot has joined #nixos
iyzsong has joined #nixos
mayhewluke has quit [Ping timeout: 252 seconds]
mayhewluke has joined #nixos
GiGa has joined #nixos
<GiGa> Hello all
<GiGa> I'm on the unstable channel and getting this during a rebuild switch: "bsddb3-6.2.6 not supported for interpreter python3.6m"
<GiGa> I'll be honest, I don't entirely understand what's caused it (I suspect a change in the BSDDB3 package, or in Python) - has anyone any advice on how to resolve please?
srhb has joined #nixos
<FRidh> GiGa: yep, seems like you found a regression (caused by 45323efb5544114ea390c5fc9174c41dfb803d5c)
Thra11 has quit [Ping timeout: 268 seconds]
srhb has quit [Client Quit]
vk3wtf has quit [Ping timeout: 264 seconds]
<GiGa> FRidh: OK, anything I can do to help (with an experience level of "novice")?
<{^_^}> [nixpkgs] @FRidh pushed to master « python.pkgs.bsddb3: enable python3 and tests »: https://git.io/fxhUE
<FRidh> GiGa: I've pushed a fix to master
zolk3ri has joined #nixos
<{^_^}> [nixpkgs] @flokli opened pull request #49682 → [18.09] gitlab fixes → https://git.io/fxhUg
<GiGa> FRidh: OK, so I need to take my local checkout of nixpkgs, rebase it on to origin/master and then tell somehow use that?
noam has joined #nixos
<GiGa> Or just wait for master to come to unstable channel?
<johanot> nix eval -f file.nix someattr <- will give me the value of "someattr" from "file.nix" given that file.nix returns an attrset. Question: Is there a way to return the root attrset of file.nix? The second arg to nix eval doesn't seem to be optional.
<GiGa> I'm not desparate to upgrade today, just wanting to help if I can
<lassulus> chris|: I haven't looked into it to be honest, just travelled a couple hundred km and attended a party
<FRidh> GiGa: if you're in no hurry, you can wait. Otherwise, you can point to a checkout of the repo or a tarball. How, depends on how you use nixpkgs.
<GiGa> FRidh: I think given my experience level waiting is best then. Thanks
<GiGa> Although I can see an issue that suggests unstable hasn't been updated in two weeks
<{^_^}> [nixpkgs] @FRidh pushed 612 commits to staging-next: https://git.io/fxhUS
<{^_^}> [nixpkgs] @FRidh pushed 48 commits to staging: https://git.io/fxhU7
thc202 has joined #nixos
alex```` has joined #nixos
<WilliamHamilton[> is anyone using agda from github (version 2.6.0) and has a nix expression to share?
<GiGa> WilliamHamilton[: I can see it in NixPkgs
<WilliamHamilton[> that's the infrastructure for packaging, the Agda package is here `pkgs/development/haskell-modules/hackage-packages.nix`
<WilliamHamilton[> but that's for the current hackage version,, and I'd like the 2.6.0 one, which is on github
<WilliamHamilton[> alternative question: what's the equivalent of callPackage, but for git?
<GiGa> WilliamHamilton[: hmm, might be out of my depth, one sec, let me find one of mine
<GiGa> WilliamHamilton[: does this help at all? References source on Github
<GiGa> (by "mine" I mean one I've contributed to)
<WilliamHamilton[> GiGa: I'll take a look, thanks!
etu has quit [Quit: WeeChat 2.2]
<{^_^}> [nixpkgs] @vcunat pushed 3 commits to master: https://git.io/fxhTc
<{^_^}> [nixpkgs] @vcunat merged pull request #49677 → thunderbird, thunderbird-bin: 60.2.1 -> 60.3.0 [Critical security fixes] → https://git.io/fxhvc
Theuni2 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
schopp0r has left #nixos ["Error from remote client"]
schopp0r has joined #nixos
__monty__ has joined #nixos
Theuni2 has joined #nixos
Theuni2 has quit [Client Quit]
xorkle has quit [Ping timeout: 245 seconds]
fendor has joined #nixos
fendor has quit [Remote host closed the connection]
fendor has joined #nixos
xorkle has joined #nixos
ottidmes has joined #nixos
<{^_^}> [nixpkgs] @joachifm merged pull request #49652 → tor: -> → https://git.io/fxplf
<{^_^}> [nixpkgs] @joachifm pushed 2 commits to master: https://git.io/fxhTK
orivej has joined #nixos
johanot has quit [Quit: WeeChat 2.2]
acarrico has quit [Ping timeout: 272 seconds]
<{^_^}> [nixpkgs] @markuskowa merged pull request #49660 → yices: 2.6.0 -> 2.6.1 → https://git.io/fxpDz
<{^_^}> [nixpkgs] @markuskowa pushed 2 commits to master: https://git.io/fxhT6
nschoe has quit [Quit: Program. Terminated.]
srhb- has joined #nixos
<{^_^}> [nixpkgs] @Zimmi48 opened pull request #49684 → dunePackage: add support function and use it in a few packages → https://git.io/fxhTM
agander has joined #nixos
<srhb-> Anyone using the new znc module managed to get a relatively declarative setup going? I'm really struggling with it
<srhb-> Gah, why is my nick wrong...
<srhb-> And why on earth does the config not update when I change settings...
<{^_^}> [nixpkgs] @markuskowa merged pull request #49672 → maxx: fix urls → https://git.io/fxphr
<{^_^}> [nixpkgs] @markuskowa pushed 2 commits to master: https://git.io/fxhTb
<{^_^}> [nixpkgs] @rht opened pull request #49685 → brave-browser: 0.25.2 -> 0.55.22 → https://git.io/fxhTA
srhb- is now known as srhb
<{^_^}> [nixpkgs] @c0bw3b merged pull request #48496 → linuxPackages.bpftrace: init at unstable-2018-10-27 → https://git.io/fxunz
<{^_^}> [nixpkgs] @c0bw3b pushed 2 commits to master: https://git.io/fxhTh
hellrazor has quit [Quit: WeeChat 2.3]
<srhb> Wut, the prestart script seems to only copy the config file over the first time.
<infinisil> srhb: Here to help
<srhb> infinisil: How do I regenerate the config?
<infinisil> services.znc.mutable = false
<{^_^}> [nixpkgs] @smaret opened pull request #49686 → [WIP] matplotlib: 2.2.3 -> 3.0.0 → https://git.io/fxhkf
srhb has quit [Quit: ZNC 1.7.1 - https://znc.in]
hlolli has joined #nixos
srhb has joined #nixos
<srhb> That's better.
<{^_^}> [nixpkgs] @gnidorah opened pull request #49687 → sway-beta module: add missing pieces → https://git.io/fxhkq
<srhb> infinisil: The comment in the preStart seems to imply the opposite
<srhb> infinisil: "# If mutable, regenerate conf file every time"
<hlolli> Is there any cheap cloud solution for individuals which has nixos image included? There's to say no manual setup.
<infinisil> srhb: Oh you're right, never noticed this
agander_ has joined #nixos
<Akii> anyone an idea what is going wrong here? https://gist.github.com/Akii/745d7c959d18cf9ee27b6d1110255f97
<Akii> can find posts from 4 months ago with the same error but no solution
agander__ has joined #nixos
agander has quit [Read error: Connection reset by peer]
agander_ has quit [Ping timeout: 240 seconds]
alex```` has quit [Ping timeout: 252 seconds]
<__monty__> Akii: Maybe it's because mac os pretends clang is gcc?
orivej has quit [Ping timeout: 252 seconds]
xorkle has quit [Ping timeout: 272 seconds]
<{^_^}> [nixpkgs] @Mic92 opened pull request #49688 → nixos-install: provide user feedback when nixos install begins. → https://git.io/fxhk5
<Akii> I'll just yolo it or get a docker container if that fails
agander__ is now known as agander
ilikeheaps has quit [Ping timeout: 244 seconds]
<{^_^}> [nixpkgs] @taku0 opened pull request #49689 → adoptopenjdk-openj9-bin: 11 -> 11.0.1 [Critical security fixes] → https://git.io/fxhkb
<domenkozar> FYI: github finally added wildcard support, so now all branches that start with staging* or release* are protected by default
fendor_ has joined #nixos
srhb has quit [Quit: ZNC 1.7.1 - https://znc.in]
srhb has joined #nixos
ilikeheaps has joined #nixos
Theuni2 has joined #nixos
lassulus has quit [Ping timeout: 252 seconds]
jbgi has joined #nixos
<elvishjerricco> Trying to upgrade to NixOS 18.09 on my desktop that I deploy with NixOps (`targetHost = "localhost";`), and I'm getting an error I can't track down:
<elvishjerricco> `error: assertion failed at /nix/store/r63y7cyalmh3vfn0bd222zacajrh6611-source/pkgs/top-level/impure.nix:80:1`
<elvishjerricco> Anyone else get this one?
<tilpner> I remember hitting that too
<tilpner> But not with nixops
<elvishjerricco> tilpner: It may have nothing to do with the NixOps stuff
<tilpner> I think it does though
<GiGa> hlolli: Not as far as I know
<elvishjerricco> tilpner: Why's that?
<hlolli> After some google-ing, it looks like this isn't difficult, going to give this a go with hetzner https://github.com/jeaye/nixos-in-place/issues/41 seems very cheap option.
<{^_^}> jeaye/nixos-in-place#41 (by nh2, 28 weeks ago, open): Instructions on how to use this to make a Hetzner Cloud snapshot (includes moving /old-root/nixos to /)
<tilpner> elvishjerricco - Something is instantiating nixpkgs improperly. I would expect that task to fall into nixops' area of responsibility
<elvishjerricco> tilpner: Could be any of the NixOS modules setting `config.nixpkgs.system`, `config.nixpkgs.localSystem`, or `config.nixpkgs.platform`. Might be NixOps, but could be anything
srhb has quit [Quit: ZNC 1.7.1 - https://znc.in]
<tilpner> Oh, I'm not sure of course. Just saying to not rule out nixops yet
srhb has joined #nixos
jasongrossman has joined #nixos
jbgi has quit [Quit: WeeChat 2.2]
<hlolli> LOL, there's a Nixos image on hetzner: An ISO Image is mounted: "NixOS 18.03 (amd64/graphical)". Just pressed boot image, and rebooot, and now I can't connect hahaha.
<tilpner> hlolli - Can't connect? Try the web console?
<steveeJ> rycee: what's the sorting for the hm activation scripts? is there any way I can way I can run something before `Activating checkLinkTargets`?
<hlolli> tilpner: I'll look for a web console. Worst case, boot ubuntu again and do the NixOps trick.
<tilpner> hlolli - I was going to suggest Vultr, DigitalOcean and Hetzner as cheap NixOS-compatible hosts, but they didn't really fulfill your no-manual-action requirement
<elvishjerricco> steveeJ: Hey, the only google result for my problem is an IRC message from you on here back in June :P Any chance you remember how to fix `error: assertion failed at /nix/store/r63y7cyalmh3vfn0bd222zacajrh6611-source/pkgs/top-level/impure.nix:80:1`?
<hlolli> yes, it seems easier that I imagined. Just short on time, need to move one website between hosts. Thanks still tilpner!
Theuni2 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<jasongrossman> hlolli: RamNode has a NixOS ISO - you have to run the installer to get it installed, though. RamNode's cheap, and I've found it well run.
<hlolli> nice, already swaped my credit card. Hetzner has a NixosISO, and I'm inside the webconsole, and it works! Just need to enable ssh service and I should be in trough my terminal.
zack_moe_ has joined #nixos
<{^_^}> [nixpkgs] @c0bw3b merged pull request #49559 → bump version librepcb-unstable: 2018-06-28 -> 2018-10-31 → https://git.io/fxNkx
<{^_^}> [nixpkgs] @c0bw3b pushed 2 commits to master: https://git.io/fxhIH
zack_moe has quit [Ping timeout: 250 seconds]
civodul has joined #nixos
<{^_^}> [nixpkgs] @mat8913 opened pull request #49690 → firefox-esr-60: 60.2.2 -> 60.3.0 [critical security fixes] → https://git.io/fxhIx
demize has quit [Quit: Do Androids Dream of Electric Sheep?]
<rycee> steveeJ (IRC): config.lib.dag.entryBefore ["checkLinkTargets"] "your code"
<{^_^}> [nixpkgs] @Mic92 closed pull request #49688 → nixos-install: provide user feedback when nixos install begins. → https://git.io/fxhk5
<jasongrossman> hlolli: Yay!
<jasongrossman> c
demize has joined #nixos
<rycee> steveeJ (IRC): I.e. home.activation.yourphase = config.lib.dag.entryBefore ["checkLinkTargets"] "your code"
agander_ has joined #nixos
hyperfekt has quit [Quit: Page closed]
agander has quit [Ping timeout: 252 seconds]
Thra11 has joined #nixos
agander_ is now known as agander
betaboon has joined #nixos
<elvishjerricco> Wtf. `manual-combined.xml fails to validate`
<elvishjerricco> Lots of: `manual-combined.xml:3079: element xref: validity error : IDREF attribute linkend references an unknown ID "opt-boot.kernel.sysctl"`
<elvishjerricco> for various different `opt-boot.*`
<elvishjerricco> er, `opt-*`
hyperfekt has joined #nixos
<Mic92> Can I extend an existing submodule without having the edit its source code?
<elvishjerricco> I would also like to know the answer to that.
<steveeJ> rycee: thank you! I'll give it a try
<rycee> Mic92 (IRC): To my knowledge you cannot.
<rycee> Mic92 (IRC): But it is possible to do something similar using a list as argument to the submodule.
<rycee> To the "submodule" function, that is.
<rycee> Mic92 (IRC): Home manager uses that for the email account options: https://is.gd/VqTLYt
<elvishjerricco> rycee: Is that effectively the same as `submodule { imports = [...]; }`?
srhb has quit [Quit: ZNC 1.7.1 - https://znc.in]
<rycee> elvishjerricco (IRC): Yeah, I believe it is pretty much exactly the same.
srhb has joined #nixos
<steveeJ> rycee: thanks to your help I can now use home-manager to configure my system's nix-channels by setting up home-manager for root in configuration.nix :D
<hyperfekt> I've made two pull requests, is there anything else I need to do to direct the attention of the right person to them or is it just a matter of waiting until someone gets to it?
<rycee> steveeJ (IRC): Cool :-)
Travankor_ has joined #nixos
Travankor_ has quit [Client Quit]
Theuni2 has joined #nixos
vidbina has joined #nixos
<{^_^}> [nixpkgs] @srhb opened pull request #49692 → nixos/google-network-daemon: systemd job type simple instead of oneshot → https://git.io/fxhto
ThatDocsLady has joined #nixos
Theuni2 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ilikeheaps has quit [Ping timeout: 244 seconds]
<{^_^}> [nixpkgs] @dotlambda pushed to master « home-assistant: 0.81.2 -> 0.81.5 »: https://git.io/fxhth
<lukego> How can a NixOS module declare that a certain additional kernel patch should be applied, such that it plays nice with other modules and nixos config done elsewhere?
<WilliamHamilton[> a question about haskell packaging: I want to patch a package in `pkgs/development/haskell-modules/hackage-packages.nix` so that it uses the latest source in a github repo instead of the hackage version. What would be the best way to do that?
<clever> WilliamHamilton[: you can first try callHackage, if the version happens to be in the db
<clever> WilliamHamilton[: basement = self.callHackage "basement" "0.0.8" {};
<lukego> clever: Is there an example where a NixOS module adds an entry to boot.kernelPatches in such a way that it won't collide with other setup somehow?
<clever> lukego: that option is a list type, so it will merge with all other values automatically
<{^_^}> [nixpkgs] @dotlambda merged pull request #49645 → pythonPackages.broadlink: init at 0.9 → https://git.io/fxxA1
<{^_^}> [nixpkgs] @dotlambda pushed 3 commits to master: https://git.io/fxhqZ
<lukego> clever: Thanks for clearing that up!
Mr_Keyser_Soze99 has joined #nixos
<WilliamHamilton[> clever: duly noted, thanks, I'm sure I'll want this at some point; what should I do instead if the source is not on hackage, but on a git repo? Is there an equivalent pattern?
<clever> WilliamHamilton[: either run cabal2nix on that repo (either url or ./.) and then haskellPackages.callPackage it, or use `haskellPackages.callCabal2nix "name" (pkgs.fetchFromGithub { ... }) {}`
periklis has quit [Ping timeout: 260 seconds]
<{^_^}> [nixpkgs] @dotlambda merged pull request #47988 → pythonPackages.fastpbkdf2: init at 0.2 → https://git.io/fxsmi
<{^_^}> [nixpkgs] @dotlambda pushed commit from @jqueiroz to master « pythonPackages.fastpbkdf2: init at 0.2 (#47988) »: https://git.io/fxhqE
astrofog has joined #nixos
orivej has joined #nixos
zolk3ri has quit [Quit: Lost terminal]
ThatDocsLady has quit [Ping timeout: 252 seconds]
<steveeJ> hyperfekt: if you changed a package with maintainers listed they'll be pinged on github
endformationage has joined #nixos
justanotheruser has quit [Ping timeout: 264 seconds]
<WilliamHamilton[> clever: thanks; follow up: https://pastebin.com/kKPe6WC5 is the Agda definition in the haskell packages. Can I modify this structure in the way you indicated (callCabal2nix "name" (fetchFromGithub ...)) keeping the `postinstall` part? In other words, how does this mix with the callPackage and mkDerivation functions there?
<clever> WilliamHamilton[: cabal2nix will return a nix file containing mkDerivation
<clever> WilliamHamilton[: callPackage and callCabal2nix will then import that with the right args
alex```` has joined #nixos
<{^_^}> [nixpkgs] @dtzWill merged pull request #49665 → tootle: 0.1.5 -> 0.2.0 → https://git.io/fxpdi
<{^_^}> [nixpkgs] @dtzWill pushed to master « tootle: 0.1.5 -> 0.2.0 »: https://git.io/fxhmS
duncan^ has joined #nixos
mayhewluke has quit [Ping timeout: 244 seconds]
alex```` has quit [Ping timeout: 246 seconds]
mayhewluke has joined #nixos
<duncan^> Mic92: Beagleboard can run mainline kernel, why do you have linux-beaglebone?
<duncan^> (X-post)
iyzsong has quit [Ping timeout: 252 seconds]
<duncan^> IIRC it's a matter of building DTBs..
NinjaTrappeur has quit [Quit: WeeChat 2.2]
<{^_^}> [nixpkgs] @flokli merged pull request #49650 → amazon-image: fix typo in comment → https://git.io/fxpcB
<{^_^}> [nixpkgs] @flokli pushed 2 commits to master: https://git.io/fxhY3
<{^_^}> Channel nixpkgs-unstable advanced to https://github.com/NixOS/nixpkgs/commit/7d24cf8ea99 (from 3 days ago, history: https://channels.nix.gsc.io/nixpkgs-unstable)
erictapen has joined #nixos
NinjaTrappeur has joined #nixos
<{^_^}> [nixpkgs] @flokli pushed commit from @srghma to release-18.09 « fix typo »: https://git.io/fxhYW
crmlt has joined #nixos
crmlt has left #nixos [#nixos]
astrofog has quit [Remote host closed the connection]
<{^_^}> [nixpkgs] @c0bw3b merged pull request #48082 → ansible: Update to latest supported releases → https://git.io/fxZ1y
<{^_^}> [nixpkgs] @c0bw3b pushed 5 commits to master: https://git.io/fxhYH
<nh2> arianvp: I really cannot help but think that some things in systemd are just bugged. Randomly, my timer no longer starts the service. It says the timer is `active (elapsed)`, but PASSED is now `n/a` and it hasn't started the cdn-rsync.service for the last 15 hours.
<clever> nh2: can you pastebin the output of `systemctl list-timers` ?
Mr_Keyser_Soze99 has quit [Ping timeout: 268 seconds]
<{^_^}> [nixpkgs] @zimbatm merged pull request #49692 → nixos/google-network-daemon: systemd job type simple instead of oneshot → https://git.io/fxhto
<{^_^}> [nixpkgs] @zimbatm pushed commit from @srhb to master « nixos/google-network-daemon: systemd job type simple instead of oneshot (#49692) »: https://git.io/fxhYx
fragamus has joined #nixos
<clever> nh2: i usually see a OnCalendar= in the timer files, so this one likely behaves differently then what i'm used to
<nh2> clever: the odd thing is that this timer worked fine before. arianvp and me did some debugging on switch-to-configuration not correctly bringing up the new ExecStart; I wrote a script to automate a nixops deploy for it. It ran fine for the first 4 hours. But now for the last 15 hours, the timer doesn't start the service any more
<clever> nh2: might it be related to the rsync failing due to an error?
<clever> systemd can be configured to not restart things when they fail due to an error, rather then exiting with 0
<nh2> unlikely, the 400 times the script ran before, each time the rsync failed as well
juhe has joined #nixos
<{^_^}> [nixpkgs] @edwtjo pushed to master « i2pd: 2.21.0 -> 2.21.1 »: https://git.io/fxhOE
<nh2> clever: looks like this bug: https://github.com/systemd/systemd/issues/6680
<{^_^}> systemd/systemd#6680 (by utezduyar, 1 year ago, open): .timer doesn't fire
<Mic92> duncan^: good point. Do you have sources for that?
<Mic92> we might remove this then all together.
<duncan^> I think it should be a matter of just using the defconfig and then run make dtbs
fragamus has quit [Ping timeout: 240 seconds]
<Mic92> rycee: to bad, I was hoping I would be able to build my secrets prototype without depending on a nixpkgs fork.
lukego_ has joined #nixos
<Mic92> duncan^: the only problem that still remains is that we would need someone to test this kernel.
<Mic92> I guess I will open a pr that removes linux-beaglebone and ask if someone wants else can to maintain a mainline version.
<duncan^> is there a generic arm mainline kernel for nixos
ddellacosta has joined #nixos
<duncan^> in that case (I assume) you could append the omap2plus defconfig to it or the equivalent
<{^_^}> [nixpkgs] @c0bw3b merged pull request #49656 → folly: 2018.08.27.00 -> 2018.10.29.00 → https://git.io/fxpaA
<{^_^}> [nixpkgs] @c0bw3b pushed 2 commits to master: https://git.io/fxhOd
vidbina_ has joined #nixos
<Mic92> duncan^: you might want to ask for access to https://github.com/nix-community/aarch64-build-box if you want to build something larger
vidbina has quit [Ping timeout: 252 seconds]
<{^_^}> [nixpkgs] @periklis opened pull request #49694 → alacritty: fix build on darwin → https://git.io/fxh3v
vidbina_ has quit [Ping timeout: 272 seconds]
betaboon has quit [Quit: WeeChat 2.2]
lukego_ has quit [Ping timeout: 256 seconds]
<{^_^}> [nixpkgs] @svenklemm opened pull request #49695 → timescaledb: (0.11.0 -> 1.0.0) → https://git.io/fxh33
<{^_^}> [nixpkgs] @c0bw3b merged pull request #49546 → shutter: 0.94 -> 0.94.2 → https://git.io/fxb9D
<{^_^}> [nixpkgs] @c0bw3b pushed commit from @fadenb to master « shutter: 0.94 -> 0.94.2 (#49546) »: https://git.io/fxh3S
cfricke has joined #nixos
lukego_ has joined #nixos
<{^_^}> [nixpkgs] @c0bw3b opened pull request #49697 → [18.09] ansible: 2.5.2 -> 2.5.11 and 2.6.2 -> 2.6.7 → https://git.io/fxhsU
lukego_ has left #nixos [#nixos]
lukego_ has joined #nixos
acarrico has joined #nixos
<lukego_> I added 'boot.kernelPatches' to my NixOS module but I'm not convinced that it's actually been applied. Is there an easy way to check? e.g. to see the build dependencies of the currently running kernel?
Itkovian has joined #nixos
Ariakenom has quit [Read error: Connection reset by peer]
<{^_^}> [nixpkgs] @vcunat pushed 3 commits to release-18.09: https://git.io/fxhsY
<symphorien> if you run nix show-derivation /nix/store/thederivationofthekernel.drv you should see your patch, I imagine
jtojnar has quit [Remote host closed the connection]
<lukego_> Thanks. How do I identify the derivation for the running kernel?
stepcut has quit []
<symphorien> looking at the output of nixos-rebuild ?
Ariakenom has joined #nixos
alex```` has joined #nixos
<nh2> clever arianvp: OK it looks like timers are totally bugged in systemd: https://github.com/systemd/systemd/issues/6680#issuecomment-435597174
hiroshi- has joined #nixos
agander has quit [Ping timeout: 240 seconds]
<{^_^}> [nixpkgs] @xeji merged pull request #49655 → lightdm-mini-greeter: 0.3.3 -> 0.3.4 → https://git.io/fxpaa
<{^_^}> [nixpkgs] @xeji pushed commit from @prikhi to master « lightdm-mini-greeter: 0.3.3 -> 0.3.4 (#49655) »: https://git.io/fxhsH
lukego_ has quit [Ping timeout: 256 seconds]
acarrico has quit [Ping timeout: 244 seconds]
hiroshi has quit [Ping timeout: 245 seconds]
hiroshi- is now known as hiroshi
<{^_^}> [nixpkgs] @xeji merged pull request #49654 → yoda: 1.7.1 -> 1.7.3 → https://git.io/fxpu5
<{^_^}> [nixpkgs] @xeji pushed commit from @veprbl to master « yoda: 1.7.1 -> 1.7.3 (#49654) »: https://git.io/fxhsj
<{^_^}> [nixpkgs] @c0bw3b merged pull request #49697 → [18.09] ansible: 2.5.2 -> 2.5.11 and 2.6.2 -> 2.6.7 → https://git.io/fxhsU
<{^_^}> [nixpkgs] @c0bw3b pushed 3 commits to release-18.09: https://git.io/fxhGU
<{^_^}> [nixpkgs] @unode opened pull request #49698 → ghc-8.2.1-binary: Patch shebang causing failures in Centos 6 → https://git.io/fxhGt
drakonis_ has quit [Ping timeout: 252 seconds]
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
drakonis_ has joined #nixos
<{^_^}> Channel nixos-18.09-small advanced to https://github.com/NixOS/nixpkgs/commit/0f4978267a0 (from 2 hours ago, history: https://channels.nix.gsc.io/nixos-18.09-small)
<gchristensen> something has broken my audio independent of changes to nixos. I've rolled back to a copy from a month ago and it is still broken.
<gchristensen> speakers work, but when I plugged in my headphones, no sound comes out. the speakers auto-mute and my headphones are detected. the equalizer is flat.
<gchristensen> any ... uh... ideas?
<hyper_ch> pulseaudio issue? check the devices settings and order maybe
fendor has quit [Quit: Leaving]
<gchristensen> any nice gui tools for that? I usually use pavucontrol
<hyper_ch> kde has it in systemsettings
<hyper_ch> I think
<{^_^}> [nixpkgs] @edwtjo pushed to master « x11basic: init at 1.26 »: https://git.io/fxhZG
<hyper_ch> I think rss2email is broken
kp__ has joined #nixos
<gchristensen> hyper_ch: do you have kde?
<hyper_ch> gchristensen: I do
<gchristensen> what is the cli program to start systemsettings?
<hyper_ch> no idea :)
<gchristensen> :) anything obvious starting with ktabtab?:)
<hyper_ch> systemsettings5
<gchristensen> nice, thank you!
<hyper_ch> good luck
<gchristensen> The program ‘systemsettings5’ is currently not installed. You can install it by typing: nix-env -iA nixos.plasma5.systemsettings .... nix-shell -p nixos.plasma5.systemsettings -> error: value is a function while a set was expected, at (string):1:94 :')
<clever> gchristensen: try flipping each of the mute flags in pavucontrol?
<clever> gchristensen: nix-env takes channel names (nixos.) but nix-shell doesnt
<gchristensen> oh yeah, haha, blinded by ... something
<clever> > nixos
<{^_^}> { config = <CODE>; options = <CODE>; pkgs = <CODE>; system = <CODE>; vm = <CODE>; vmWithBootLoader = <CODE>; }
<clever> so you refered to this function, i'm surprised to even see that in the pkgs tree
<gchristensen> it was added a bit ago by roberth I think
hyper_ch has quit [Quit: ZNC 1.7.1 - https://znc.in]
<gchristensen> ok so now the headphones are just clicking at me, haha.
<clever> gchristensen: also try alsamixer, use f6 i think to change cards
<gchristensen> adjusting the volume of Built-In Audio Analog Stereo doesn't change the clicking sound, but muting it makes it stop
hyper_ch has joined #nixos
<roberth> gchristensen: correct. Didn't expect that. Can I blame nix-env?
<gchristensen> mmm...not sure
<gchristensen> :)
<hyper_ch> gchristensen: got the sound issue solved?
<{^_^}> [nixpkgs] @eadwu opened pull request #49699 → glava: 1.5.5 -> 1.5.8 → https://git.io/fxhnT
<clever> roberth: i just use import (pkgs.path + "/nixos") { configuration = ...; }
<clever> though that ignores overrides+overlays
<roberth> yeah the point of it is to streamline the use case where you want to keep the same pkgs and use it for NixOS
<gchristensen> hyper_ch: the headphones are making this crazy clicking sound. sounds like lip smacks.
agander has joined #nixos
<hyper_ch> either replace ears or headphones :)
<roberth> clever: it should also evaluate a bit quicker because you don't call nixpkgs twice or more
<clever> roberth: yeah
<gchristensen> they work now!
<{^_^}> [nixpkgs] @globin merged pull request #49682 → [18.09] gitlab fixes → https://git.io/fxhUg
<{^_^}> [nixpkgs] @globin pushed 2 commits to release-18.09: https://git.io/fxhnO
<clever> i think {^_^} also auto-called pkgs.nixos with {} before printing it
<gchristensen> muting everything, unmuting everything, then turning volmes up & down seems to have fixed it. crazy :)
<hyper_ch> it's a cheat code :)
<clever> because it dis-agrees with the source
<gchristensen> thank you hyper_ch, clever!
<hyper_ch> like iddqd and idkfa
<roberth> clever: it certainly auto-called
<clever> > ({ a ? 42 }: a)
<{^_^}> <LAMBDA>
<roberth> > { a ? 42, ... }: a
<{^_^}> <LAMBDA>
<roberth> > { a ? 42 }: { inherit a; }
<{^_^}> <LAMBDA>
Thra11 has quit [Ping timeout: 252 seconds]
<clever> > let foo = ({ a ? 42 }: a); in foo
<{^_^}> <LAMBDA>
<hyperfekt> clever: btw, it's a thing now (or rather as soon as it is merged, #49671). thank you again for all your help :D
<{^_^}> https://github.com/NixOS/nixpkgs/pull/49671 (by hyperfekt, 11 hours ago, open): bcachefs root support
<roberth> > lib.attrsets
<{^_^}> { attrByPath = <CODE>; attrNames = <PRIMOP>; attrVals = <CODE>; attrValues = <PRIMOP>; catAttrs = <PRIMOP>; chooseDevOutputs = <CODE>; collect = <CODE>; filterAttrs = <CODE>; filterAttrsRecursive = <C...
<roberth> oh well duh. I'm giving up and accepting {^_^} behavior
<clever> hyperfekt: i prefer zfs, but always nice to see more choices available in nixos
Izorkin has quit [Quit: ZNC 1.7.1 - https://znc.in]
elgoosy has joined #nixos
<clever> hyperfekt: boot.kernelPackages = pkgs.linuxPackages_testing_bcachefs;
<clever> hyperfekt: this line will likely cause problems
<clever> hyperfekt: its best to use a kernel package from config.boot.kernelPackages, which has to be added to the right region of all-packages.nix, or it will wind up being of the wrong kernel version
<clever> everything else looks great
<gchristensen> the : stuff feels weird I think
<clever> hyperfekt: oh wait, no, misread that nvm
<clever> hyperfekt: thats changing out the entire kernel package set, and the kernel version
<clever> hyperfekt: yeah, i see it now, its using a fork of the linux kernel, with the BCACHEFS_FS flag enabled
<infinisil> > :v nixos
hotfuzz has quit [Read error: Connection reset by peer]
<{^_^}> nixos = import <nixpkgs/nixos> { configuration = nixosConfigBase; }
<infinisil> clever: roberth: ^^
<hyperfekt> clever: oh, what a relief, i was already looking into how i was gonna get the fork into the normal kernel packages.
<clever> infinisil: neat
<infinisil> > :v nixosConfigBase
<{^_^}> nixosConfigBase = { boot.loader.grub.device = "nodev"; fileSystems."/".device = "x"; imports = [ nixosConfig ]; }
<infinisil> > :v nixosConfig
<{^_^}> nixosConfig = { fileSystems."/".options = [ "test" ]; }
kp__ has quit [Ping timeout: 272 seconds]
<infinisil> You can reassign nixosConfig to test some config, the `nixos` value will use the updated value
hotfuzz has joined #nixos
<hyperfekt> gchristensen: do you mean it's existence at all, or the way it is handled in nixos? for the former, i don't see a way around it beyond maintaining a fork of bcachefs :/
<gchristensen> huh?
<clever> hyperfekt: if the source was kept as an external module, it could be built on any kernel version
<clever> hyperfekt: you would need a makefile and kconfig like in here, to begin with: https://github.com/cleverca22/v3d2
<clever> and any additional files would be in the same dir
<clever> but if its modifying linux internals, it becomes harder to maintain it as an external module
<gchristensen> specifically this : stuff :) https://github.com/NixOS/nixpkgs/pull/49671#discussion_r230562083
<clever> ah
agander has quit [Ping timeout: 268 seconds]
<rnhmjoj[m]> do you know why shells spawned by ranger have python, file and setuptools in the PATH? it's a bit of an issue because it shadows my python installation with the library i need.
<averell> if i add a beta version of an existing package, what do i put in the maintainers field?
<gchristensen> averell: presumably you :)
<averell> only? and does that mean i have to get myself added?
<hyperfekt> gchristensen: I see. The alternative would be reworking stage-1-init.sh to receive a list of devices separately in addition to the list of mounts.
drakonis has joined #nixos
<nh2> clever: how did it end up with the .xz nix signature problem at nixcon?
<hyperfekt> clever: I'm not really experienced at all with kernel development, there's definitely a lot of changes in the kernel as far as I have read. What of that qualifies as the 'internals' you refer to, I'm unsure.
drakonis_ has quit [Ping timeout: 252 seconds]
<clever> nh2: i narrowed the bisect down to a certain region of commits, then entirely forgot about it! lol
<nh2> clever: I think you should first file the issue so that we can coordinate it there, I'm sure dtz also wants to read about it
<clever> nh2: sure
kp__ has joined #nixos
<roberth> infinisil: that clears things up, thanks!
<{^_^}> nix#2512 (by cleverca22, 7 seconds ago, open): <nix/fetchurl.nix> auto-uncompresses xz files
<nh2> clever: thanks! Do you know from which issue/link this started, so we can link where we found it?
<clever> nh2: i think it was when we where trying to solve gold issues with musl on ghc?
<clever> and because my nix was newer, and i lacked the binary cache, i ran into the issue
<nh2> clever: ah right, I have the issue. Do you remember which of dtz's file URLs made the problem?
<clever> nh2: but dtz's host isnt to blame, its working perfectly
<nh2> clever: sure, just so that we can give a concrete example
<clever> i already have a `nix-build` command in the issue, but you could just put that attr into a -E flag
<{^_^}> [nixpkgs] @c0bw3b merged pull request #49165 → consul: 0.9.3 -> 1.3.0 with vendored UI → https://git.io/fxShz
<{^_^}> [nixpkgs] @c0bw3b pushed 3 commits to master: https://git.io/fxhcx
<clever> and it could also be a smaller file for faster repro, it just needs to be a url ending in .xz
drakonis_ has joined #nixos
vidbina_ has joined #nixos
agander has joined #nixos
<nh2> that bisection range is somewhere upwards of nix 2.1, right?
<clever> the first good commit is on 2.1.3, as is the first bad commit
<clever> so i would expect 2.1.3 and master to be broken
ivegotasthma has joined #nixos
<srhb> globin: flokli: Hey guys. Did you manage to test migrations on the gitlab changes?
kp__ has quit [Ping timeout: 272 seconds]
<ivegotasthma> hello, I just finished setting up nixos the way I want it. I want to ask if I should use home manager to create a completely stateless system, from a config standpoint. Will I be able to remove unused ~ configuration files with home-manager?
<nh2> clever: I'll put that info in the issue, as that's probably what people will google
<clever> nh2++
<{^_^}> nh2's karma got increased to 2
<WilliamHamilton[> what's wrong with this invocation in the nix repl?
<WilliamHamilton[> ```nix-repl> haskellPackages.callCabal2nix "Agda" (fetchFromGitHub { "owner": "agda"; "repo": "agda"; "rev": "ac38171335f3cdd4be92326c1b3aace484320803"; "sha256": "1y3x78shzci6xz1n52djjy70y9bilsm9zmcqpn8cb1498cha2wrl"} ) {};```
<srhb> WilliamHamilton[: { attr = value; }
<WilliamHamilton[> I get `error: syntax error, unexpected ':', expecting '.' or '=', at (string):1:64`
<clever> WilliamHamilton[: you gave it json, not nix
<WilliamHamilton[> ok, so, no quotes?
<clever> = not :
<WilliamHamilton[> I already used semicolons
<WilliamHamilton[> wow, I'm blind, thank you guys!
<c0bw3b_> owner = "agda"; repo = "agda"; rev = "ac...";
<{^_^}> [nixpkgs] @roberth pushed to roberth-patch-1 « rabbitmq module: Update documentation after proofreading »: https://git.io/fxhCi
ddellacosta has quit [Remote host closed the connection]
<{^_^}> [nixpkgs] @roberth pushed to roberth-patch-1 « rabbitmq module: Update documentation after proofreading »: https://git.io/fxhC1
<WilliamHamilton[> also, is there a function to convert from json to nix syntax for this cases or it is commonly done manually?
<clever> WilliamHamilton[: builtins.fromJSON
<WilliamHamilton[> clever: thanks
Thra11 has joined #nixos
<WilliamHamilton[> how would I call nix-prefetch-github in the nix repl?
<clever> ,tofu WilliamHamilton[
<{^_^}> WilliamHamilton[: To get a sha256 hash of a new source, you can use the Trust On First Use model: use probably-wrong hash (for example: 0000000000000000000000000000000000000000000000000000) then replace it with the correct hash Nix expected.
<{^_^}> [nixpkgs] @averelld opened pull request #49701 → RStudio 1.2 preview init → https://git.io/fxhC5
<WilliamHamilton[> so, you're saying i should not use nix-prefetch-github in the repl because I can use 0000?
<clever> you can either use it in a normal shell, or just use the wrong hash in a nix repl
<ottidmes> if a package has a default config, should I put it in $out/share? or what is best practice in that regard? I have a matching service module for it which supplies the real config to be placed in /etc
<WilliamHamilton[> clever: thanks, I'm almost done, now if I write
<WilliamHamilton[> ` nix-shell -p "with import <nixpkgs> {}; haskellPackages.callCabal2nix "Agda" (fetchFromGitHub {"owner"= "agda"; "repo"= "agda"; "rev"= "ac38171335f3cdd4be92326c1b3aace484320803"; "sha256"= "0000000000000000000000000000000000000000000000000000"; } ) {}" `
mayhewluke has quit [Ping timeout: 268 seconds]
<WilliamHamilton[> I'm still missing something. callPackage?
<catern> you know, git and nix seem quite similar to me
mayhewluke has joined #nixos
<hyper_ch> both feature the letter "i" in the middle
<catern> in both cases people complain about the UI being "bad", but my opinion is just that the UI requires an investment of effort and after that is actually really good, and both of them are really supremely powerful
<clever> WilliamHamilton[: callCabal2nix handles the callPackage for you
<WilliamHamilton[> so why does that invocation say `error: undefined variable 'Agda' at (string):1:151`?
<clever> WilliamHamilton[: you have doublequotes inside doublequotes
<clever> nix-shell -p with import <nixpkgs> {}; haskellPackages.callCabal2nix Agda (fetchFromGitHub {owner= agda; repo= agda; rev= ac38171335f3cdd4be92326c1b3aace484320803; sha256= 0000000000000000000000000000000000000000000000000000; } ) {}
<WilliamHamilton[> right, let me fix that
<clever> WilliamHamilton[: so the command turned into this
<clever> singlequotes on the outside is the simplest solution
<elvishjerricco> How do you build emacs packages locally? In 18.03 I used melpaBuild and set `recipeFile = null`. But now I HAVE to provide a `recipe` arg, and I don't know what to put in there.
<flokli> srhb: I didn't use the gitlab module for now, as tests failed. I could give it a try, migrating from an older Debian omnibus installation (although that might not count). I think globin and fpletz have a gitlab around, where we could dogfood it with the changes :-)
<srhb> flokli: Ah. I'm asking since the changes seem to break migrations for me on master, so I'm worrying that the changes were merged _very fast_ into 18.09 and will break peoples' gitlab instances.
<srhb> I will test further as soon as I can.
<srhb> At least rollback was succesful.
Izorkin has joined #nixos
Izorkin has quit [Client Quit]
<{^_^}> [nixpkgs] @peti merged pull request #49653 → drop old ghc versions 7.10.3, 8.0.2, and 8.4.3. → https://git.io/fxpER
<{^_^}> [nixpkgs] @peti pushed 3 commits to master: https://git.io/fxhWV
<{^_^}> [nixpkgs] @worldofpeace opened pull request #49702 → lollypop: 0.9.607 -> 0.9.610 → https://git.io/fxhWD
deanman has quit [Ping timeout: 250 seconds]
<{^_^}> [nixpkgs] @FRidh opened pull request #49703 → nixos: hardware.nvidia.package option for selecting nvidia package → https://git.io/fxhW7
deanman has joined #nixos
fendor_ is now known as fendor
alex```` has quit [Ping timeout: 244 seconds]
<flokli> srhb: I was assuming globin tested the migrations before merging... If it breaks, we should probably revert 18.09 and fix mkgrations in master first
Izorkin has joined #nixos
<ivegotasthma> can nix run a different kernel than linux?
<{^_^}> [nixpkgs] @FRidh merged pull request #49128 → nvidia-x11: 410.66 -> 410.73 → https://git.io/fxyHz
<{^_^}> [nixpkgs] @FRidh pushed commit from @eadwu to master « nvidia-x11: 410.66 -> 410.73 »: https://git.io/fxhle
Izorkin has quit [Client Quit]
<clever> ivegotasthma: the darwin kernel and windows kernel are also supported, though windows has some issues still
<flokli> srhb: can you share some logs on what failed how?
Izorkin has joined #nixos
<{^_^}> [nixpkgs] @matthewbauer opened pull request #49704 → make-derivation: enable pie hardening with musl → https://git.io/fxhlk
Izorkin has quit [Client Quit]
<ivegotasthma> clever: freebsd/openbsd? I'm a bit surprised windows supper was faster than the BSDs.
<ivegotasthma> s/supper/support
<clever> ivegotasthma: ive seen traces of bsd in nixpkgs, and i think darwin is bsd-ish, but ive not heard much about them actually being used
<drakonis_> there's someone porting nix to freebsd but its not building yet
<drakonis_> so, support's not really there yet
<drakonis_> they have way too much software that demands patches to work
cmacrae has joined #nixos
<drakonis_> the bsds are all different from one another, so grouping them together is a disservice
<{^_^}> [nixpkgs] @matthewbauer pushed to master « grpc: only set -Wno-error on clang »: https://git.io/fxhlB
<{^_^}> [nixpkgs] @joachifm merged pull request #49620 → nixos/rspamd: Add support for included files → https://git.io/fxxI5
<{^_^}> [nixpkgs] @joachifm pushed 2 commits to master: https://git.io/fxhla
cnidario has joined #nixos
ottidmes has quit [Ping timeout: 252 seconds]
Izorkin has joined #nixos
Izorkin has quit [Client Quit]
<{^_^}> [nixpkgs] @matthewbauer pushed to master « release-cross: use ghc844 for tests »: https://git.io/fxh8L
Izorkin has joined #nixos
Izorkin has quit [Client Quit]
Izorkin has joined #nixos
Izorkin has quit [Client Quit]
Izorkin has joined #nixos
cfricke has quit [Quit: WeeChat 2.2]
orivej has quit [Ping timeout: 252 seconds]
Itkovian has joined #nixos
o1lo01ol1o has joined #nixos
zolk3ri has joined #nixos
<o1lo01ol1o> I have a project whose (two) binaries are produced by running a couple commands inside two shells, shell.foo and shell.bar. I would like to add another attribute to my default.nix that simply runs those two commands (each with access only to the one shell) and then copies the binaries to a new directory. How should I go about doing this?
<clever> o1lo01ol1o: does nix-build work on those derivations yet?
aleph- has joined #nixos
<o1lo01ol1o> I can nix-shell -A shell.foo and it builds
<o1lo01ol1o> is that what you mean?
<clever> o1lo01ol1o: usualy, -A provides a shell suitable for building foo, not for using foo
<{^_^}> [nixpkgs] @matthewbauer pushed to master « krb5: don't pull in keyutils with bionic »: https://git.io/fxh8Q
Ariakenom has quit [Quit: Leaving]
<aleph-> join #nixos-chat
<aleph-> Woops
Ariakenom has joined #nixos
<o1lo01ol1o> clever: right, nix-build want's to know where $src is when I tell it to build the shell.foo
juhe has quit [Remote host closed the connection]
<clever> o1lo01ol1o: you need to modify the derivation to add src = ./.; or similar
eljih has joined #nixos
ottidmes has joined #nixos
<o1lo01ol1o> clever: well, shell.foo is not a set, it's a list, so I can't put a src attribute on there
<clever> o1lo01ol1o: you need to add src to every item in the list then
<clever> o1lo01ol1o: or make a new derivation, that has that list on its buildInputs
<o1lo01ol1o> I'm leaning towards the latter
eljih has quit [Client Quit]
<o1lo01ol1o> clever: I would just use mkDerivation with src = ./.; the builder as my two commands in a shell script and the list of build inputs?
hlolli has quit [Ping timeout: 252 seconds]
<clever> o1lo01ol1o: you generally want to use buildPhase and installPhase, rather then builder
<clever> builder is a lot more limited
<o1lo01ol1o> k, Can I specify which build input to use during which phase? I'm not certain that wires won't get crossed if they're both active
<clever> every derivation is isolated and can only access its own buildInputs
<o1lo01ol1o> what if shell.foo has the binary foobar and shell.bar also has a different foobar, and my build command uses each?
<clever> then you use ${shell.foo}/bin/foobar and ${shell.bar}/bin/foobar in the buildCommand
<clever> or you rename things to be more sane
<o1lo01ol1o> heh, it's haskell's ghc and ghcjs shells, so there's a lot of overlap
hlolli has joined #nixos
<o1lo01ol1o> but ok, yes, I'll try that
<o1lo01ol1o> thanks for the help!
<clever> o1lo01ol1o: it would probably be easyer if you gist all of your nix files ane give a better example of what you want done
<clever> yep
<o1lo01ol1o> clever: it may come to that, but I'm also learning nix so struggling through has some of it's own benefits.
edwtjo has quit [Ping timeout: 272 seconds]
edwtjo has joined #nixos
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<clever> o1lo01ol1o: pkgs.runCommand "name" { buildInputs = [ foo ]; } "build script" is also of use
o1lo01ol1o has quit [Remote host closed the connection]
o1lo01ol1o has joined #nixos
<catern> hmmm
<catern> does anyone know of a good, standard way for a C library to expose helper binaries to programs using the library?
<catern> I'd put the binaries into libexec, I guess, but it looks like the standard is to put them into a nested directory, not sure, anyone know?
<{^_^}> [nixpkgs] @Mic92 merged pull request #49694 → alacritty: fix build on darwin → https://git.io/fxh3v
<{^_^}> [nixpkgs] @Mic92 pushed 2 commits to master: https://git.io/fxh4X
crmlt has joined #nixos
crmlt has left #nixos [#nixos]
zack_moe has joined #nixos
zack_moe_ has quit [Ping timeout: 246 seconds]
xorkle has joined #nixos
<{^_^}> [nixpkgs] @Mic92 opened pull request #49707 → linux_beagleboard: remove → https://git.io/fxhBE
ddellacosta has joined #nixos
agander has quit [Ping timeout: 268 seconds]
FRidh has quit [Quit: Konversation terminated!]
jD91mZM2 has quit [Quit: WeeChat 2.2]
<WilliamHamilton[> clever in the end, I used https://pastebin.com/4Qg7fub7 as my command, but I get a failure of the test suite even if I used dontCheck. Do you happen to know the reason?
<clever> WilliamHamilton[: that is a list containing the function and a derivation
<clever> not the result of applying the function to the derivation
<clever> lists in nix are greedy and need () to force the order correctly
alex```` has joined #nixos
<WilliamHamilton[> clever: :o let me try with parentheses. But then, why didn't it report an error? What's the meaning of the expression if `pkgs.haskell.lib.dontCheck` is seen as an element of the list?
<clever> WilliamHamilton[: there is another function, that filters all haskell dependencies, to exclude non-haskell things
<clever> it winds up excluding functions as well, causing it to not fail
<WilliamHamilton[> and this function that removes functions from the list is called by ghcWithPackages?
<clever> yep
<WilliamHamilton[> clever: wow, I would never have guessed!
<clever> same, i tracked it down a few months ago when somebody else made the same mistake, and i wanted to know how it didnt fail
o1lo01ol1o has quit [Remote host closed the connection]
<WilliamHamilton[> clever: and is this a pattern that nixpkgs uses often (filtering what is not expected to take only the elements with the right type) in your experience?
<clever> only haskell does it, as far as ive seen
jtojnar has joined #nixos
<{^_^}> [nixpkgs] @jtojnar merged pull request #49702 → lollypop: 0.9.607 -> 0.9.610 → https://git.io/fxhWD
<{^_^}> [nixpkgs] @jtojnar pushed 2 commits to master: https://git.io/fxhRg
cmacrae has quit [Ping timeout: 252 seconds]
hyperfekt has quit [Ping timeout: 256 seconds]
Itkovian has joined #nixos
ddellacosta has quit [Remote host closed the connection]
fragamus has joined #nixos
<{^_^}> [nixpkgs] @periklis opened pull request #49708 → nix-index: fix darwin build → https://git.io/fxhRS
jtojnar_ has joined #nixos
greymalkin has quit [Remote host closed the connection]
greymalkin has joined #nixos
jtojnar has quit [Ping timeout: 244 seconds]
jtojnar_ is now known as jtojnar
amfl has quit [Read error: Connection timed out]
amfl has joined #nixos
fragamus has quit [Remote host closed the connection]
nschoe has joined #nixos
<nschoe> Hi :) Anybody managed to install nixOS on the rPi 3 recently?
<nschoe> I'm following https://nixos.wiki/wiki/NixOS_on_ARM#NixOS_installation_.26_configuration, but I have issues with the serial console
<clever> nschoe: you may need to disable bluetooth to make the serial work right
<clever> thats done in config.txt with one of the dtb overlays
<nschoe> clever, thanks, I'll try that.
<WilliamHamilton[> clever, the build just finished without errors now, I'm very pleased by the power of nix. I have some other doubts regarding the final touches
<nschoe> clever, quick question, though (because it's not clear in the tutorial), right now I have dd the *.img file. When I take a look at the configuration.nix on the sdcard, i'ts just very basic: just a single `import` and that's all. When they explain to use a given template for configuration.nix, they mean after we managed to log into it, right?
<clever> nschoe: import or imports? and what path is it importing?
<WilliamHamilton[> if I do `nix-build -i -p '...'` that `-i` option will install the package in my channel so that I can call just `agda` in the command line? And because cabal2nix has the post-installation patch for agda I will also have the emacs mode installed automatically?
<clever> WilliamHamilton[: nix-build doesnt have -i or -p flags
joncfoo has quit [Ping timeout: 252 seconds]
<clever> WilliamHamilton[: but you can do something like `nix-env -f default.nix -iA foobar` to install something into your profile, it will get merged into ~/.nix-profile/
joncfoo has joined #nixos
<nschoe> clever, imports = [ <nixpkgs/nixos/modules/installer/cd-dvd/sd-image-aarch64.nix> ];
<clever> nschoe: that will merge everything defined in that file with the other things you add to configuration.nix
<{^_^}> [nixpkgs] @Mic92 opened pull request #49710 → linux-samus: remove → https://git.io/fxhEs
<nschoe> clever, yes. But https://nixos.wiki/wiki/NixOS_on_ARM#NixOS_installation_.26_configuration says "Note: the default configuration.nix will contain something like imports = [ <nixos/modules/installer/cd-dvd/sd-image-armv7l-multiplatform.nix> ]; do not include that in your final installation or you will experience interesting problems. It is only for building the installation image!"
v0|d has quit [Read error: Connection reset by peer]
<clever> nschoe: ah, then the default config is likely wrong
<nschoe> I'm not sure what they mean by "building the installation image"
<WilliamHamilton[> clever: just to be sure, in the case of agda doing `nix-env -i` will install also the emacs mode?
<clever> WilliamHamilton[: if emacs is searching the right areas of ~/.nix-profile/, and the package had those files in its $out, yes
<clever> nschoe: you will need the right entries under boot.loader and fileSystems."/" and "/boot" to make sure it still boots, those will depend on how exactly the image is built
GiGa has quit [Quit: Leaving]
<nschoe> clever, okay, so that means right now the image is not built, right?
hlolli_ has joined #nixos
<clever> nschoe: if you found that inside a .img, then it is built, but not very well
<nschoe> clever, it's the stock result of dd-ing the .img on the sd card
fendor_ has joined #nixos
<clever> yeah, it sounds like its just poorly written
hlolli has quit [Ping timeout: 252 seconds]
fendor has quit [Ping timeout: 246 seconds]
<nschoe> clever, okay. Well I have to go now, but thanks. I'll try tomorrow with my head cleared.
<nschoe> I have been tryoing too hard for too long ^^
nschoe has quit [Quit: Program. Terminated.]
<{^_^}> [nixpkgs] @Mic92 merged pull request #49398 → Implement rfc0035: default `name` from `pname` → https://git.io/fx7Ia
<{^_^}> [nixpkgs] @Mic92 pushed 7 commits to master: https://git.io/fxhEr
fendor_ has quit [Ping timeout: 268 seconds]
{^_^} has quit [Ping timeout: 250 seconds]
{^_^} has joined #nixos
<ottidmes> I am trying to use multiple srcs but I get the following error: cp: cannot create directory 'source/mrxfafc0z0g64zmzni1fgbidbb2109i5-source': Permission denied
<ottidmes> do not know how to unpack source archive /nix/store/mrxfafc0z0g64zmzni1fgbidbb2109i5-source
vk3wtf has joined #nixos
{^_^} has quit [Ping timeout: 250 seconds]
{^_^} has joined #nixos
<ottidmes> nvm, I have to approach this package differently anyway
cnidario has quit [Remote host closed the connection]
mayhewluke has quit [Ping timeout: 252 seconds]
mayhewluke has joined #nixos
cnidario has joined #nixos
<{^_^}> [nixpkgs] @plchldr opened pull request #49712 → expand nspawn unit file validation logic to allow all currently supported fields → https://git.io/fxhuD
jperras has joined #nixos
cnidario has quit [Remote host closed the connection]
<globin> flokli, srhb: i updated our production instance without issues
<globin> srhb: do you have a stacktrace?
ivegotasthma has quit [Ping timeout: 272 seconds]
ivegotasthma has joined #nixos
ivegotasthma has quit [Client Quit]
zack_moe has quit [Ping timeout: 246 seconds]
stepcut has joined #nixos
<stepcut> Is there someway to override nixpkgs.lib.inNixShell? You can see what I am trying to do here -- but it is not working -- it seems to ignore my attempts to update it, https://pastebin.com/C3hbr1Xd
endformationage has quit [Ping timeout: 272 seconds]
<Zajcev_> guys, I'd like to modify pg_hba.conf and I guess I can do that with services.postgresql.authentication and "lib.mkForce", but how? thanks
sigmundv has joined #nixos
Thra11 has quit [Ping timeout: 240 seconds]
sigmundv has quit [Ping timeout: 240 seconds]
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
acarrico has joined #nixos
Dedalo has joined #nixos
Akii has quit [Ping timeout: 244 seconds]
vidbina_ has quit [Ping timeout: 244 seconds]
fresheyeball has joined #nixos
<fresheyeball> hey out there
vidbina_ has joined #nixos
obadz has quit [Ping timeout: 240 seconds]
obadz has joined #nixos
<fresheyeball> I am getting this running nix-build for a haskell project
<fresheyeball> only it builds just fine with cabal build inside the nix-shell
alex```` has quit [Ping timeout: 252 seconds]
<roberth> stepcut: stuff inside nixpkgs still references the old lib. I recommend against inNixShell. If A depends on B and B uses inNixShell for magic, what happens when you open a shell for A? You get the shell variation of B, but B is just a dependency for which you don't want the inNixShell == true variation
<roberth> just put a shell.nix in A and B
__monty__ has quit [Quit: leaving]
<stepcut> roberth: well - the inNixShell is not in my code -- or things would be a lot easier. And, indeed, I am running into a situation where I want the non-inNixShell variation -- hence my desire to find some way to trick it
<roberth> fresheyeball: seems like an instance ofhttps://github.com/NixOS/nixpkgs/issues/41340
<{^_^}> #41340 (by nh2, 22 weeks ago, open): GCC fails because argument list created by nix is too long
<stepcut> the other option would be if I could somehow set IN_NIX_SHELL="" before doing the import :-/
<clever> stepcut: i try to avoid using inNixShell, it just makes things behave in unpredictable ways
<clever> just do the right thing from shell.nix to begin with, dont try to do magic things that guess what you want
<stepcut> clever: I never use it myself - but I don't control the ouput of cabal2nix
<clever> you can ignore the default.nix cabal2nix generates, and only use the main file it produces
<clever> just load it with haskellPackages.callPackage
<stepcut> I want to use the generated `cabal2nix . --shell shell.nix` -- but I also want to call overrideCabal on the derivation it produces
<clever> skip --shell and just write a better file
<fresheyeball> roberth: is there a way around?
<stepcut> clever: that sounds horrible -- this is not something I have to do just once
<clever> stepcut: its a 2 lines, the example --shell makes is far more complex then it has to be
<clever> stepcut: and you dont even really need to run cabal2nix, because of callCabal2nix: https://github.com/input-output-hk/cardano-chain/blob/devops-1118/pkgs.nix#L11-L15
<stepcut> hmm
<fresheyeball> ottidmes: yeah I tried adding ulimit -s unlimited, to patchPhase
<clever> fresheyeball: what command is generating a cmdline too long?
<roberth> fresheyeball: is your nixpkgs up to date?
aleph- has quit [Ping timeout: 264 seconds]
c0bw3b_ has left #nixos [#nixos]
<roberth> this may help, but it's a mass-rebuild https://github.com/NixOS/nixpkgs/pull/49552
<{^_^}> #49552 (by matthewbauer, 3 days ago, open): setup.sh: avoid running the same hook twice
<clever> fresheyeball: how many deps does this package have?
<fresheyeball> clever: not many frankly
<ottidmes> I am not sure how to packge the bitwarden frontend. I have npm setup in my NixOS, so I can use it the non-Nix way, then building is a matter of 'npm install; npm run dist' and the frontend is build, but how do I handle packinging something like that. Are there any advanced node packages examples that do things like patching and building webapps?
<clever> ottidmes: yarn2nix is what you want, i think the recent terraform PR's have an example
<fresheyeball> clever: https://pastebin.com/D6xCS3ki
<clever> fresheyeball: thats over 50, lol
<fresheyeball> clever: doesn't seem like that many too me
<clever> fresheyeball: some of angerman's recent changes might fix it, but i'm not sure what branch they are on
<fresheyeball> at work it's waaaaay bigger
<{^_^}> [nixpkgs] @globin pushed to master « gitlab module: fix config handling »: https://git.io/fxh2T
drk_lrd_877_ has quit [Remote host closed the connection]
<ottidmes> clever: do you have a link, I cannot find any reference to terraform that mentions node/npm/yarn?
<fresheyeball> clever: what is this about angerman?
<clever> fresheyeball: he has been fixing some ghc problems lately, including things like commandline too long
<{^_^}> [nixpkgs] @globin pushed to release-18.09 « gitlab module: fix config handling »: https://git.io/fxh2t
<fresheyeball> 153d32cd9fac7e885979426b0e86b560a661a1ac <- this is my nixpkgs
vidbina_ has quit [Ping timeout: 268 seconds]
<clever> ottidmes: cant find it on nixpkgs, but i can pull up the other example i have
<fresheyeball> clever: has it been merged? It would be lovely to simply update my hash and having everything link again
<clever> fresheyeball: not sure, he has a PR open with a large number of improvements
<ottidmes> clever++
<{^_^}> clever's karma got increased to 40
<ottidmes> clever: that example is exactly what I was looking for, thanks!
ddellacosta has joined #nixos
<ottidmes> git submodules, I need to use fetchgit rather than fetchFromGithub for that, right?
<clever> ottidmes: fetchFromGithub "supports" submodules now
justanotheruser has joined #nixos
<clever> ottidmes: if you pass fetchSubmodules=true; then it will internally switch to pkgs.fetchgit for you
<clever> while keeping the same API
<ottidmes> clever: ok, I just read that too on the relevant issue, but that means what seemed like a temp fix, is still what is being used
<fresheyeball> clever++
<{^_^}> clever's karma got increased to 41
<fresheyeball> fresheyeball--
<fresheyeball> ah, I can't hurt me
<ottidmes> haha, I was wondering if that would work, but did not dare to try it out :P
<stepcut> clever: this seems to do the trick, (import <nixpkgs> {}).haskell.packages.ghcjs84.developPackage {root = ./.; }
<fresheyeball> ottidmes: can you hurt me?
<ottidmes> fresheyeball--
<ottidmes> no, I cannot :)
<fresheyeball> I feel like I should be able to hurt me. This karma system doesn't fill my use-case. Much anger
<ottidmes> why so negative? :P Personally I think it matches this channel, people come here with problems and their frustrations, and (mostly) end up with their problems solved and frustrations gone ;)
thc202 has quit [Ping timeout: 240 seconds]
<ddellacosta> so, I failed to get my wireless set up when I first installed NixOS on my laptop, so I'm only now trying to get my wireless network working, but I'm in a bit of a catch-22 situation where running `nixos-rebuild switch` requires network access so...not sure what to do. Any suggestions? Surely others have encountered this before?
<fresheyeball> ottidmes: I was just curious and in a silly mood
<fresheyeball> not serious
<clever> ddellacosta: got an ethernet cord?
<Zajcev_> guys, how to make pam to look another location for module?
<Zajcev_> Nov 04 00:36:07 nixosV vsftpd[23936]: PAM unable to dlopen(/nix/store/4fzqlzw0dyzvn36q5vlkika2v8jrjnhh-linux-pam-1.3.0/lib/security/pam_pgsql.so): /nix/store/4fzqlzw0dyzvn36q5vlkika2v8jrjnhh-linux-pam-1.3.0/lib/security/pam_pgsql.so: cannot open shared object file: No such file or directory
<ddellacosta> clever: this laptop requires an adapter which I don't have :-(
<clever> ddellacosta: boot the install media again, mount the existing fs to /mnt (and /mnt/boot if you have it), edit the config, and nixos-install
<clever> ddellacosta: nixos-install is just a helper that runs nixos-rebuild under a chroot for you
<Zajcev_> pam_pgsql is simply not there ...
<ddellacosta> clever: I was worried that was the only solution, kind of a pain. Seems like this could be better addressed. In any case, thanks! I do appreciate the help despite my complaining.,
<Zajcev_> /nix/store/a9ms7xygw024q0dzx58wdg05crqba302-pam_pgsql-
<clever> ddellacosta: the only other option is to mess with nix-copy and a usb stick, which is more complex
<joepie91> ,locate alsa.pc
<{^_^}> Found in packages: alsaLib.dev
<joepie91> hrm.
<ddellacosta> clever: yeah that sounds worse, haha
<clever> joepie91: just add pkgconfig and alsaLib to your inputs, and it should work
<joepie91> oh. it helps if I'm in the nix-shell...
<clever> ,locate pam_pgsql.so
<joepie91> <.<
<{^_^}> Found in packages: pam_pgsql
jperras has quit [Ping timeout: 245 seconds]
<emily> ddellacosta: nixos-rebuild should maybe warn you if you have no ethernet port and wireless isn't enabled in the configuration...
<clever> Zajcev_: its over in pam_pgsql, sounds like a bug in vsftpd's package
<emily> in the same way the password config stuff is checked
<joepie91> ,locate cmake
<Zajcev_> clever: great ...
<clever> joepie91: cmake!
<joepie91> lol
<{^_^}> Found in packages: vc, wt, uhd, glm, hpx, itk, stp, vxl, yarp, pbrt, qca2, hdf4, lean, neko, belr, sfml, cgal, ogre, glfw, dlib, poco, xbmc, root, lief, grpc, ortp, atom, mar1d, faust, fcitx, fcppt, qjson, cmake, cmark, torch, stxxl, tulip, bzrtp, falcon, xapian, bcunit, zeromq, shogun, libLAS, clblas, clingo, bloaty, llvm_4, cmocka, mongoc, spdlog, cppzmq, prison, csound, assimp, strigi, bullet, nanomsg, qca-qt5, eccodes, ucommon, and 316 more
<joepie91> .... not helpful
drakonis_ has quit [Ping timeout: 252 seconds]
<clever> joepie91: cmake is in cmake
<ddellacosta> emily: yeah that would be helpful, or else maybe it should be more obviously highlighted in the installation instructions
* joepie91 tries `cargo build` again
<joepie91> clever: thanks :P
<ddellacosta> I'll make a PR or whatever I can to the NixOS docs after I'm done and suggest that
<emily> ddellacosta: I also kinda wish the install media came with more usable wireless tools
<emily> since messing with wpa_supplicant every install sucks a bit
orivej has joined #nixos
<ddellacosta> emily: ugh yeah it's been a while since I've set up wireless on linux (years and years) and I'm not loving it
<Zajcev_> clever: is there anything what can I do?
<clever> Zajcev_: investigate how to configure it to look elsewhere for pam modules, and then fix either the package or nixos module
<ottidmes> ddellacosta: agreed, I still have to set it up on my freshly installed NixOS laptop, for the time being I just use the ethernet port
<ddellacosta> ottidmes: yeah, wish I'd thought of that. Live and learn for me I guess...
<joepie91> No package 'freetype2' found
<joepie91> eh...
<joepie91> but I have freetype in my package list...
<joepie91> buildInputs list*
<clever> joepie91: and pkgconfig in the nativeBuildInputs?
<ottidmes> fresheyeball: sorry, I have a tendancy to respond seriously to remarks, even if I know its not meant to be taken fully serious
<Zajcev_> clever: thank you
<joepie91> clever: ... I think I forgot to reinit my shell
jperras has joined #nixos
<clever> joepie91: also, if you ever installed pkgconfig in nix-env or systemPackages, remove it asap
<joepie91> yeah freetype works now lol
<clever> joepie91: it doesnt work right, and only leads to more confusion when it half works due to not being in the nix-shell
<joepie91> clever: ah, what's the specific issue with it?
<clever> pkgconfig has a setuphook, that will scan all buildInputs, and add them to the pkgconfig search path
<clever> so it magically finds all inputs
<clever> that only works under nix-shell and nix-build
<ottidmes> ddellacosta: just like the time I had to setup internet for my grandma, the offical instructions to do so were... online, long live mobile networks I guess...
<joepie91> aha
<ddellacosta> haha, that is painful
<{^_^}> [nixpkgs] @roberth opened pull request #49714 → haskellPackages: Add shell completions → https://git.io/fxhaf
<clever> if you wrongly install it with nix-env/systemPackages, then it just never finds buildInputs
<joepie91> noted, thanks
<{^_^}> Channel nixos-18.09-small advanced to https://github.com/NixOS/nixpkgs/commit/386ff58ba3f (from 7 hours ago, history: https://channels.nix.gsc.io/nixos-18.09-small)
drakonis has quit [Ping timeout: 244 seconds]
<clever> joepie91: and to make things worse, when you do forget to add it to your nix-shell inputs, it claims freetype2 not found, when it exists
<joepie91> also, I am currently in undeclared-dependency hell
<clever> joepie91: if you uninstall pkgconfig, then it will instead say pkgconfig not found, and you will then add it, and all is well
<{^_^}> [nixpkgs] @Mic92 merged pull request #49707 → linux_beagleboard: remove → https://git.io/fxhBE
<{^_^}> [nixpkgs] @Mic92 pushed 2 commits to master: https://git.io/fxhaT
jperras has quit [Ping timeout: 268 seconds]