2018-08-24

<clever> sir_guy_carleton: nix run unstable.firefox
<clever> for general development, or just to use a random package?
<clever> sir_guy_carleton: do you want to build X or use X in the shell?
<clever> though, a recent nixos may have changed that
<clever> but that only works for channels root has made
<clever> sir_guy_carleton: nix-shell '<unstable>' -A firefox
<clever> yeah
<clever> cmcdragonkai: a common mistake i see people making, is that they shove things into propagatedBuildInputs, and then wonder why it didnt magically appear in PATH
<clever> cmcdragonkai: python then bends the rules a bit, and uses this to generate $PYTHONPATH at buildtime, and then copies the build-time value of PYTHONPATH into the wrapper
<clever> but its only a compile-time thing, and has no impact on the runtime
<clever> B will behave as-if you had done buildInputs = [ A hello ];
<clever> and then package B has buildInputs = [ A ];
<clever> so if package A has propagatedBuildInputs = [ hello ];
<clever> the normal rules for propagatedBuildInputs, is that it is only in the inputs of other derivations
<clever> python bends the rules a bit
<clever> propagatedBuildInputs has nothing to do with the normal runtime deps
<clever> bash just runs things, lol
<clever> python packages already have a requirements.txt
<clever> ELF files have a clear list of every .so file they need to start
<clever> and then you dont even need a wrapper
<clever> or have a `export PATH=@finalpath@` in the script itself, and run substitute-in-place over it
<clever> then --set may be better
<clever> > lib.makeBinPath [ coreutils openssh gnupg ]
<clever> cmcdragonkai: i would do a single --prefix, using lib.makeBinPath
<clever> and chrome just deadlocks and never opens anything when you apply libredirect
<clever> and the libredirect on teamspeak then tries to apply redirections to chrome
<clever> but the env vars are inherited the whole way down
<clever> it uses xdg-open i think
<clever> should the web-browser have its path sanitized?
<clever> teamspeak for example is a gui, that can then launch the web-browser when you click links
<clever> cmcdragonkai: depends on what it is, and if it can launch child procs that are long-lasting
<clever> dont ask why, i never used the container backend
<clever> Glenn: its ssh'ing into root@localhost, and running netcat on the privateip of the container, i think

2018-08-23

<clever> wrong a name tab-completed
<clever> adamantium: nixos already sources it for you
<clever> ashkitten: if your using a single-user nix, it will be in ~/.nix-profile/
<clever> ashkitten: probably, just make sure to also build the 64bit version, and confirm that one still works, and has 64bit libs
<clever> ls -l /proc/*/fd/* | grep pipe | egrep '31629937|31629939' --color
<clever> then its piped into something else, youll need to find another process with an id + or - 1 of that
<clever> for both 1 and 2?
<clever> elvishjerricco: find the pid of xmonad, and run `ls -l /proc/PID/fd/` and then look at where 1 and 2 point
<clever> elvishjerricco: are you sure its actually starting xmonad?
<clever> elvishjerricco: did you check journalctl -f -u display-manager?
<clever> the steam FHS package deals with most of that
<clever> Drakonis[m]: yeah
<clever> tertle||eltret: nix-channel --update, as root
<clever> Drakonis[m]: some programs like skype and steam only use 32bit libs half the time
<clever> yep, so now if $system == i686-linux, copy the 32bit ones
<clever> because its using builder.sh, i dont think it will co-operate with nix-shell very well
<clever> you can also just throw an ls command into it to poke around
<clever> `set -x` will show everything builder.sh is doing
<clever> use -E and give it a full expression to import and override
<clever> --arg doesnt support overrides
<clever> ashkitten: add `set -x` to the top of builder.sh and re-build it with nix-build
<clever> you need to modify it to do the right thing if a lib32 exists
<clever> ashkitten: it just builds whatever is in the lib dir i think
<clever> Drakonis[m]: its silently providing 64bit libs in a 32bit build
<clever> also, why is that in builder.sh?, that could have just been attributes on the derivation
<clever> ashkitten: read builder.sh in the same directory
<clever> ah, then you need to modify the build process to select the 32bit libs from the combo 32+64 .run
<clever> ashkitten: the source doesnt exist, so it can only get the arch that is within the .run file
<clever> ashkitten: your if statement is flawed
<clever> ashkitten: if the version is not older then 396, it will not give 32bit libs
<clever> ashkitten: can you pastebin your diff?
<clever> ashkitten: this module should be automating it all and doing things properly
<clever> ashkitten: cant think of anything else to check
<clever> ashkitten: where are you looking to find the 64bit libs, and expect to see 32bit libs?
<clever> ashkitten: what do you have videoDrivers set to?
<clever> ashkitten: ah, that one i would expect to see working, but nixos should deal with that automatically when you have 32bit dri enabled
<clever> ashkitten: not on linuxPackages
<clever> ashkitten: that override is setting a flag that doesnt exist
<clever> ashkitten: how are you telling nix to generate a 32bit build?
<clever> same with xfce
<clever> ashkitten: are you importing nixpkgs at any point?, how are you requesting a 32bit version?
<clever> Watcher7: #nixos-on-your-router may also be of interest to you
<clever> Watcher7: yep
<clever> Watcher7: cat /mnt/etc/nixos/configuration.nix and throw it into a pastebin
<clever> Watcher7: the current working directory
<clever> ls -l
<clever> Watcher7: what does pwd and ls output?
<clever> Watcher7: check again, when you mounted vda1 onto /mnt, it hid all the old config
<clever> ij: i pass them thru builtins.toJSON
<clever> you must mount the filesystems to /mnt before you run nixos-generate-config
<clever> and grub cant find the device behind that tmpfs
<clever> your installing nixos to the tmpfs in ram
<clever> Watcher7: thats why everything is failing
<clever> Watcher7: did you mount vda1 to /mnt/ ?
<clever> drakonis__: systemd-boot lacks A LOT of the options grub has
<clever> Watcher7: and also the output of `mount`
<clever> Watcher7: and what is the exact command you ran, and its full output, when it fails?
<clever> drakonis__: depends on which driver your using
<clever> Watcher7: can you pastebin the output of lsblk ?
<clever> ah, misread your question a bit
<clever> so vda1 is ext4
<clever> the alignment for the partitions ensures that gap
<clever> but it needs a gap between the partition table and the 1st partition
<clever> it doesnt need its own boot partition
<clever> Watcher7: grub needs a partition table to install into the root of a disk
<clever> can you pastebin the nix expression?
<clever> does that return a set? a package?
<clever> it should return whatever the default.nix returns
<clever> ah, that should be good then
<clever> how many args did you pass to callPackage?
<clever> mkaito: why is it wrapped with mkDerivation?
<clever> mkaito: can you pastebin one as an example?
<clever> adamantium: the snapshot itself is basically free, but it means anything you delete with nix-collect-garbage will then not be deleted, for a week
<clever> adamantium: and then i can turn snapshots off for just /nix
<clever> adamantium: i keep nix on its own filesystem, within the same pool
<clever> but a non-root setup means that you cant do automatic cleanup, which would solve the issue chreekat was asking earlier
<clever> emily: ive not tried a multi-user non-root install
<clever> erickomoto: ah, for multi-user, then you do need the daemon running as root, and i can see why he would want nosetuid then to be paranoid
<clever> erickomoto: and if you never gain root, you cant make setuid or device nodes, so both flags are pointless
<clever> erickomoto: in general, you can just mkdir /nix ; chown erickomoto /nix, and then do the entire install without ever gaining root
<clever> but the fake uid 0 can still use ping, which is a cabability
<clever> ah, weird
<clever> sudo should be equally broken then, it uses the same system
<clever> ah
<clever> emily: what does `type ping` return?
<clever> so such things would already be broken on a normal nix install
<clever> emily: nix also has the rule that there be no setuid binaries in the store
<clever> chreekat: if you are using multi-user nix, then the nix-daemon will kill every proccess in the build user after the build has finished
<clever> and maybe unload the bluetooth BEFORE you go into standby
<clever> Judson1: see if you can fix it with unload-module and load-module via pactl instead
<clever> prior to that, i wasnt using capture in the browser enough to notice the patterns
<clever> i figured it out on the 3rd day after needing to use hangouts for daily voice chats
<clever> exactly
<clever> it claims you lack a capture device
<clever> yeah, only capture
<clever> Judson1: and you must then restart chromium
<clever> Judson1: note, that chromium looses the ability to capture any audio if pulseaudio has restarted
<clever> and a value of 0 prevents nix-env from performing a DoS attack
<clever> yeah
<clever> which would result in over 10,000 requests for .narinfo files
<clever> chreekat: and depending on your search terms, it may wind up checking every single package in all of nixpkgs
<clever> chreekat: when using something like `nix-env -qA`, it can check the binary cache to see if things are pre-built or not
<clever> have fun trying to watch a movie!
<clever> the app also does not disable automatic shutoff, so the OS will try to shut off the ipad mid-stream :P
<clever> hyper_ch: and the ipad app streams the content, so if the ipad turns off at any point, the stream breaks
<clever> hyper_ch: my cable box supports some weird form of upnp, but it only works with a specific ipad app, the android variant of the app does not work
<clever> thats nearly as many nixos users in here, lol
<clever> hyper_ch: and there are 400 experts in #plex
<clever> gchristensen: ah, didnt think they would have a room here
<clever> Izorkin: nixpkgs will then use one of those kernels to generate pkgs.linuxPackages_4_17
<clever> Izorkin: you must override the kernel in pkgs, not linuxPackages_4_17
<clever> once you register the server as being under your control, you can go back to the normal IP:32400/web, or even app.plex.tv
<clever> hyper_ch: and if you disable it?
<clever> hyper_ch: nothing obvious, any ad blocker?
<clever> Izorkin: you need to override the kernel the set was made from
<clever> Izorkin: linuxPackages_4_17 is a set, not a package, so override cant be used on it directly
<clever> i can see events like this
<clever> Aug 23, 2018 16:01:41.270 [0x7f46087fe700] DEBUG - Request: [192.168.2.15:57022 (Subnet)] GET /library/sections/2/all?type=2&includeCollections=1 (7 live) TLS Page 50-166 GZIP Signed-in Token (cleverca22@gmail.com)
<clever> hyper_ch: check the logs in here
<clever> /var/lib/plex/Plex Media Server/Logs
<clever> and it sometimes claims season 264, because the dir has h264 in the name, lol
<clever> there is no "fix match" option on a series, until i view it details, then go back to the page where that menu option previously didnt exist
<clever> ive also found the matching in plex to be rather anoying, and the UI to be somewhat glitchy
<clever> hyper_ch: anything in the journal?
<clever> hyper_ch: you may have broken the datadir on the remote end when changing the user, check journalctl
<clever> hyper_ch: id say remove the user and group also, it already has its own user&group
<clever> hyper_ch: what is your configuration.nix?
<clever> hyper_ch: open /web
<clever> hyper_ch: on the not-headless machine
<clever> yeah
<clever> disasm: the root problem, is that it listens on both, but gives 127.0.0.1 special access for the initial config
<clever> hyper_ch: once you do that ssh, http://127.0.0.1:32400, will connect to ssh, which relays it to 127.0.0.1 on the plex machine
<clever> i left datadir at the default
<clever> which re-routes the localhost 32400, to the localhost of the remote machine
<clever> hyper_ch: you must use `ssh IP -L 32400:127.0.0.1:32400` to cheat that
<clever> hyper_ch: you can only configure it via 127.0.0.1:32400/web
<clever> hyper_ch: i installed plex ~2 days ago, and had the exact same problem
<clever> simukis: ^^
<clever> Izorkin: a bug should also be fixed on nix
<clever> Izorkin: on the remote system, do `realpath /run/current-system/sw/bin/nix-store`, then on the nixops machine, add that bin dir to the start of `$PATH`, and the deploy should pass
<clever> Izorkin: you may need to downgrade the nix on the deployer side temporarily
<clever> Izorkin: command 9 was added on august 3rd, and greatly reduces memory usage
<clever> Izorkin: if you ssh into root on that machine, what does `type nix-store` return?
<clever> Izorkin: what version of nix is running on that remote machine?
<clever> thankfully, they have now made that illegal :P
<clever> you cant undo asking to cancel it :P
<clever> nope!
<clever> "ok, just cancel it in 2 weeks"
<clever> and it turns out, if you dont give a 2 week notice, you have to pay a hefty fee
<clever> ldlework: i heard of a story of somebody that wanted to cancel their cable tv service
<clever> lol
<clever> but if you test it from the outside, you will find its already working
<clever> ldlework: some routers wont allow you to access the forwarded port from inside the network
<clever> yeah
<clever> infinisil: yeah
<clever> and 80% of the time, i ask that after a user is confused about things not persisting between reboots
<clever> ldlework: ive asked people before if /boot is mounted, they claim yes, then i ask for a pastebin of `mount`, its not mounted :P
<clever> the rev you choose for kitty needs a dbus
<clever> the latest version of nixpkgs, does not supply a dbus, to the kitty package
<clever> ldlework: nixpkgs master does not give kitty dbus
<clever> ldlework: the dbus error sounds normal, the package depends on dbus now, but didnt depend on dbus before
<clever> the imports key cant depend on config
<clever> ldlework: add the kitty overlay on line 7
<clever> then just omit the .unstable
<clever> unstable = import <unstable> { overlays = [ unstableOverlay ]; };
<clever> what if you set the overlay on the unstable nixpkgs
<clever> dbus is in both files of the gist i linked
<clever> you had dbus in here when you copied it
<clever> because its a new dependency that it previously didnt need
<clever> looks like it should just work
<clever> 16968 kitty = callPackage ../applications/misc/kitty { };
<clever> anything you dont set, wont be changed
<clever> dont set buildInputs
<clever> you must use oldAttrs to get the original value and append
<clever> because overrideAttrs overrides all attrs
<clever> so you deleted the old inputs
<clever> ldlework: line 15, you replaced buildInputs, not appended
<clever> ldlework: try this one
<clever> > python3Packages.wrapPython
<clever> ldlework: ah, its called wrapPython
<clever> top-level/python-packages.nix: wrapPython = callPackage ../development/interpreters/python/wrap-python.nix {inherit python; inherit (pkgs) makeSetupHook makeWrapper; };
<clever> wrap-python.nix:} ./wrap.sh
<clever> pkgs/development/interpreters/python/wrap.sh:wrapPythonPrograms() {
<clever> ldlework: is python int he inputs?
<clever> ldlework: thats part of python
<clever> ldlework: did you add makeWrapper to the inputs?
<clever> yep
<clever> so its as simple as imports = [ ./qemu.nix ]; and the overlay is automatic
<clever> timclassic: this module adds qemu-userland options, and adds its own overlay to the cfg
<clever> timclassic: oh, what i do, is i put my overlay, inside the module
<clever> overlays cant touch the module system
<clever> timclassic: modules can only be loaded by adding them to the imports section of configuration.nix
<clever> ashkitten: that can happen if your local copy of nixpkgs is not on a rev that the cache contains
<clever> ashkitten: put it into $NIX_PATH, possibly via the nixos option nix.nixPath
<clever> > pkgs.python36Packages.Mako
<clever> Packages, not packages
<clever> pkgs.python36packages.Mako
<clever> what error?
<clever> yeah
<clever> or put it within the native inputs on 117, which is already doing the append for you
<clever> lines 107-117 of the pastebin
<clever> then you need to investigate how llvm is checking for mako, and why its failing
<clever> buildInputs = attrs.buildInputs ++ [ python36packages.Mako ];
<clever> it needs to append to buildInputs
<clever> trevthedev: that wont work
<clever> trevthedev: add it to the inputs for the derivation that needs it
<clever> trevthedev: because nix is pure and doesnt care what you have installed
<clever> LnL: buildEnv has a post hook, where you can just put a for loop that renames the symlinks
<clever> LnL: you could do another buildEnv, and then rename its symlinks, but thats a bit messy
<clever> oh, and the firewall is disabled, so neither has any effect
<clever> and there is zero point in trying to open udp port 22, ssh only listens on tcp
<clever> trevthedev: port 22 is opened automatically by the ssh module
<clever> firewall.allowedTCPPorts = list ++ [ 22 ];
<clever> trevthedev: you still have an override on llvm, line 84

2018-08-22

<clever> trevthedev: can you gist your config?
<clever> why are you trying to override llvm?
<clever> you may want to override patches = [];
<clever> those overrides are likely to blame then
<clever> did you put any overrides on llvm?
<clever> trevthedev: the error is the first line you pasted, one of the patches doesnt apply
<clever> adamantium: you can if you want to
<clever> sir_guy_carleton: is there any extra neovim's in your systemPackages on another nix file?
<clever> i cant remember why i dug that up last time
<clever> you can also use this expression to generate a naked vimrc file that uses vam, and then just run your-flavor-of-vim -c ${that} in a writeScriptBin
<clever> 2018-07-10 20:55:35< clever> nix-repl> :b vimUtils.vimrcFile { customRC = "foo"; vam.pluginDictionaries = [ { names = [ "vim-nix" ]; } ]; }
<clever> sir_guy_carleton: unknown
<clever> sir_guy_carleton: https://github.com/cleverca22/nixos-configs/blob/master/vim.nix#L33-L39 is how i did my plugins
<clever> ive not used that flag before
<clever> sir_guy_carleton: i'm guessing its something in the area of packages.nvim
<clever> sir_guy_carleton: can you gist your current nix expressions?
<clever> sir_guy_carleton: you need to set the buildPhase and installPhase
<clever> but all of the fields in spec.json match up 1:1 to the config in a jobset on the UI
<clever> i dont think it is
<clever> and then hydra configures everything else itself
<clever> sevcsik: for this project, i setup the declarative spec with "toxvpn/spec.json" and a git input of "https://github.com/cleverca22/hydra-configs"
<clever> so if you ever have to setup another hydra with the same project, it goes much faster
<clever> sevcsik: declarative projects let you put nearly all of the hydra config on github
<clever> sevcsik: have you seen how declarative projects work? they help greatly
<clever> nice
<clever> and login is fully working as well?
<clever> sevcsik: the Host: line in the cfg i linked gets the URL's to come up correctly, and the forwarded-proto keeps the protocol right as well
<clever> it may even be posting the login to that 2nd local hydra
<clever> you may have a 2nd hydra on the local machine, which is serving those URL's
<clever> sevcsik: you need this config on your nginx: https://github.com/cleverca22/nixos-configs/blob/master/nas-hydra.nix#L56-L62
<clever> sevcsik: you are sending localhost based URL's for those, one min
<clever> sevcsik: all of the css and js is failing to load
<clever> sevcsik: is the URL publicly visible?
<clever> sevcsik: yes
<clever> sevcsik: try checking the network console in chrome, via f12, while you login
<clever> viric: so the deprecated manager works better, lol
<clever> sevcsik: did you create a user from the terminal?
<clever> sevcsik: did the login work before?
<clever> sevcsik: journalctl -f -u hydra-server
<clever> viric: yeah, i use slim on my desktop
<clever> nh2: yeah, thats one option
<clever> viric: it works for me with sddm and slim
<clever> viric: when did you last modify the groups?