gchristensen changed the topic of #nixos to: Received email spam? Hydra bug, fixed! Sorry! It should be solved by now | Share the output of nix-shell -p nix-info --run nix-info to help us help you. || https://nixos.org || Latest NixOS: https://nixos.org/nixos/download.html || Latest Nix: https://nixos.org/nix/download.html || Logs: https://botbot.me/freenode/nixos/ || #nixos-dev, #nix-darwin, #nixos-aarch64, #nixos-chat
<boomshroom> Fingers crossed nixos-install doesn't overwrite the store and profiles.
<disasm> clever: right, I'm just wondering why I'm not pulling things from cache.nixos.org that are successfully building in hydra since upgrading to 18.03.
mahalel_ has quit [Remote host closed the connection]
mahalel_ has joined #nixos
Itkovian has joined #nixos
<clever> disasm: can you give me an example of a storepath its building?
<disasm> /nix/store/xfdjqyvp7h0n63y8xdx4wg6n7jj9kfwr-zziplib-0.13.68.drv
<clever> i need the outpath of it
<{^_^}> [nixpkgs] @dezgeg pushed to master « sssd: 1.16.0 -> 1.16.1 »: https://git.io/vxepn
<clever> URL: nar/0rn7rf3y36cvhv2gypbkl8vq18f2p3vql7v9bm1sgdv4nr8s962x.nar.xz
<disasm> --prefix=/nix/store/0gfi5kk4bajp748iiv6bdswbzrqmc12p-zziplib-0.13.68
<clever> disasm: zziplib is on the binary cache
<clever> disasm: what is in your nix.conf?
hakujin has quit [Ping timeout: 276 seconds]
hakujin has joined #nixos
<clever> substituters = https://hydra.iohk.io
<clever> disasm: you dont have cache.nixos.org listed
<clever> so it never uses it
mahalel_ has quit [Remote host closed the connection]
<disasm> oh, I betcha everything I wanted from 17.09 was in iohk binary cache. I thought it was additive in that it kept the default cache (it kept the key oddly enough)
<clever> yeah, something sometimes doesnt work right with that in the binary cache
jrolfs has joined #nixos
<{^_^}> [nixpkgs] @YorikSar opened pull request #36835 → bitcoin-unlimited: fix build with boost 1.66 → https://git.io/vxepV
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<boomshroom> I find it funny how wifi is hell to set up, but USB tethering works out of the box.
alaricsp has quit [Ping timeout: 260 seconds]
<clever> boomshroom: for all of my systems, the wifi drivers just work, and all i have to do is copy /etc/wpa_supplicant.conf in to make it live
<clever> and hardware.wireless.enable = truee;
jrolfs has quit [Ping timeout: 240 seconds]
* deepfire waves at disasm and clever
* clever waves back
<boomshroom> clever: That assumes you have a wpa_supplicant premade and available without internet.
<clever> boomshroom: yeah
<clever> boomshroom: if your on wep or open, you can manually connect using iwconfig
alaricsp has joined #nixos
<{^_^}> [nixpkgs] @markuskowa opened pull request #36836 → beegfs: introduce patches for kernel 4.14 → https://git.io/vxepb
<disasm> hi deepfire :)
<boomshroom> This time I did actually have a wpa_supplicant available offline from the previous NixOS installation, but I'm using USB tethering anyways because its still easier.
hakujin has quit [Ping timeout: 265 seconds]
hakujin has joined #nixos
mahalel_ has joined #nixos
mahalel_ has left #nixos [#nixos]
mahalel_ has joined #nixos
<Lisanna> ...what is tarball-ttl?
<clever> Lisanna: the ttl for builtins.fetchTarball
<Lisanna> it's a nix.conf option?
<clever> yes
<Lisanna> and it means "if I've already downloaded a tarball within tarball-ttl, don't bother checking the server for a new one"?
<clever> yep
<Lisanna> what's its default value?
<clever> an hour i think?
<Lisanna> oh god
<clever> ?
<Lisanna> okay, this was the thing that gave me about an hour of grief yesterday
<clever> heh
<Lisanna> nix-channel from nix 2.0 started using downloadCached
<MichaelRaskin> .~
<Lisanna> which also uses tarball-ttl
<clever> ahh
<Lisanna> so that's why my channel wasn't updating
hakujin has quit [Ping timeout: 256 seconds]
<clever> Lisanna: ~/.cache/nix/tarballs/ is where it stores the cache
<clever> if you wipe that, it will re-check everything
<Lisanna> I assumed that channels were usable as more or less direct pipes full of nix code, and so I have been using a nix channel in a code->deploy->run cycle
<Lisanna> Is this desirable for channel tarballs to have a TTL?
<boomshroom> I reinstalled nixos and now I'm stuck on SDDM. I don't have a password yet and ctrl-alt-f1 doesn't seem to do anything.
<Lisanna> ...I think most users would expect nix-channel --update to update the channel
<Lisanna> regardless of any ttl, especially if they didn't set it
<boomshroom> Need to use fn key.
<samueldr> hi o/ anyone seen another weird cut-off github message from {^_^}?
<Lisanna> samueldr the recent ones seem fine?
<samueldr> yeah, that's why I'm asking
<samueldr> since I did nothing
hakujin has joined #nixos
<samueldr> oh, sphalerite, seen your earlier messages, I'm logging here https://logs.nix.samueldr.com/nixos
<samueldr> (and anyone else)
<samueldr> if botbot's making trouble, looks like I'll need to make sure to keep logging :)
mahalel_ has quit [Remote host closed the connection]
mahalel_ has joined #nixos
<Acou_Bass> eyy folks, just a quick question - lets say i have two channels in my user profile, so ive done nix-channel --add urltonixos-17.09 nixpkgs, AND nix-channel --add urltonixos-unstable unstable... if i want to install something from unstable, is it just a case of nix-env -iA unstable.packagename... and then if i want it from stable, nix-env -iA nixpkgs.packagename?
freeman42x]NixOS has quit [Ping timeout: 256 seconds]
mahalel_ has quit [Client Quit]
mahalel_ has joined #nixos
<samueldr> oh, and earlier, (if any of those people are here now) there were weird "PR merge" where the author of the PR was used instead of the user who closed the PR, I'm working on it
* samueldr is scanning the whole backlog for cut-off messages to understand why
mahalel_ has quit [Remote host closed the connection]
hlolli has quit [Remote host closed the connection]
<clever> samueldr: i checked it here, nothing looks cut off
<gchristensen> samueldr: quite likely a bug in the ircbot gateway component
<Lisanna> Acou_Bass yes
<gchristensen> I've seen it happen that way if the message ends with a trailing neweline
<clever> 2018-03-11 16:44:45-{^_^}:#nixos- → 880311a5 by @grahamc: Merge pull request #36824 from dupgit/patch-
<clever> samueldr: maybe that?
mahalel_ has joined #nixos
<Acou_Bass> thanks, just checking ;D
<clever> 2018-03-11 16:52:11-{^_^}:#nixos- → 0edb8f7b by @dtzWill: Merge pull request #36810 from dtzWill/fix/glibc-with-mus
<clever> or that?
<samueldr> that's the kind of wrroneous messages I'm talking about
<samueldr> yes to both of them
<samueldr> they're not all problematic though :/
<samueldr> this too : <{^_^}> → 9a68c569 by @7c6f434c: Merge pull request #34067 from mogorman/gnucla
<dtz> (sounds like y'all are debugging a bot --no issues re:that commit right? just saw ping and wanted to make sure ;))
<clever> dtz: yeah, the commit messages are a bit cut off
<samueldr> in my particular case, no, a replacement of nixos_github bot
<samueldr> ah, I thought you said "a bot *that commits*" lol
<boomshroom> How do I get /nix/var/nix/profiles/default out of my path?
<dtz> Haha okay!
<samueldr> though, not all commit subjects, and those that are cut-off don't look any weirder
<boomshroom> It's the biggest issue with the shared store.
<clever> boomshroom: when i shared a store between platforms, i had a profiles.x86 and a profiles.arm, and then i used fileSystems."/nix/var/nix/profiles" = { fsType = "bind"; ...
<clever> to bind-mount the right one
<{^_^}> [nixpkgs] @knedlsepp opened pull request #36837 → pythonPackages.shapely: fix darwin build → https://git.io/vxejl
<samueldr> gchristensen: I'll look into stripping/chomping all \n, just in case, but it would be weird to have \n from the github API responses
<boomshroom> clever: And both were seen by the garbage collector as roots?
<clever> boomshroom: i had to manually add both to the gcroots directory
<samueldr> though, since you said that you've seen that happen previously, could it be something else than newlines? only that now we're using the bot more often, it happens more often?
jrolfs has joined #nixos
<samueldr> the weirdest part for me is how it cut off something *not* at the last char earlier on
<gchristensen> samueldr: oh?
<samueldr> (UTC) 19:24 <{^_^}> [nixpkgs] @dezgeg pushed to release-18.03 « nix: Fix build on ARMv
<samueldr> this should've finished with "6 »"
<samueldr> maybe even an URL
<samueldr> yes, with an URL
<{^_^}> [ofborg] @grahamc reopened pull request #42 → (scratch PR for live-fire testing) → https://git.io/vxejM
mahalel_ has quit [Remote host closed the connection]
mahalel_ has joined #nixos
<boomshroom> What even is the 'default' profile? I've only seen more than 2 generations of it on my single user mac installation.
jrolfs has quit [Ping timeout: 240 seconds]
mahalel_ has quit [Client Quit]
mahalel_ has joined #nixos
{^_^} has quit [Remote host closed the connection]
{^_^} has joined #nixos
{^_^} has quit [Changing host]
{^_^} has joined #nixos
mahalel_ has quit [Remote host closed the connection]
<{^_^}> [nixpkgs] @markuskowa opened pull request #36838 → texmaker: 4.5 -> 5.0.2 → https://git.io/vxejh
<boomshroom> Also, trying to access the terminal as my user causes both bash and fish to freak out. Bash still kind of works, but fish boots me out completely.
mahalel_ has joined #nixos
mahalel_ has quit [Client Quit]
<boomshroom> bash: cannot set terminal process group (7231): Inappropriate ioctl for device
mahalel_ has joined #nixos
hakujin has quit [Ping timeout: 255 seconds]
mahalel_ has left #nixos [#nixos]
<{^_^}> [nixpkgs] @pjan opened pull request #36839 → updates gist to 4.6.2 → https://git.io/vxveJ
<boomshroom> Fish has a less descriptive version of the same error.
mahalel_ has joined #nixos
{^_^} has quit [Remote host closed the connection]
{^_^} has joined #nixos
{^_^} has quit [Changing host]
{^_^} has joined #nixos
oahong has quit [Ping timeout: 256 seconds]
mahalel_ has quit [Remote host closed the connection]
mahalel_ has joined #nixos
oahong has joined #nixos
mahalel_ has quit [Client Quit]
<clever> boomshroom: the default profile is roots nix-env profile
mahalel_ has joined #nixos
hakujin has joined #nixos
<boomshroom> clever: That explains why it didn't get very much action. I either installed something on my user, or through configuration.nix.
<boomshroom> I already switched root to use /nix/var/nix/profiles/per-user/root/nixos\
spear2 has joined #nixos
<clever> ah yeah, if you just mess with the ~/.nix-profile symlink, you can get nix-env to use any profile
<clever> that would probably be simpler, if you only have 1 nixos
<clever> but in my case, i was running arm and x86 nixos from the same disk
<boomshroom> clever: profiles/nixos-arm and profiles/nixos-x86?
<clever> yeah
<clever> or rather
<clever> profiles-arm/system
<clever> nixos-rebuild cant be told to use a different profile
<boomshroom> clever: right. You can temporarilly, but you can't switch like with nix-env.
<clever> yeah
<clever> so i used a bind-mount at boot time to change the whole profiles directory
<boomshroom> Add that to the issue tracker? :P
jwynn6 has joined #nixos
<clever> boomshroom: each nixos had a unique value for its device=
<rawtaz> soooo. is it proper to say that NixOS == nixpkgs?
<disasm> rawtaz: same github repo, but nixpkgs is used for nix on non-nixos systems as well
<rawtaz> from what i can tell, NixOS is basically a collection of packages in a structured way, or how to describe it.
<clever> rawtaz: yeah, nixos is just a single giant package called top-level, that depends on a ton of other specialized packages
<rawtaz> disasm: why is it not in its own repo?
<clever> and when ran, it boots a distro
<boomshroom> rawtaz: nixpkgs is a collection of build scripts. Some of those scripts just happen to make an OS. That's one way to think of it.
<rawtaz> yeah
<disasm> rawtaz: they share the same packages. Managing forks of different packages for nixos/non-nixos would be a major pain.
<rawtaz> but if NixOS is basically just a top package and/or a bunch of scripts, then that should be decoupled from the nixpkgs and not be a problem to maintain separately
alexteves has quit [Quit: Leaving]
<clever> rawtaz: the commands you use to start a daemon depend on what version it is
<clever> rawtaz: so it helps if they are tracked in the same repo and kept in sync with eachother
<rawtaz> bah :P
<rawtaz> okay
hakujin has quit [Remote host closed the connection]
<rawtaz> so why isnt Nix in nixpkgs too?
<clever> nixpkgs is just build scripts, not source code
<rawtaz> perhaps one day it'll be clearer
<clever> where to get the source, how to build and configure it, what depends on what
<boomshroom> rawtaz: Becuase Nix is an actual program. Nixpkgs does contain the instructions to build Nix though.
<boomshroom> It's the same reason the Arch repository is separate from Pacman.
<boomshroom> And just like Pacman, Nix can be configured to use alternate repositories.
<rawtaz> yeah
<rawtaz> its just the NixOS being inside Nixpkgs repo that's super weird
<rawtaz> the other things are fine :)
<gchristensen> rawtaz: only because no other OS's package manager is super useful outside of that OS
hakujin has joined #nixos
<boomshroom> rawtaz: ya. I'm actually having trouble finding where the binaries for nixos-rebuild et al are stored.
<boomshroom> clever: Thank you.
infinisil has quit [Quit: Configuring ZNC, sorry for the join/quits!]
infinisil has joined #nixos
<{^_^}> [nixpkgs] @pjan merged pull request #36839 → updates gist to 4.6.2 → https://git.io/vxveJ
<{^_^}> [nixpkgs] @grahamc pushed 2 commits to master: https://git.io/vxve7
<{^_^}> → 68f60bcc by @pjan: updates gist to 4.6.2
<{^_^}> → 418b9ecb by @grahamc: Merge pull request #36839 from pjan/update_gis
mk-fg has quit [Ping timeout: 256 seconds]
<clever> thats a bit ugly, but it makes sense
mahalel_ has quit [Remote host closed the connection]
<rawtaz> boomshroom: :D
mahalel_ has joined #nixos
mk-fg has joined #nixos
mk-fg has quit [Changing host]
mk-fg has joined #nixos
<Lisanna> is there a nice way to watch a nix build log as it's building? not being able to see what's going on with the new nix build command makes me nervous :p
<clever> Lisanna: i still use nix-build and it outputs the logs
<Lisanna> yeah I know I can use nix-build, but I like the new syntax (no -A, channels are easier to specify)
<Lisanna> is there any significant difference between the two other than UI?
<clever> ive not really used `nix` for building for building yet
<MichaelRaskin> Not printing the resulting path
<clever> ive only use `nix copy`, `nix copy-sigs` and `nix repl`
<MichaelRaskin> Both theory and observation tell me that deep underneath there is a single operation called «realise the derivation» that is called by both
<Lisanna> MichaelRaskin oh geez, your'e right, it doesn't even print the result path
jrolfs has joined #nixos
hiroshi has quit [Ping timeout: 255 seconds]
<MichaelRaskin> There is an issue about that, though
<Lisanna> ...that doesn't seem particularly useful
<Lisanna> okay, good
<clever> nix-store -r /nix/store/foo.drv builds a derivation
<MichaelRaskin> Well, you can always --link!
<clever> nix-instantiate '<nixpkgs>' -A hello turns a nix expression into a derivation
<clever> nix-build was a perl script that used both
<Lisanna> oh right, I completely forgot about ./result x_x
<clever> then the perl was purged, and nix-build became a c++ binary that just called the library functions in turn
hakujin has quit [Ping timeout: 256 seconds]
mk-fg has quit [Ping timeout: 245 seconds]
<joepie91> hm. TIL: Nix has an 'or' operator
<joepie91> also not in the documentation..
aristid has quit [Ping timeout: 260 seconds]
<Lisanna> joepie91 you mean ||?
<joepie91> no, I mean a literal 'or'
jrolfs has quit [Ping timeout: 245 seconds]
<Lisanna> ...where?
<joepie91> configureFlags = attrs.configureFlags or "" + "--enable-wayland=no";
<joepie91> that's from some version of all-packages.nix
<joepie91> and it's highlighted specially by my syntax highlighter too
<Lisanna> OR_KW
<Lisanna> ...oh
<joepie91> presumably a lower-precedence version of ||?
<joepie91> er
<joepie91> higher*
<Lisanna> 'true or false' doesn't evaluate
leat has quit [Ping timeout: 240 seconds]
<Lisanna> yeah I have no clue what it means
<clever> Lisanna: `or` is for attributes, it will use the right side of the attribute on the left doesnt exist
hakujin has joined #nixos
<Lisanna> Oh, that's handy!
infinisil has quit [Quit: Configuring ZNC, sorry for the join/quits!]
<clever> or { return OR_KW; }
<clever> from lexer.l
infinisil has joined #nixos
<clever> and how the parser parses it
<rawtaz> joepie91: if it is not in the documentation, would it be sensible to file a PR for it?
<clever> any simple expression, even (import <nixpkgs>{})
<clever> a .
<clever> and then an attribute path
<clever> (import <nixpkgs>{}).hello
<clever> then `or` and another expr_select
<joepie91> rawtaz: I'll file an issue once I'm done mapping out the stuff that's missing :P
<joepie91> quite a few things are
<Lisanna> it's documented
<joepie91> where?
<Lisanna> first row
<clever> ah
<clever> line 360 of parser.y also has special handling, to let you call a function named `or`
<joepie91> ... okay, I did not parse that as referring to an 'or' operator, and it does not explain it...
<joepie91> well
<joepie91> not clearly at least
<Lisanna> "If the attribute doesn’t exist, return def if provided, otherwise abort evaluation."
<Lisanna> idk, it seems pretty clear to me
<Lisanna> I didn't know about it because I don't read the nix manual cover-to-cover regularly ^^
<boomshroom> How can I unmark GC roots?
<joepie91> Lisanna: it doesn't explain what part the or operator is responsible for or that it's an optional additioin
<joepie91> like, you can defend it as technically correct
<joepie91> but it's not well-explained
<clever> boomshroom: is it just a result symlink you left somewhere?
<joepie91> anyway, dinner
<Lisanna> or isn't its own operator, it's only valid when used in combination with the dot operator
<Lisanna> I think the documentation is fine, but file an issue if you think it can be improved!
<clever> ExprSelect(const Pos & pos, Expr * e, const AttrPath & attrPath, Expr * def)
<boomshroom> clever: no, I mean paths in the profiles directory. In addition to the `default` in need of purging, I may have copied over more symlinks than I did environments that they point to.
mahalel_ has left #nixos [#nixos]
<clever> boomshroom: ah, nix-env -p /path/to/profile --delete-generation 5
<clever> --delete-generations*
<boomshroom> clever: Thank you.
<Lisanna> clever yeah, it's a pretty weird / specific syntax
rardiol1 has joined #nixos
* clever heads off to bed
<rawtaz> bai clever
hakujin has quit [Ping timeout: 265 seconds]
<boomshroom> Cannot delete current generation
aristid has joined #nixos
xcmw has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
hakujin has joined #nixos
hiroshi has joined #nixos
jrolfs has joined #nixos
<{^_^}> [nixpkgs] @kmicklas opened pull request #36841 → callCabal2nix: Use host platform rather than build platform for evaluation of Cabal files → https://git.io/vxvfG
vidbina has quit [Ping timeout: 276 seconds]
hakujin has quit [Ping timeout: 252 seconds]
<{^_^}> [nixpkgs] @grahamc merged pull request #36805 → gauge: 0.9.6 -> 0.9.7 → https://git.io/vxeWv
<{^_^}> [nixpkgs] {"login"=>"grahamc", "id"=>76716, "avatar_url"=>"https://avatars3.githubusercontent.com/u/76716?v=4", "gravatar_id"=>"", "url"=>"https://api.github.com/users/grahamc", "html_url"=>"https://github.com/grahamc", "followers_url"=>"https://api.github.com/users/grahamc/followers", "following_url"=>"https://api.github.com/users/grahamc/following{/other_user}", "gists_url"=>"https://api.github.com/users/grahamc/gists{/gist_id}", "starred_url"=>"https:/
<{^_^}> → 35b7632b by @vdemeester: gauge: 0.9.6 -> 0.9.7
<{^_^}> → 1bfe2511 by @grahamc: Merge pull request #36805 from vdemeester/update-gauge-097
<gchristensen> hooo buddy samueldr
jrolfs has quit [Ping timeout: 252 seconds]
mk-fg has joined #nixos
mk-fg has quit [Changing host]
mk-fg has joined #nixos
mutantmell has quit [Quit: WeeChat 1.3]
<Lisanna> lol
<samueldr> just... don't push :)
nuncanada has quit [Ping timeout: 260 seconds]
<samueldr> (no!, please push!)
mk-fg has quit [Ping timeout: 260 seconds]
hakujin has joined #nixos
xcmw has joined #nixos
<{^_^}> [nixpkgs] @7c6f434c opened pull request #36842 → nixpkgs/tests: create nixos-independent PoC tests → https://git.io/vxvf9
<disasm> well nixos handles upgrades fantastically, but a lot of packages/services I have setup like to really break reverse compatibility requiring lots of debugging, lol :)
<{^_^}> [nixpkgs] @grahamc closed pull request #36610 → librime: 1.2.9 -> 1.2.10 → https://git.io/vAxAW
<{^_^}> [nixpkgs] @grahamc pushed 3 commits to master: https://git.io/vxvf7
<{^_^}> → 1c99ffcb by @sifmelcara: librime: 1.2.9 -> 1.3.0
<{^_^}> → cbf735e3 by @sifmelcara: brise: change platforms to linux only
<{^_^}> → e065aa94 by @grahamc: Merge pull request #36808 from sifmelcara/update/librime
<{^_^}> [nixpkgs] @grahamc merged pull request #36808 → librime: 1.2.9 -> 1.3.0 → https://git.io/vxe89
hakujin has quit [Ping timeout: 256 seconds]
markus1199 has joined #nixos
Supersonic has quit [Disconnected by services]
Supersonic112 has joined #nixos
n0qqe has joined #nixos
Supersonic112 is now known as Supersonic
markus1189 has quit [Ping timeout: 268 seconds]
noqqe has quit [Ping timeout: 260 seconds]
n0qqe is now known as noqqe
xcmw has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
jrolfs has joined #nixos
leat has joined #nixos
rosa has joined #nixos
hakujin has joined #nixos
asuryawanshi has joined #nixos
mk-fg has joined #nixos
mk-fg has quit [Changing host]
mk-fg has joined #nixos
rosa has quit [Client Quit]
infinisil has quit [Quit: Configuring ZNC, sorry for the join/quits!]
infinisil has joined #nixos
jrolfs has quit [Ping timeout: 256 seconds]
<{^_^}> [nixpkgs] @pjan opened pull request #36843 → updates youtube-dl → https://git.io/vxvJw
mtjmullen has joined #nixos
Guestv has joined #nixos
hakujin has quit [Ping timeout: 265 seconds]
asuryawanshi has quit [Ping timeout: 240 seconds]
<Guestv> Probably dumb question here please. Is it advisable to store my own modules in /etc/nixos ?
mk-fg has quit [Ping timeout: 256 seconds]
<Guestv> I mean for a user
<Guestv> First time installer here.. so i'm a little confused
<Lisanna> Guestv what do you mean by "store"?
<Lisanna> you mean store the nix code for them there and then in your imports have imports = [ /etc/nixos/mymodule.nix ] ?
<disasm> Guestv: that's what I do
<Guestv> Yes please
<Guestv> oh ok
<Guestv> Thanks
<Lisanna> I wouldn't necessarily do that... how are you adding them to /etc/nixos?
mk-fg has joined #nixos
mk-fg has quit [Changing host]
mk-fg has joined #nixos
<Guestv> writing them there myself
<infinisil> Guestv: yeah that's totally okay, I personally have a ton of my own modules in my system configuration (but my system config isn't in /etc/nixos, but that doesn't matter anyways)
<Lisanna> ok, if you're writing them there yourself, that sounds fine
<disasm> Guestv: if you're looking for structural inspiration, I recently pushed my configs to github :) https://github.com/disassembler/nixos-configurations
hakujin has joined #nixos
* infinisil takes a peek
ottidmes has quit [Ping timeout: 268 seconds]
<Guestv> Thanks Sam
blonkhart has quit [Quit: WeeChat 1.9.1]
<disasm> infinisil: those are my laptops. The other machines in the house use nixops: https://github.com/disassembler/network -- I'm debating merging the two git repos and having deployments under the root of the configurations one
<infinisil> disasm: you mean use nixops for all of them?
<Guestv> Do you by chance have a postgres setup in there somewhere?
owickstrom has quit [Remote host closed the connection]
owickstrom has joined #nixos
<infinisil> Guestv: This is probably the best resource for postgresql: https://nixos.wiki/wiki/PostgreSQL
<disasm> infinisil: my laptops are just git cloned off that first repo, nixops is used for my desktop (optina) and router (portal)
hakujin has quit [Ping timeout: 260 seconds]
orivej has joined #nixos
<infinisil> I personally use nixops for all my machines, which is admittedly a bit slow
<infinisil> But the advantage is that I can refer to other machines configs like this and do fancy stuff with it
<boomshroom> infinisil: How's NixOps for managing user machines?
<infinisil> boomshroom: There are some minor things you need to watch out for, but it works well
mtjmullen has quit [Quit: WeeChat 1.9.1]
<disasm> I wish I could figure out why my zfs home and tmp partitions fail to mount with 18.03 and then the system just hangs :(
<infinisil> disasm: Do you have the error?
<boomshroom> On my new NixOS installation, bash keeps acting funny and attempting to use fish fails entirely.
hakujin has joined #nixos
pkill9 has quit [Ping timeout: 256 seconds]
<boomshroom> It has trouble setting the terminal correctly. bash complains about an innapropiate ioctl.
<disasm> infinisil: filesystem 'zroot/root/tmp' can not be mounted: Read-only file system
Guestv has quit [Ping timeout: 260 seconds]
<infinisil> disasm: Well my immediate guess is that readonly could be turned on, can be checked with `zfs get readonly zroot/root/tmp`
<boomshroom> As a part of my expanding quest to Nix the world, has anyone tried getting Nix to run on Plan 9?
<disasm> boomshroom: is that still around?
<boomshroom> disasm: Kind of, I think?
<boomshroom> Wikipedia says the latest release was in 2015.
hakujin has quit [Ping timeout: 276 seconds]
mbrgm has quit [Ping timeout: 260 seconds]
boomshroom has quit [Quit: WeeChat 2.0]
boomshroom has joined #nixos
<boomshroom> Note to self "/quit" exits all of weechat, not just the current channel.
<infinisil> Indeed :P
hakujin has joined #nixos
mbrgm has joined #nixos
<boomshroom> infinisil: To make sure it doesn't happen again, what is the command to leave a single channel?
<infinisil> boomshroom: /part
<boomshroom> I would not have guessed that.
<disasm> /close works too :)
<boomshroom> Thank you.
mtjmullen has joined #nixos
jrolfs has joined #nixos
hakujin has quit [Ping timeout: 260 seconds]
<boomshroom> The plan 9 subreddit isn't dead. That's for sure.
Poca has joined #nixos
<disasm> infinisil: you're using grub with efi, right?
<infinisil> Yeah
<disasm> infinisil: can you paste your config? I'm getting You must set the option ‘boot.loader.grub.devices’ or 'boot.loader.grub.mirroredBoots' to make the system bootable. Which if it's just generating an efi image, I'm not sure why it requires those.
hakujin has joined #nixos
<infinisil> You just need to set it to "nodev" with efi
mk-fg has quit [Ping timeout: 256 seconds]
<disasm> ah :)
jrolfs has quit [Ping timeout: 264 seconds]
<infinisil> There might be some stuff unnecessary in there though
<Lisanna> Is it possible to perform a "nix copy --to" as part of a nix build?
<infinisil> Lisanna: You seem to be all about doing stateful stuff in Nix again!
<Lisanna> I'm running into errors in my test of it, so wondering if someone's already worked out a solution
<Lisanna> infinisil what, no!!!
<infinisil> (That's totally stateful though
<infinisil> )
<Lisanna> I want to generate a binary cache in nix, that's all
<Lisanna> as part of a nix build
<Lisanna> that's not stateful!
<boomshroom> When I started looking up the current state of Plan 9, I was not expecting to see a Touhou character.
<Lisanna> I want to do nix copy --to $out
<infinisil> Lisanna: Ah, so you want to use `nix copy` to fetch stuff for a nix build?
hamishmack has quit [Ping timeout: 260 seconds]
<infinisil> Well I have no idea how that would work, it's one of those recursive nix things
<Lisanna> well, I mean, not really fetch stuff, just to generate a binary cache as a build product that I can then move elsewhere
<Lisanna> and use on other systems
<Lisanna> yeah, I know it's one of those recursive nix things, but it seems like a useful thing ot be able to do, but nix tries to create /nix/var in the sandbox, so just wondering if someone's already pathcleared this usecase
hakujin has quit [Ping timeout: 276 seconds]
<Lisanna> ...in theory it *shouldn't* *need* to write anything to /nix to do this...
<Lisanna> I guess I could just do this in a runInLinuxVM
<infinisil> Heh yeah, but even parsing a nix file wants write access to /nix/var
<Lisanna> actually, yeah, runInLinuxVM sounds like the way to go here
<Lisanna> I *know* nix copy --to works in there :)
<infinisil> I'm not sure how I feel about meta-nix.. especially when using runInLinuxVM is a solution..
<Lisanna> what do you mean?
<infinisil> I feel like this should be handled with nix commands, not the nix language
<Lisanna> Nix is about building things... it's a language for how to build stuff. I want to build a binary cache that I then shove onto a bootable disk image. a binary cache is a thing to be built, so of course I want to use nix to build it
<infinisil> Hmm.
jakeLand has joined #nixos
hakujin has joined #nixos
<infinisil> Fair enough
<jakeLand> hi everyone, I am new to nixos and I am trying to override an attribute of a dependency of one of my installed packages (changed a build flag). I've figured out how to override it in my configuration.nix, I think, but how do I make sure that my installed package uses this overriden dependency? when I do nix-rebuild it doesn't recompile the dependency, however if I put it in enviroment.systemPackages then it does compile with the new
<infinisil> Lisanna: Oh, actually nixpkgs/nixos does something like this, hold on
<Lisanna> infinisil I'm sure it does :)
<Lisanna> jakeLand something something overlays
<Lisanna> (I am a very helpful person)
<infinisil> Lisanna: What is a binary cache really? What does it contain?
<Lisanna> it's the result of nix copy --to /some/path, or nix-push if using 1.11
<Lisanna> it's the one that has .nar and .narinfo files in it
<infinisil> jakeLand: you can use `nixpkgs.overlays = [ (self: super: { hello = super.hello.override { ... }; }) ];`
<jakeLand> thanks guys, I will look into this overlay thing. hopefully it will not be too difficult
<Lisanna> yeah, I think I'm just going to write a "mkNixBinaryCache" function that uses runInLinuxVM. Unfortunately that means it won't work on anything other than linux though, which is a little bit sad
jrolfs has joined #nixos
<infinisil> Lisanna: Hmm, so I guess the somewhat clean way to do it would be to implement this specific nix copy usage in a nix build with the standard commands, but that's not really nice
hakujin has quit [Ping timeout: 256 seconds]
<infinisil> And the actual solution would be to fix these friggin nix commands to they work recursively, since they obviously should
<Lisanna> recursive nix = expose a /nix filesystem, but only containing the packages in the closure of your build, and make it read-only
<Lisanna> ...and fix the commands to not need to write to /nix to perform what the user sees as a read operation
<infinisil> Lisanna: but all of /nix is read-only available in builds already?
<Lisanna> is the database available?
<infinisil> oh right, only /nix/store
<Lisanna> yeah
asuryawanshi has joined #nixos
<Lisanna> don't want to enable that by default though, since most builds do not need recursive nix, and generating that /nix would take some time
<disasm> I know we don't need it because we can rollback to a previous version (how I'm currently booted) but what would be the best way to get into a system that doesn't boot fully to a tty to live debug on nixos?
mk-fg has joined #nixos
mk-fg has quit [Changing host]
mk-fg has joined #nixos
hakujin has joined #nixos
jrolfs has quit [Ping timeout: 256 seconds]
<Lisanna> ...actually recursive nix could be implemented with a custom builder today and runInLinuxVM... write a builder which runs the nix installer to get a base install and add the closure of a build with exportReferencesGraph to the nix store, shove the whole thing in a $out, and use unshare to map that to your nix build's sandbox env at /nix
<Lisanna> (with creating a /nix install done in runInLinuxVM)
<infinisil> disasm: I feel like there should be a way to boot using an usb stick instead, then call the init of the machine to start it nontheless, but without having to use its bootloader
<infinisil> ah, but i guess you mean it not working after the bootloader
<infinisil> I have no idea
asuryawanshi has quit [Ping timeout: 256 seconds]
<Lisanna> heck you could even make it writable with unshare, then a bunch more nix commands might work (Although the written results obviously won't persist)
<infinisil> Lisanna: That's horrible
<Lisanna> infinisil I know :p
<infinisil> Somebody needs to take some time to implement recursive nix
<infinisil> (But there is nobody that does it)
<infinisil> Well I need to sleep, good night!
<Lisanna> goodnight :)
hakujin has quit [Ping timeout: 265 seconds]
<disasm> infinisil: night!
<{^_^}> [nixpkgs] @lloydhazlett opened pull request #36844 → spideroak: 6.1.9 -> 7.0.1 → https://git.io/vxvky
<boomshroom> Apparently Plan9 doesn't have symlinks. While it should be possible to implement Nix-like functionality, it would require a lot of effort to convert the symlinks into mount points.
jakeLand has quit [Quit: Page closed]
hakujin has joined #nixos
<boomshroom> Looking at nix's output, I'd really like to fix my terminal.
jrolfs has joined #nixos
hakujin has quit [Ping timeout: 260 seconds]
tertle||eltret has quit [Quit: Connection closed for inactivity]
jrolfs has quit [Ping timeout: 256 seconds]
hakujin has joined #nixos
<boomshroom> Fixed my terminal. All it needed was an update.
hakujin has quit [Ping timeout: 240 seconds]
disasm has quit [Quit: WeeChat 1.9.1]
hakujin has joined #nixos
<pjan> If anyone of the maintainers has some time to review / comment on https://github.com/NixOS/nixpkgs/pull/36790, it would be greatly appreciated. I have some room this week to address remarks & create the follow-up pr's mentioned before I'll be out for holidays.
hamishmack has joined #nixos
hakujin has quit [Ping timeout: 245 seconds]
disasm has joined #nixos
hakujin has joined #nixos
hakujin has quit [Ping timeout: 265 seconds]
disasm has quit [Quit: WeeChat 2.0]
disasm has joined #nixos
disasm has quit [Client Quit]
disasm has joined #nixos
mmlb has quit [Ping timeout: 260 seconds]
mmlb has joined #nixos
blankhart has quit [Ping timeout: 256 seconds]
blankhart has joined #nixos
<boomshroom> What's the state on true declarative use environment management? I think there was a tool to let it manage the home directory, but I'm not sure.
hakujin has joined #nixos
schoppenhauer has quit [Ping timeout: 264 seconds]
jrolfs has joined #nixos
disasm has quit [Quit: WeeChat 2.0]
schoppenhauer has joined #nixos
<adisbladis[m]> boomshroom: There are more than one tool :) I'm using home-manager
<adisbladis[m]> It works pretty great
hakujin has quit [Ping timeout: 252 seconds]
<boomshroom> It just seems like there are so many competing versions it's hard to choose.
<Lisanna> Ugh... the only downside with using runInLinuxVM to do a nix copy is that it's *so slow*
jrolfs has quit [Ping timeout: 264 seconds]
<{^_^}> [nixpkgs] @jbedo opened pull request #36845 → singularity: 2.4 -> 2.4.2 → https://git.io/vxvLR
disasm has joined #nixos
hakujin has joined #nixos
<Lisanna> how do I make my derivation output a special message to nix build like stdenv.mkDerivation does for each phase?
hakujin has quit [Ping timeout: 256 seconds]
rogue_koder has joined #nixos
hakujin has joined #nixos
<Lisanna> i.e. when nix build displays: "[1/0/1 built] building mypackage (configurePhase): checking for thing..."
<Lisanna> I don't see it ):
scribbler has joined #nixos
hakujin has quit [Ping timeout: 276 seconds]
<Lisanna> oh, well, found the commit that enables it: https://github.com/NixOS/nix/commit/0ac35b67b8035ad1166df3fb199590bc346d71f2
rujavada has joined #nixos
<rujavada> Hi everyone.
<rujavada> So as a thought experiment i deleted the nix config file and tried to rebuild
<Lisanna> found it :)
<rujavada> It failed
<rujavada> So there's that. Is there a way to regenerate from what i already have ?
<Lisanna> rujavada "thought experiment"? are you running nix in your mind? :p
hakujin has joined #nixos
<rujavada> Lol wrong choice of word?
<Lisanna> what do you mean regenerate? nix.conf isn't really generated... you write it
<rujavada> I wanted to see what would occur if i delete i tried to rebuild.
<rujavada> Yeah, well i don't remember everything i had in there
<Lisanna> so... are you asking how to un-delete it?
<{^_^}> [nixpkgs] @Ericson2314 merged pull request #36841 → callCabal2nix: Use host platform rather than build platform for evaluation of Cabal files → https://git.io/vxvfG
<{^_^}> [nixpkgs] @Ericson2314 pushed 2 commits to master: https://git.io/vxvtC
<{^_^}> → f34e136a by @kmicklas: callCabal2nix: Use host platform rather than build platform for evaluation of Cabal files
<{^_^}> → 13c5a570 by @Ericson2314: Merge pull request #36841 from obsidiansystems/cabal2nix-system-hostPlatform
<rujavada> not undelete per say
<rujavada> I'm asking if the system can sort of generate the exact same config i had?
<rujavada> because everything seems to be running well
<rujavada> And i have no configuration whatsoever in that directory
<Lisanna> are you on NixOS?
<rujavada> Yup
<gillmanash> I think you might be stuffed rujavada. I know captain hindsight is never that helpful, but maybe next time try `mv configuration.nix configuration.nix.bkp`
<rujavada> That's fine. Nothing lost
<rujavada> I'm just playing around and learning
<gillmanash> There may be a way somebody else knows of, but I haven't come accross it in the past
<rujavada> Scoured the manual and didn't see one either
hakujin has quit [Ping timeout: 256 seconds]
<Lisanna> idk, try "touch /etc/nix/nix.conf; nixos-rebuild switch"
<Lisanna> nix.conf on nixos is generated by nixos-rebuild
<rujavada> Already tried
<rujavada> Wouldn't build
<Lisanna> wait... did you delete /etc/nix/nix.conf or /etc/nix/configuration.nix?
<rujavada> configuration.nix
<Lisanna> ...oh.
<Lisanna> well that was kind of a silly thing to do wasn't it :p
<Lisanna> ...I mean, I guess you should just write a new one.
<rujavada> Not really. I'm new to Nix
<Lisanna> that really is the top-level file for NixOS
<rujavada> just writing all my do's and dont's
<Lisanna> try nixos-generate-config
<rujavada> Wasn't going to keep it anyway because i've been struggling with stack for a couple of hours
<gillmanash> FYI, I'd imagine most people keep theirs under version control at some level
<rujavada> Tried. Told me it wouldn't overwrite
<rujavada> But there's nothing in there
<Lisanna> rm -f /etc/nixos/configuration.nix; nixos-generate-config
<elvishjerricco> rujavada: I think you can still do nixos-generate config to get back a sensible minimal config. Obviously this won't bring back your old config
<elvishjerricco> rujavada: What kind of issues have you had with Stack? Bear in mind you *have* to use Stack's `--nix` argument on NixOS. You can make this the default by enabling nix in the global stack config
<rujavada> writing /etc/nixos/etc/nixos/hardware-configuration.nix... warning: not overwriting existing /etc/nixos/etc/nixos/configuration.nix
<elvishjerricco> --force, I believe
<Lisanna> Ohhhhh
<rujavada> except there's nothing there
<Lisanna> that's strange
<Lisanna> nixos-generate-config --root /
<rujavada> Yup tried that too
<Lisanna> what does it say?
<rujavada> same
<Lisanna> "warning: not overwriting existing /etc/nixos/configuration.nix"?
<Lisanna> does it say that exactly?
<rujavada> yup
<rujavada> when a create the file myself it allows me though
<boomshroom> My laptop is ready for action!
<Lisanna> ...and you are confident that /etc/nixos/configuration.nix does not exist
<rujavada> Just recreated one
<rujavada> It's empty
<boomshroom> How can I minimise battery usage when the lid is closed?
<elvishjerricco> rujavada: Just do nixos-generate-config --root /path/to/some/temp/dir and manually move the resulting files into the right place
<Lisanna> Sounds like nixos-generate-config needs a bug filed against it then
<Lisanna> if such a basic usecase is broken
<rujavada> Done. generated a generic conf
hakujin has joined #nixos
Rusty1_ has quit [Quit: Konversation terminated!]
<rujavada> Damn i don't remember all the packages.
jrolfs has joined #nixos
loonquawl has joined #nixos
jmeredith has quit [Quit: Connection closed for inactivity]
hakujin has quit [Ping timeout: 256 seconds]
<{^_^}> [nix] @masaeedu opened pull request #1965 → Wrap thread local in function for Cygwin → https://git.io/vxvqv
rujavada has quit [Quit: Page closed]
<infinisil> Well I'm still awake even though I said like an hour ago I'd go to sleep
<infinisil> boomshroom: hibernation
<infinisil> which won't use any power, at all
<boomshroom> infinisil: I know that's a thing. I don't know how to use or enable it.
jrolfs has quit [Ping timeout: 264 seconds]
<infinisil> boomshroom: you only need swap, preferably more than amount of RAM you have, then `systemctl hibernate`
woffs has quit [Quit: Gateway shutdown]
<boomshroom> I don't think my laptop is configured with any swap.
<infinisil> boomshroom: Run `swapon -s` to find out if you have a swap
blankhart has quit [Ping timeout: 252 seconds]
<{^_^}> [nixpkgs] @jbedo opened pull request #36846 → snpeff: 4.3t -> 4.3q → https://git.io/vxvql
blankhart has joined #nixos
guibou has quit [Ping timeout: 256 seconds]
hakujin has joined #nixos
scribbler has quit [Quit: scribbler]
endformationage has quit [Quit: WeeChat 1.9.1]
<boomshroom> No output
blankhart has quit [Quit: WeeChat 1.9.1]
<infinisil> Yeah then you don't have swap and can't do hibernation (unless you add some)
<infinisil> systemctl suspend (which is what a lid close normally does) is the next best power saving thing
rauno has quit [Ping timeout: 268 seconds]
hakujin has quit [Ping timeout: 256 seconds]
<boomshroom> infinisil: Would you recomend a deticated partition, or a ZFS volume? I've reads that you can do swap within a zfs pool and I'm wondering if you have any experience.
<Lisanna> Cool! I now have a working mkNixBinaryCache function :)
hakujin has joined #nixos
<infinisil> boomshroom: I haven't tried it myself, but I heard from other people that using zfs for swap is a bad idea
<Lisanna> ...and it's not even that slow!
<boomshroom> infinisil: My laptop has 4 GB memory. Would you recommend 5 GB swap?
jrolfs has joined #nixos
<infinisil> I just got reminded by the internet that the consensus is 2x the amount of RAM to use for swap
Myrl-saki has joined #nixos
<infinisil> so 8GB in your case
<infinisil> I have 8GB RAM and 7GB swap and that's probably why I'm having so many problems with it (even though it occasionally works)
Fare has quit [Ping timeout: 240 seconds]
<Lisanna> infinisil oh, really? why?
asuryawanshi has joined #nixos
hakujin has quit [Ping timeout: 265 seconds]
<Lisanna> (why 2x)?
<boomshroom> Unfortunately, my mac partition is so full, I only have 11 GB unused. Using 8 GB for swap (just for NixOS) would only give 3 GB for future mac files.
hakujin has joined #nixos
<boomshroom> I understand that pretty much any usage of swap other than for hibernation can slow down the computer a lot.
<infinisil> Lisanna: Hmm, it might be a myth, reading https://askubuntu.com/questions/49109/i-have-16gb-ram-do-i-need-32gb-swap#49138
<infinisil> boomshroom: Check this link out too ^^
Fare has joined #nixos
<infinisil> boomshroom: Nah, adding swap never makes anything slower
jrolfs has quit [Ping timeout: 260 seconds]
<Lisanna> lol wtf
<boomshroom> That I agree with. Using it means you're going to disk instead of RAM, which is slow.
<Lisanna> >7GB of RAM => 7GB of swap space
<Lisanna> >8GB of RAM => 4GB of swap space
<boomshroom> At least on a hard disk, which my desktop uses but I'm pretty sure my laptop doesn't.
<infinisil> Lisanna: haha, needs some smoothing
aarvar has joined #nixos
<infinisil> boomshroom: no, swap doesn't mean it does to disk instead of ram
<infinisil> boomshroom: you should probably look up how swap works
<Lisanna> or you could take starblue's advice: "The square root of the RAM in GB, rounded up to a power of two"
asuryawanshi has quit [Ping timeout: 240 seconds]
<infinisil> Lisanna: that sound nice
<boomshroom> Sorry, I associate swapping with the grinding noise my hard drive makes when it's under stress.
<Lisanna> It definitely sounds technical.
hakujin has quit [Ping timeout: 255 seconds]
<infinisil> Lisanna: But.. I need to hibernat, so it needs to be at least the amount of RAM
<infinisil> boomshroom: that is not correct, you're thinking of thrashing
<infinisil> which is closely related to swap, but not a direct consequence of it
rauno has joined #nixos
hakujin has joined #nixos
drakonis has joined #nixos
<boomshroom> I'm tempted to try taking my laptop to an Apple store and see what they can do to trim some of the disk usage. They'd probably just tell me to delete the other partitions. :P
<infinisil> Lisanna: Ah yeah the top answer mentions that swap for hibernation should be about 2 times amount of RAM
<Lisanna> boomshroom install DaisyDisk, really nice tool for seeing what's taking up your space
<boomshroom> It seems like 8GB is the recommended amount for 4GB + hibernation.
<Lisanna> ugh, nix repl doesn't show logs of failed IFDs
<infinisil> boomshroom: are you using zfs? You could enable compression for it, which helps a lot to reduce usage
Myrl-saki has quit [Read error: Connection reset by peer]
<infinisil> boomshroom: and you could choose the most space-efficient compression algorithm for it
<boomshroom> NixOS refused to nixos-rebuild switch when the root was set to compress.
Myrl-saki has joined #nixos
<infinisil> refuses? That can't be, nixos shouldn't even know that zfs compresses
hakujin has quit [Ping timeout: 252 seconds]
MP2E has quit [Remote host closed the connection]
<boomshroom> infinisil: It probably had something to do with the fact that the configuration was still based on btrfs.
<boomshroom> Also, 13$?! I'm a linux user! I like my software free!
<infinisil> Well yeah, then it's not a problem of zfs compression, but one of switching from btrfs to zfs
<boomshroom> The half off student discount could be nice though.
<Lisanna> boomshroom you're a mac user and complaining about something being expensive?
<Lisanna> (I kid, I'm also a macfag ^^)
reinzelmann has joined #nixos
<boomshroom> Lisanna: I have a macbook because I needed to round out the trifecta. Also it's unix so it's automatically better than Windows (I mean, Nix doesn't run natively on Windows.)
<boomshroom> 18GB for a backup of my 3DS SD card.
blankhart has joined #nixos
jrolfs has joined #nixos
<infinisil> boomshroom: 128GB macbook model?
<boomshroom> 256
<boomshroom> Wow, the Android SDK is big.
jrolfs has quit [Ping timeout: 256 seconds]
<dtz> too big for hydra, I rem seeing something re:that other day
<dtz> multiple outputs maybe? O:)
<boomshroom> I just realised I can cast from my desktop to my TV.
rauno has quit [Ping timeout: 256 seconds]
<dtz> :D
<boomshroom> Watching a speedpaint set to shin megami tesei music on the TV while my laptop sends a gargantuan tarball to my desktop.
<boomshroom> The transfer won.
MP2E has joined #nixos
ma27 has joined #nixos
<dtz> xD
<{^_^}> [nixpkgs-channels] @edolstra pushed 28 commits to nixos-unstable-small: https://git.io/vxvYB
<{^_^}> → d289736d by @knedlsepp: pythonPackages.regex: fix darwin build
<{^_^}> → 91a94534 by @knedlsepp: pythonPackages.construct: Fix darwin build
<{^_^}> → 57d862fe by @knedlsepp: pythonPackages.discid: fix darwin build
MercurialAlchemi has joined #nixos
rauno has joined #nixos
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/6c5aa934dda (from 12 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
drakonis has quit [Read error: Connection reset by peer]
hakujin has joined #nixos
BombStrike has quit [Quit: ZNC 1.6.5 - http://znc.in]
rauno has quit [Remote host closed the connection]
Guanin has quit [Ping timeout: 256 seconds]
<boomshroom> I just freed a good 30GB with 2 files.
Myrl-saki has quit [Ping timeout: 252 seconds]
hakujin has quit [Ping timeout: 256 seconds]
Guanin has joined #nixos
rogue_koder has quit [Quit: Konversation terminated!]
ma27 has quit [Ping timeout: 265 seconds]
<{^_^}> [nixpkgs] @badi opened pull request #36847 → qtlocation: fix darwin build → https://git.io/vxvOZ
mounty has quit [Ping timeout: 256 seconds]
jrolfs has joined #nixos
alex`` has joined #nixos
ma27 has joined #nixos
jrolfs has quit [Ping timeout: 264 seconds]
<boomshroom> How do I set my laptop to hibernate after idle for too long?
wrl_ has joined #nixos
wavewave_ has joined #nixos
blankhart has quit [Quit: WeeChat 1.9.1]
boomshroom has quit [Quit: WeeChat 2.0]
boomshroom has joined #nixos
<boomshroom> Wrong command again.
<boomshroom> Also, it would be a good idea to export the zpool when powering down right?
ok2`` has joined #nixos
shachaf_ has joined #nixos
pmeunier_ has joined #nixos
siel_ has joined #nixos
aminb_ has joined #nixos
codedmart_ has joined #nixos
<adisbladis[m]> boomshroom: Why would you wanna export your pool?
shachaf has quit [Disconnected by services]
shachaf_ is now known as shachaf
<boomshroom> adisbladis[m]: So I could import it when booting into Mac.
<disasm> infinisil: wiped the machine and rebuilt, this time with luks and no zfsUnstable. On 18.03 finally! I'm pretty sure rebuilding /home and /tmp probably would have solved the problem as well, but I think I'll wait a bit before trying zfsUnstable again.
swflint_away has joined #nixos
akapav_ has joined #nixos
ma27 has quit [Ping timeout: 256 seconds]
genesis has quit [*.net *.split]
Profpatsch has quit [*.net *.split]
pareidolia has quit [*.net *.split]
lukec has quit [*.net *.split]
pmeunier has quit [*.net *.split]
siel has quit [*.net *.split]
aminb has quit [*.net *.split]
Obscurity[m] has quit [*.net *.split]
wavewave has quit [*.net *.split]
Drakonis[m] has quit [*.net *.split]
codedmart has quit [*.net *.split]
nz has quit [*.net *.split]
johs has quit [*.net *.split]
retrry[m] has quit [*.net *.split]
nckx has quit [*.net *.split]
jack[m] has quit [*.net *.split]
EB[m] has quit [*.net *.split]
peterhoeg has quit [*.net *.split]
johnazoidberg has quit [*.net *.split]
spacekitteh[m] has quit [*.net *.split]
ok2` has quit [*.net *.split]
petersjt014[m] has quit [*.net *.split]
rnhmjoj[m] has quit [*.net *.split]
Vincenzo[m] has quit [*.net *.split]
Dezgeg[m] has quit [*.net *.split]
thra11[m] has quit [*.net *.split]
m4rv33n[m] has quit [*.net *.split]
revoltmedia[m] has quit [*.net *.split]
randomenduser[m] has quit [*.net *.split]
AdamSlack[m] has quit [*.net *.split]
bugabinga[m] has quit [*.net *.split]
paperdigits[m] has quit [*.net *.split]
imbigbrother[m] has quit [*.net *.split]
rihardsk[m] has quit [*.net *.split]
mudri[m] has quit [*.net *.split]
bennofs[m] has quit [*.net *.split]
pierrebeaucamp[m has quit [*.net *.split]
freeman42x has quit [*.net *.split]
equalunique[m] has quit [*.net *.split]
wrl has quit [*.net *.split]
NinjaTrappeur has quit [*.net *.split]
cybrian has quit [*.net *.split]
fyuuri has quit [*.net *.split]
Dezgeg has quit [*.net *.split]
alanz has quit [*.net *.split]
swflint has quit [*.net *.split]
akapav has quit [*.net *.split]
mt has quit [*.net *.split]
swflint_away is now known as swflint
wavewave_ is now known as wavewave
siel_ is now known as siel
codedmart_ is now known as codedmart
cybrian has joined #nixos
nckx has joined #nixos
oida has quit [*.net *.split]
<boomshroom> Woah!
alanz has joined #nixos
pie___ has joined #nixos
pareidolia has joined #nixos
pbogdan has quit [Ping timeout: 260 seconds]
<boomshroom> Goodnight!
boomshroom has quit [Quit: WeeChat 2.0]
astsmtl has quit [Ping timeout: 240 seconds]
owickstrom has quit [Ping timeout: 260 seconds]
pie__ has quit [Ping timeout: 240 seconds]
owickstrom has joined #nixos
pbogdan has joined #nixos
astsmtl has joined #nixos
astsmtl has quit [Changing host]
astsmtl has joined #nixos
petersjt014[m] has joined #nixos
Profpatsch has joined #nixos
freeman42x has joined #nixos
spacekitteh[m] has joined #nixos
genesis has joined #nixos
NinjaTrappeur has joined #nixos
jack[m] has joined #nixos
johnazoidberg has joined #nixos
pie___ has quit [Ping timeout: 264 seconds]
imbigbrother[m] has joined #nixos
Dezgeg has joined #nixos
revoltmedia[m] has joined #nixos
retrry[m] has joined #nixos
EB[m] has joined #nixos
peterhoeg has joined #nixos
randomenduser[m] has joined #nixos
rnhmjoj[m] has joined #nixos
paperdigits[m] has joined #nixos
rihardsk[m] has joined #nixos
lukec has joined #nixos
mudri[m] has joined #nixos
Drakonis[m] has joined #nixos
jrolfs has joined #nixos
bugabinga[m] has joined #nixos
thra11[m] has joined #nixos
m4rv33n[m] has joined #nixos
AdamSlack[m] has joined #nixos
Dezgeg[m] has joined #nixos
pierrebeaucamp[m has joined #nixos
Vincenzo[m] has joined #nixos
Obscurity[m] has joined #nixos
bennofs[m] has joined #nixos
equalunique[m] has joined #nixos
mt has joined #nixos
nz has joined #nixos
johs has joined #nixos
aarvar has quit [Ping timeout: 256 seconds]
jsoo has joined #nixos
<Lisanna> Can I set the date in the nix sandbox, in order to increase reproducibility?
jrolfs has quit [Ping timeout: 256 seconds]
<Lisanna> or, I guess the right word is "determinism"
jsoo has quit [Quit: Mutter: www.mutterirc.com]
jsoo has joined #nixos
<MoreTea> Lisanna: that's a bit tricky, since some build programs depend on timestamps (like make), and they can get confused if all the filestamps are the same.
pie___ has joined #nixos
<Lisanna> oh, darn
<MoreTea> and once you allow different timestamps, you have indeterminism in any case, right? because your build will proably not take exactly as long as mine.
<Lisanna> yeah, you're right
<MoreTea> I once made exactly the same suggestion :)
pie___ has quit [Ping timeout: 240 seconds]
rauno has joined #nixos
jsoo_ has joined #nixos
logzet has joined #nixos
jrolfs has joined #nixos
davidlt has joined #nixos
Tucky has joined #nixos
jsoo has quit [Ping timeout: 265 seconds]
slyfox has quit [Quit: no need to cry]
jrolfs has quit [Ping timeout: 256 seconds]
slyfox has joined #nixos
jsoo_ has quit [Quit: Mutter: www.mutterirc.com]
leat has quit [Ping timeout: 256 seconds]
Fare has quit [Ping timeout: 256 seconds]
jsoo_ has joined #nixos
jsoo_ has quit [Client Quit]
mounty has joined #nixos
asuryawanshi has joined #nixos
loonquawl has quit [Quit: loonquawl]
deepfire has quit [Ping timeout: 260 seconds]
georges-duperon has joined #nixos
<{^_^}> [nixpkgs] @bjornfor pushed commit from @ryantm to master « sdcc: 3.6.0 -> 3.7.0 »: https://git.io/vxvZe
<fearlessKim[m]> I want to override the value of deployment.libvirtd.template to debug_domain so I do ` (libvirtd-local args) // { deployment.libvirtd.template = debug_domain; };` but the result seems not like what I expect. How can I see it ? I tried lib.traceShowVal but it just shows "trace: x is attr set { deployment= }"
<{^_^}> [nixpkgs] @bjornfor closed pull request #36685 → sdcc: 3.6.0 -> 3.7.0 → https://git.io/vAhZo
jsoo has joined #nixos
coot has joined #nixos
deepfire has joined #nixos
jsoo has quit [Quit: Mutter: www.mutterirc.com]
thc202 has joined #nixos
jsoo has joined #nixos
zzamboni has joined #nixos
dkao has joined #nixos
<{^_^}> [nixpkgs] @7c6f434c pushed commit from @pjan to master « youtube-dl: 2018.03.03 -> 2018.03.10 »: https://git.io/vxvZ5
Itkovian has joined #nixos
jrolfs has joined #nixos
jsoo has quit [Ping timeout: 276 seconds]
<{^_^}> [nixpkgs] @Izorkin opened pull request #36849 → nftables: 0.8.2 -> 0.8.3 → https://git.io/vxvnf
<{^_^}> [nixpkgs] @7c6f434c closed pull request #36843 → updates youtube-dl → https://git.io/vxvJw
<sphalerite> gchristensen: maybe the topic should be changed to point to samueldr 's logs at https://logs.nix.samueldr.com/nixos rather than botbot, since it still doesn't seem to be fixed?
roberth has joined #nixos
mounty_ has joined #nixos
jrolfs has quit [Ping timeout: 276 seconds]
woffs has joined #nixos
rauno has quit [Ping timeout: 268 seconds]
vidbina has joined #nixos
guibou has joined #nixos
zzamboni has quit [Quit: Leaving.]
zzamboni has joined #nixos
jensens has joined #nixos
civodul has joined #nixos
<Lisanna> sphalerite gchristensen I think it should be changed regardless!
mounty_ has quit [Ping timeout: 260 seconds]
Itkovian has quit [Remote host closed the connection]
rindvieh has joined #nixos
Itkovian has joined #nixos
<{^_^}> [nixpkgs] @peti merged pull request #36781 → ghc841: semi-automated updates → https://git.io/vAj5N
<{^_^}> [nixpkgs] @peti pushed 16 commits to master: https://git.io/vxvc4
<{^_^}> → bb9704ca by @deepfire: ghc841: intern changes to PKGDB, except manual: esqueleto haskell-src
<{^_^}> → 06a19792 by @deepfire: ghc841: gc outdated overrides
<{^_^}> → b77b9d56 by @deepfire: ghc841: deriving-compat released [1/2]
MichaelRaskin has quit [Quit: MichaelRaskin]
hakujin has joined #nixos
hakujin has quit [Ping timeout: 256 seconds]
asuryawanshi has quit [Remote host closed the connection]
asuryawanshi has joined #nixos
<fearlessKim[m]> ok solved it with lib.recursiveUpdate
leat has joined #nixos
<{^_^}> [nixpkgs-channels] @edolstra pushed 17 commits to nixos-unstable-small: https://git.io/vxvcb
<{^_^}> → 35b7632b by @vdemeester: gauge: 0.9.6 -> 0.9.7
<{^_^}> → 82644c15 by @dtzWill: glibc: patch to fix building w/musl-based stdenv
<{^_^}> → 54c4c183 by @dtzWill: glibc: make fetchpatch optional ("? null"), just in case.
tpanum has joined #nixos
tpanum has left #nixos [#nixos]
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/1bfe2511ca1 (from 7 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
orivej has quit [Ping timeout: 240 seconds]
roberth has quit [Ping timeout: 264 seconds]
jrolfs has joined #nixos
reinzelmann has quit [Ping timeout: 245 seconds]
<mpickering> "darwin.apple_sdk.utmp" raises a missing attribute warning? What am I missing, I can see it defined - https://github.com/NixOS/nixpkgs/blob/master/pkgs/os-specific/darwin/apple-sdk/default.nix
jrolfs has quit [Ping timeout: 240 seconds]
<mpickering> ah it is libs.utmp
Itkovian has quit [Ping timeout: 256 seconds]
rauno has joined #nixos
pie_ has joined #nixos
<{^_^}> [nixpkgs] @teto opened pull request #36850 → Virtualisation → https://git.io/vxvWw
reinzelmann has joined #nixos
Izorkin_ has joined #nixos
<Lisanna> ah shoot, I forgot what the magic env variable I could set to change where "nix" treats as the store...
MercurialAlchemi has quit [Ping timeout: 256 seconds]
<Lisanna> e.g., to get nix path-info on a local binary cache
<clever> Lisanna: NIX_REMOTE?
MercurialAlchemi has joined #nixos
reinzelmann has quit [Client Quit]
Izorkin has quit [Ping timeout: 240 seconds]
<Lisanna> clever YES! thank you!
<Lisanna> clever oh cool, you have that convo saved :)
zzamboni has quit [Remote host closed the connection]
<Lisanna> I should probably bookmark that gist
Itkovian has joined #nixos
<{^_^}> [nixpkgs] @volth opened pull request #36851 → [wip] processing3: init at 3.3.6 → https://git.io/vxvl6
Myrl-saki has joined #nixos
zzamboni has joined #nixos
<the-kenny> I keep my nixpkgs as a symlink to a git repository under ~/.nix-defexpr/nixpkgs. This means I have to prefix all packages with the new `nix` tool with `nixpkgs.`. Is there a common way to get around this prefixing? I can accept either no prefix or `pkgs.` as a prefix :)
robstr_ has joined #nixos
<clever> Lisanna: my plan is to go thru every example i scraped off irc, and document how each works
roberth has joined #nixos
pkill9 has joined #nixos
<makefu> the-kenny: wouldn't it be possible to symlink your <nixpkgs> to ~/.nix-defexpr instead?
<the-kenny> makefu: likely, but I'm not sure if this will cause any issues
<makefu> i also saw something like creating a default.nix in this folder with the import 'import <nixpkgs>' as contents
<clever> [clever@amd-nixos:~]$ cat .nix-defexpr/test/foo/default.nix
<clever> import /home/clever/apps/nixpkgs
<clever> makefu: this creates a foo channel, so i can now nix-env -iA foo.hello
<clever> all nix-env cares about, is that it returns a function, which returns a package set, so you can import any nixpkgs you want, fetched any way you want
<clever> or have entirely custom package sets
<{^_^}> [nixpkgs] @rprospero opened pull request #36852 → sasview: init at 4.1.2 → https://git.io/vxv4Y
spear2 has quit [Quit: Leaving]
jrolfs has joined #nixos
thblt has quit [Ping timeout: 260 seconds]
jrolfs has quit [Ping timeout: 256 seconds]
Ivanych has quit [Ping timeout: 248 seconds]
<the-kenny> clever: Thanks! That sounds reasonable. Can you show me your NIX_PATH for completeness?
<clever> the-kenny: no changes at all to it
<clever> the-kenny: nix-env is the only tool that reads ~/.nix-defexpr/ and the structure is un-documented
<the-kenny> ahh
tkral has joined #nixos
rauno has quit [Ping timeout: 265 seconds]
vidbina has quit [Ping timeout: 265 seconds]
__Sander__ has joined #nixos
<makefu> clever: maybe these quirks could be added to the wiki if it is not really fitting the manual
zzamboni has quit [Quit: Leaving.]
zzamboni has joined #nixos
<clever> makefu: that would be a good idea
<clever> i'm also not sure what nix2 has done with ~/.nix-defexpr/
asymmetric has joined #nixos
vidbina has joined #nixos
ixxie has joined #nixos
zzamboni has quit [Client Quit]
zzamboni has joined #nixos
<hyper_ch> hmmm, did something change with spice-gtk? I get: USB redirection error: cloud not redirect ....: Failed to execute child process "/nix/store.....-spice-gtk-0.3.4/bin/spice-client/glib-usb-acl-helper" (no such file or directory)
<hyper_ch> didn't get that before
<clever> hyper_ch: does that path exist?
<hyper_ch> yes
<clever> hyper_ch: and if you run file on it?
<clever> ah, a symlink
<clever> does the target of that symlink exist?
<hyper_ch> it's not in /run/wrappers/bin
<clever> then you need to enable something
<hyper_ch> like?
Izorkin has joined #nixos
MP2E has quit [Remote host closed the connection]
<hyper_ch> ah, spice-gtk wasn't installed
<clever> installing it wont make it setuid root
<clever> you need the config listed in the comment to make it setuid
rauno has joined #nixos
<hyper_ch> ok
Izorkin_ has quit [Ping timeout: 265 seconds]
<hyper_ch> but it did work before
<clever> where you running things as root?
<hyper_ch> no
<clever> something must have changed
<hyper_ch> but I did have this in my config
<hyper_ch> # Make smartcard reader and label printer accessible to everyone, so they can be passed to the VM
<hyper_ch> services.udev.extraRules = ''
<hyper_ch> SUBSYSTEM=="usb", ATTR{idVendor}=="072f", ATTR{idProduct}=="90cc", GROUP="users", MODE="0777"
<hyper_ch> SUBSYSTEM=="usb", ATTR{idVendor}=="04f9", ATTR{idProduct}=="2043", GROUP="users", MODE="0777"
<hyper_ch> '';
<clever> ah
<hyper_ch> do I need to take that out now?
<clever> i'm guessing that previously, the helper to gain root and not need udev was just missing
<clever> so it relied on the udev rules
<clever> you might be able to remove those udev rules now and still have it work
<hyper_ch> and I guess I need to restart the vm?
<hyper_ch> or restart computer?
<clever> yeah
<clever> just the vm
<hyper_ch> rebuild switch didn't give error
vidbina has quit [Ping timeout: 256 seconds]
<hyper_ch> turned off vm
<hyper_ch> now starting it again
<{^_^}> [nixpkgs] @selaux opened pull request #36854 → chromedriver: 2.35 -> 2.36 → https://git.io/vxvRN
<makefu> clever: i've created a first stub https://nixos.wiki/wiki/Nix-env
<hyper_ch> USB redirection error: Could not redirect ACS CCID USB Reader: Device is in use by another application
robstr_ has quit []
<clever> makefu: for a more complete explanation of how things work
<clever> makefu: nix-env will recursively search ~/.nix-defexpr/ until it either finds <name>.nix or <name>/default.nix, and then create a channel with <name>
<clever> makefu: any directories above that (like test in my example) are basically ignored, but allow you to symlink into dirs owned by other users
<makefu> go ahead and refine the text, ive just copy pasted your text :D
<clever> and it will search within it
<clever> i dont have a wiki acct
<makefu> you can use your github acc
orivej has joined #nixos
<hyper_ch> makefu: it works
<clever> makefu: pasted more into it
<makefu> neat!
<makefu> i've cleaned up your input :)
<clever> +1
<clever> behind the scenes, nix-channel will manage a nix-env profile
<clever> so when you add/update a channel, it basically just does nix-env -p <foo> -i /path/to/channel
<clever> [root@amd-nixos:~]# nix-env -p /nix/var/nix/profiles/per-user/root/channels -q
<clever> dummy
<clever> nixos-18.03pre129076.831ef4756e3
<clever> [root@amd-nixos:~]# nix-env -p /nix/var/nix/profiles/per-user/root/channels --list-generations
dan_b has joined #nixos
<Lisanna> argh, shoot, nix-store --delete doesn't even work on binary caches
<clever> 58 2018-02-24 15:04:58 (current)
<Lisanna> that really sucks
<clever> the dummy is somehting i added
<clever> i cant remember how though
jrolfs has joined #nixos
<__Sander__> woohoo
<__Sander__> I managed to fix node-gyp on macOS again
hlavaty has joined #nixos
<__Sander__> it's an incredibly messy codebase
<__Sander__> sigh
<__Sander__> I hope I don't have to touch it again anytime soon
<__Sander__> they pretend that they can "simulate" xcode builds without having xcode installed
<__Sander__> but that's not the case
<__Sander__> there is still a xcodebuild -version invocation somewhere
<__Sander__> I managed to "fake" it
freeman42x]NixOS has joined #nixos
betaboon has joined #nixos
jrolfs has quit [Ping timeout: 265 seconds]
hiberno has quit [Quit: WeeChat 1.6]
winem_ has joined #nixos
ThatDocsLady has joined #nixos
<{^_^}> [nixpkgs] @Mic92 opened pull request #36855 → travis: add update instructions → https://git.io/vxvuP
<adisbladis[m]> __Sander__: node-gyp is such a hack :P
<__Sander__> yes it is
asymmetric has quit [Ping timeout: 248 seconds]
ma27 has joined #nixos
sigmundv has joined #nixos
raynold has quit [Quit: Connection closed for inactivity]
<LnL> I wish we could also just fix it in one place
rardiol1 has left #nixos [#nixos]
ottidmes has joined #nixos
humanoyd has joined #nixos
hiberno has joined #nixos
<{^_^}> [nixpkgs] @zimbatm merged pull request #36855 → travis: add update instructions → https://git.io/vxvuP
<{^_^}> [nixpkgs] @zimbatm pushed commit from @Mic92 to master « travis: add update instructions (#36855) »: https://git.io/vxv2J
<Mic92> more unicode
<sphalerite> I'm trying to get Agda working, but it says the package is marked as broken. I can't seem to find *where* it's marked as broken though, I don't see anyhting to that effect in pkgs/development/haskell-modules/hackage-packages.nix
<sphalerite> Any ideas?
<{^_^}> [nixpkgs-channels] @edolstra pushed 23 commits to nixos-unstable-small: https://git.io/vxv2n
<{^_^}> → 1c99ffcb by @sifmelcara: librime: 1.2.9 -> 1.3.0
<{^_^}> → cbf735e3 by @sifmelcara: brise: change platforms to linux only
<{^_^}> → bb9704ca by @deepfire: ghc841: intern changes to PKGDB, except manual: esqueleto haskell-src
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/a9e1ae3d298 (from 2 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
asymmetric has joined #nixos
jrolfs has joined #nixos
pie_ has quit [Ping timeout: 264 seconds]
<{^_^}> [nixpkgs] @7c6f434c merged pull request #35122 → dar: 2.5.9 -> 2.5.14 → https://git.io/vA8sl
<{^_^}> [nixpkgs] @7c6f434c pushed 3 commits to master: https://git.io/vxv2r
<{^_^}> → 242e7727 by @Izorkin: dar: 2.5.9 -> 2.5.14
<{^_^}> → 3bca698d by @Izorkin: dar 2.5.14: fix build on Darwin
<{^_^}> → 1e4358e5 by @7c6f434c: Merge pull request #35122 from Izorkin/dar-2.5.14
Myrl-saki has quit [Read error: Connection reset by peer]
<aminechikhaoui> gchristensen: not sure if the nixpkgs-channels updates are really useful ^, since there is already a notification when the channel advances
rindvieh has left #nixos [#nixos]
jrolfs has quit [Ping timeout: 240 seconds]
<{^_^}> [nixpkgs] @garbas opened pull request #36856 → Update vim plugins → https://git.io/vxva0
<{^_^}> [nixpkgs] @garbas closed pull request #36829 → LanguageClient-neovim: 2017-12-05 -> 2018-03-06 → https://git.io/vxe7G
<makefu> gchristensen: i really like the new notifications for PRs. maybe we could only keep opened and closed PRs as well as notifications for merges but stip the single commit messages
<{^_^}> [nixpkgs] @Mic92 merged pull request #36783 → resilio: Adjust option description to 3d17573 → https://git.io/vAjdH
<{^_^}> [nixpkgs] @Mic92 pushed 2 commits to master: https://git.io/vxva9
<{^_^}> → 86ca6179 by @Infinisil: resilio: Adjust option description to 3d17573
<{^_^}> → 9881934b by @Mic92: Merge pull request #36783 from Infinisil/desciption/update/resilio
<{^_^}> [nixpkgs] @Mic92 pushed commit from @Infinisil to release-18.03 « resilio: Adjust option description to 3d17573 »: https://git.io/vxvax
mrkgnao has joined #nixos
<gchristensen> Good call amine, I can fix that. Makefu, samueldr made it:) suggest that too him?
zzamboni has quit [Quit: Leaving.]
<etu> Oh, it's not the github bot anymore :D
<etu> nice
<makefu> what did i do? :D
asymmetric has quit [Quit: Leaving]
<sphalerite> makefu: we don't need notifications for pushes to nixpkgs-channels
<sphalerite> err
simukis has joined #nixos
<sphalerite> no what *christensen was saying is you should suggest it to samueldr
jrolfs has joined #nixos
zzamboni has joined #nixos
<{^_^}> [nixpkgs] @svanderburg pushed to master « nodejs: patch bundled node-gyp module to not invoke xcodebuild on macOS causing modules with native dependencies to fail »: https://git.io/vxvwj
jrolfs has quit [Ping timeout: 240 seconds]
zzamboni has quit [Ping timeout: 240 seconds]
<gchristensen> Yeah that
bennofs has joined #nixos
<gchristensen> Or send a PR! https://github.com/samueldr/github-to-irc
zzamboni has joined #nixos
ThatDocsLady has quit [Remote host closed the connection]
hotfuzz_ has joined #nixos
ThatDocsLady has joined #nixos
<infinisil> sphalerite: regarding the broken agda thing, you could add -v to the nix command to see what it evaluates
hotfuzz has quit [Ping timeout: 248 seconds]
<sphalerite> infinisil: I already found it, thanks though
bennofs has quit [Ping timeout: 255 seconds]
zzamboni has quit [Ping timeout: 240 seconds]
fpletz has quit [Changing host]
fpletz has joined #nixos
zzamboni has joined #nixos
fiddlerwoaroof has quit [Ping timeout: 255 seconds]
xcmw has joined #nixos
chisui has joined #nixos
fk_ has joined #nixos
<fk_> Hi. Submitted https://github.com/NixOS/nixpkgs/pull/36335 .. Can someone take a look please?
hlavaty has quit [Remote host closed the connection]
GlennS has quit [Remote host closed the connection]
<{^_^}> [nixpkgs] @maximedenes opened pull request #36857 → pythonPackages.sphinx_rtd_theme: 0.1.9 -> 0.2.5b2 → https://git.io/vxvKN
<Li[m]> question: can I use 'override' and 'overrideAttrs' on the same pkg?
<CrazedProgrammer> yes, you can combine them in any way like for example (package.overrideAttrs (old: { })).override { }
<Li[m]> ahh I see
<Li[m]> thanks
<CrazedProgrammer> np :)
<{^_^}> [nixpkgs] @Mic92 pushed 2 commits to master: https://git.io/vxv6c
<{^_^}> → c10b7a7d by @Izorkin: nftables: 0.8.2 -> 0.8.3
<{^_^}> → 859fe162 by @Mic92: Merge pull request #36849 from Izorkin/nftables-0.8.3
<{^_^}> [nixpkgs] @Mic92 merged pull request #36849 → nftables: 0.8.2 -> 0.8.3 → https://git.io/vxvnf
<{^_^}> [nixpkgs] @Mic92 pushed commit from @Izorkin to release-18.03 « nftables: 0.8.2 -> 0.8.3 »: https://git.io/vxv64
reinzelmann has joined #nixos
<clever> Li[m]: i think override has to go first, so just swap the order from that example
jrolfs has joined #nixos
coot has quit [Quit: coot]
knupfer has joined #nixos
<{^_^}> [nixpkgs] @limeytexan opened pull request #36858 → perlPackages.FileFnMatch: init at 0.02 → https://git.io/vxvir
reinzelmann has quit [Quit: Leaving]
jrolfs has quit [Ping timeout: 245 seconds]
lush has joined #nixos
vidbina has joined #nixos
{`-`} has joined #nixos
<{^_^}> [nixpkgs] @Mic92 merged pull request #36858 → perlPackages.FileFnMatch: init at 0.02 → https://git.io/vxvir
<{^_^}> [nixpkgs] @Mic92 pushed 2 commits to master: https://git.io/vxvPf
<{^_^}> → 1b01d493 by @limeytexan: perlPackages.FileFnMatch: init at 0.02
<{^_^}> → a6c5515c by @Mic92: Merge pull request #36858 from limeytexan/perlPackages-File-FnMatch
<{^_^}> [nixpkgs] @nlewo merged pull request #36826 → houdini 16.0.671->16.5.405 → https://git.io/vxeM5
<{^_^}> [nixpkgs] @nlewo pushed 2 commits to master: https://git.io/vxvPI
<{^_^}> → d13c1f06 by @kwohlfahrt: houdini 16.0.671->16.5.405
<{^_^}> → d4c20ad1 by @nlewo: Merge pull request #36826 from kwohlfahrt/houdini
reinzelmann has joined #nixos
taktoa has quit [Remote host closed the connection]
xcmw has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
georges-duperon has quit [Quit: Leaving]
<{^_^}> [nixpkgs] @svanderburg pushed to release-17.09 « nodejs: patch bundled node-gyp module to not invoke xcodebuild on macOS causing modules with native dependencies to fail »: https://git.io/vxvPb
Shell has joined #nixos
<Shell> Does anyone have any idea how to pick a single .jar out of javaPackages?
<Shell> I can get junit, but everything else seems to require an extra step, and I’m not quite sure what that step is.
<{^_^}> [nixpkgs] @svanderburg pushed to release-18.03 « nodejs: patch bundled node-gyp module to not invoke xcodebuild on macOS causing modules with native dependencies to fail »: https://git.io/vxvXG
knupfer has quit [Ping timeout: 264 seconds]
{^_^} has quit [Remote host closed the connection]
{^_^} has joined #nixos
logzet has quit [Ping timeout: 245 seconds]
logzet has joined #nixos
knupfer has joined #nixos
<{^_^}> [nixpkgs] @limeytexan opened pull request #36859 → perlPackages.PkgConfig: init at 0.19026 → https://git.io/vxv17
<{^_^}> [nixpkgs] @garbas merged pull request #36856 → Update vim plugins → https://git.io/vxva0
<{^_^}> [nixpkgs] @garbas pushed 2 commits to master: https://git.io/vxvMW
<{^_^}> → 31db03ea by @johnchildren: LanguageClient-neovim: 2017-12-05 -> 2018-03-06
<{^_^}> → bf25a020 by @garbas: vimPlugins: updating all vim plugings
zzamboni has quit [Quit: Leaving.]
ma27 has quit [Ping timeout: 252 seconds]
<gchristensen> ok aminechikhaoui, I dropped the messages for nixpkgs-channels, thanks!
ma27 has joined #nixos
<{^_^}> [nixpkgs] @vbgl merged pull request #36857 → pythonPackages.sphinx_rtd_theme: 0.1.9 -> 0.2.5b2 → https://git.io/vxvKN
<{^_^}> [nixpkgs] @vbgl pushed commit from @maximedenes to master « pythonPackages.sphinx_rtd_theme: 0.1.9 -> 0.2.5b2 »: https://git.io/vxvDT
knupfer has quit [Ping timeout: 256 seconds]
<{^_^}> [nixpkgs] @garbas opened pull request #36860 → Update vim plugins for 18.03 → https://git.io/vxvD4
bkchr has joined #nixos
knupfer has joined #nixos
jrolfs has joined #nixos
blankhart has joined #nixos
<bkchr> With the new nix, does the memory usage increase for huge builds? I want to execute a nox-review, but it always dies with "Too many heap sections: Increase MAXHINCR or MAX_HEAP_SECTS". I already tried to set the GC_INITIAL_HEAP_SIZE=8G but that kills my PC, because it uses 14GB RAM. Any ideas what else I coudl change?
xcmw has joined #nixos
jrolfs has quit [Ping timeout: 245 seconds]
<{^_^}> [nixpkgs] @etu closed pull request #36751 → asyncio: move to own directory → https://git.io/vAjGq
<{^_^}> [nixpkgs] @etu closed pull request #36749 → amqp: move to own directory → https://git.io/vAjsg
<{^_^}> [nixpkgs] @etu closed pull request #36737 → Python polib fixes → https://git.io/vAjky
<{^_^}> [nixpkgs] @etu closed pull request #36730 → pymongo_2_9_1: Move to own file → https://git.io/vAjJE
<gchristensen> etu: what's up?
<octe> 1/31
<etu> gchristensen: Just removed some branches. Nothing much. I will come back with new pr's later that are more organized.
<gchristensen> ok, cool
<etu> I was trying to set up a matrix homeserver (with synapse) this weekend and it doesn't work very well :(
<etu> I get internal server errors back when trying to do federation.
pie_ has joined #nixos
<etu> I should probably take it with the guys in #matrix to help me find out what's wrong. But it's probably some dependency issue. Message in journalctl: http://ix.io/W46
hoverbear has quit [Changing host]
hoverbear has joined #nixos
hoverbear has joined #nixos
<gchristensen> X|
<Lisanna> is it possible for multiple .narinfo files to reference the same .nar in their URL field?
<gchristensen> I ... don't think so ...
knupfer has quit [Ping timeout: 264 seconds]
<etu> gchristensen: Yeah, I should also move my only ubuntu install to nixos today :)
<Lisanna> ...well, that's what I've got
<Lisanna> 267drk3isaz2iiksk2pp9lyjbdhb42ys.narinfo and 5ji1qznlg9fj2wvf8halyik3namdzan9.narinfo both have "URL: nar/1ydyak3r62j32rwfb1mdp8hxrdsfglaxwq3k6xp7dxsvsnqca5pl.nar.xz"
<Lisanna> different derivers too...
pie_ has quit [Ping timeout: 260 seconds]
<gchristensen> can you paste the contents of the narinfo?
alexteves has joined #nixos
<Lisanna> gchristensen sure, one sec
<ottidmes> bkchr: Are you sure you are not introducing some recursion somewhere. I have had that error recently, but it was caused by my overlays. In one of my overlays I imported my unstable checkout of nixpkgs, but because how I had setup my overlays, it would apply the same overlays to that nixpkgs, so it would again import the unstable nixpkgs in the unstable nixpkgs because of the overlay, etc.
<bkchr> ottidmes: Hmm. I also have some overlays activated, but `nixos-rebuild` works.
<disasm> sphalerite: do you know how I can get pkgs.python27Packages.websocket_client into the PYTHONPATH for weechat with the new changes? This is what I did before that doesn't work on 18.03.
peacememories has joined #nixos
itzmjauz_ has joined #nixos
knupfer has joined #nixos
<gchristensen> Lisanna: ooh I guess that does make sense, and your outputs are binary reproducible -- congratulations :P
<sphalerite> disasm: the manual explains it
dejanr has quit [Quit: ZNC 1.6.5 - http://znc.in]
jrolfs has joined #nixos
<Lisanna> gchristensen ok, cool
<ottidmes> bkchr: Maybe not that particular problem, but maybe similar kind of recursion is happening with nox-review (or whatever you call when you run into this error)
<Lisanna> figured it was something like that
<disasm> sphalerite: oh, lol, probably should have looked at that first :) sorry!
<bkchr> ottidmes: yeah. I will need to debug that
vidbina has quit [Ping timeout: 248 seconds]
jrolfs has quit [Ping timeout: 256 seconds]
<{^_^}> [hydra] @edolstra pushed to master « Add a hydra.conf option to enable email notification »: https://git.io/vxvHd
<{^_^}> [nixops] @rbvermaa pushed to master « gce: set set_host_keys to false, to prevent google-instance-setup service from replacing the host key on first reboot. »: https://git.io/vxvHb
<{^_^}> [nixpkgs] @NeQuissimus pushed to master « linux: 4.16-rc4 -> 4.16-rc5 »: https://git.io/vxvHA
<gchristensen> hydra and nixops notifications :o
peacememories has quit [Quit: Textual IRC Client: www.textualapp.com]
xcmw has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<sphalerite> \o/
peacememories has joined #nixos
Rusty1_ has joined #nixos
<etu> gchristensen: I like not having github bot joining and leaving all the time.
<gchristensen> all thanks to samueldr!
<etu> Yeah, he seems to have done a good job. :)
disasm has quit [Quit: WeeChat 2.0]
disasm has joined #nixos
peacememories has quit [Client Quit]
iyzsong has joined #nixos
knupfer has quit [Ping timeout: 260 seconds]
xcmw has joined #nixos
disasm has quit [Client Quit]
disasm has joined #nixos
disasm has quit [Client Quit]
<aminechikhaoui> woot \o/
lush has quit [Quit: Konversation terminated!]
mounty_ has joined #nixos
xcmw has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
dejanr has joined #nixos
disasm has joined #nixos
<{^_^}> [nixpkgs] @peti pushed to release-18.03 « hackage-packages.nix: fix gnome3-related evaluation errors »: https://git.io/vxv7j
Poca has quit [Ping timeout: 240 seconds]
<gchristensen> can someone mention me on a Nixpkgs PR? any PR will do
obadz- has joined #nixos
<aminechikhaoui> gchristensen: done
obadz has quit [Ping timeout: 264 seconds]
obadz- is now known as obadz
<aminechikhaoui> huh that PR shows a conflict :D https://github.com/NixOS/nixpkgs/pull/33643
knupfer has joined #nixos
knupfer has quit [Remote host closed the connection]
<etu> aminechikhaoui: You could define a date as version of the package but still use a revision for what to clone, it's a bit nicer to see how old something is
xcmw has joined #nixos
<{^_^}> [nixpkgs] @peti pushed 13 commits to master: https://git.io/vxv59
<{^_^}> → 86058acb by @peti: haskell-ChasingBottoms: drop obsolete override
<{^_^}> → 83de07c6 by @peti: LTS Haskell 10.9
<{^_^}> → 1278dba2 by @deepfire: ghc841: bv fix released
<etu> aminechikhaoui: I usually pick "date-of-commit"
disasm has quit [Quit: WeeChat 2.0]
knupfer has joined #nixos
<gchristensen> thanks aminechikhaoui!
<{^_^}> [nixpkgs] @garbas merged pull request #36860 → Update vim plugins for 18.03 → https://git.io/vxvD4
<{^_^}> [nixpkgs] @garbas pushed 2 commits to release-18.03: https://git.io/vxv5F
<{^_^}> → 11caaec8 by @johnchildren: LanguageClient-neovim: 2017-12-05 -> 2018-03-06
<{^_^}> → d4fc1a0c by @garbas: vimPlugins: updating all vim plugings
Fare has joined #nixos
<aminechikhaoui> etu: ah good idea
disasm has joined #nixos
<etu> aminechikhaoui: And a rebase on current master seems like a good idea to resolve conflicts :p
<Lisanna> yay! My Nix CI system now has a declarative binary cache with garbage collection (suck it hydra!!)
<gchristensen> heck yes!
mounty_ has quit [Ping timeout: 264 seconds]
<etu> Hmm, I think I know what's my issue with my matrix-synapse server is now :)
disasm has quit [Quit: WeeChat 2.0]
disasm has joined #nixos
<etu> Matrix doesn't support SNI. The proxy I have set up requires SNI.
jrolfs has joined #nixos
hlolli has joined #nixos
wrl_ is now known as wrl
BurNiinTRee[m] has joined #nixos
<aminechikhaoui> etu: alright should be fixed now
<Lisanna> ...and right after I finish, eelco points me to perl scripts he wrote forever ago that do the same thing... lol
<etu> aminechikhaoui: You could have amended the change to your previous commit to not have two commits for "one change", it's still possible to squash but requires some more git magic :) And the merge commit could have been avoided with a rebase :)
<gchristensen> every program which will ever exist has already been written in perl, probably
<hlolli> how do I create shell with both runtime and buildtime dependencies, I'm trying to compile app with emscripten and I was hoping that `nix-shell '<nixpkgs>' -A emscripten` could provide my with for example path to $EMSCRIPTEN_ROOT, which would be the buildDeps.
<etu> aminechikhaoui: But I'm not allowed to merge anyways. I'm just a person with opinions on git usage. :p
<monotux> hmm, my nftables rules are evaluated before the network is up, is this really intended? my interfaces are not defined then it seems, so loading the rules fails and has to be done manually (which sucks on a router)
<aminechikhaoui> etu: that's already great :D thanks will do so next time
jrolfs has quit [Ping timeout: 276 seconds]
<aminechikhaoui> niksnut: I thought there was a plan to implement https://github.com/NixOS/nix/issues/1968 with dynamoDB for S3BinaryCache for example, isn't that still the plan ?
<{^_^}> [nixpkgs] @johnchildren opened pull request #36861 → go-langserver: init at unstable-2018-03-05 → https://git.io/vxvbl
<{^_^}> [nixops] @AmineChikhaoui merged pull request #866 → add labels option to be applied to GCE instances → https://git.io/vAezc
<{^_^}> [nixops] @AmineChikhaoui pushed 2 commits to master: https://git.io/vxvbD
<{^_^}> → 8a8d8572 by @AmineChikhaoui: add labels option to be applied to GCE instances, this seems to be
vidbina has joined #nixos
<{^_^}> → 21955d1f by @AmineChikhaoui: Merge pull request #866 from AmineChikhaoui/gcp-labels
szicari has joined #nixos
orivej has quit [Ping timeout: 240 seconds]
Myrl-saki has joined #nixos
coot has joined #nixos
pie_ has joined #nixos
Neo-- has joined #nixos
xcmw has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
AluisioASG has quit [Remote host closed the connection]
<pjan> 1
jrolfs has joined #nixos
xcmw has joined #nixos
blankhart has quit [Ping timeout: 264 seconds]
<gchristensen> 2
justan0theruser has quit [Ping timeout: 276 seconds]
zopsi has quit [Ping timeout: 240 seconds]
r5d has quit [Quit: WeeChat 1.9.1]
<monotux> 3
<gchristensen> uh oh
<clever> 4
<monotux> dammit
<monotux> 5
<monotux> \o/
<monotux> 6
<monotux> ok enough for me
jrolfs has quit [Ping timeout: 248 seconds]
<LnL> ?
__Sander__ has quit [Quit: Konversation terminated!]
zopsi has joined #nixos
erasmas has joined #nixos
jmeredith has joined #nixos
freeman42x]NixOS has quit [Ping timeout: 248 seconds]
asuryawanshi has quit [Ping timeout: 256 seconds]
oida has joined #nixos
hakujin has joined #nixos
pie_ has quit [Ping timeout: 264 seconds]
FRidh[m] has joined #nixos
coot_ has joined #nixos
disasm has quit [Quit: WeeChat 2.0]
disasm has joined #nixos
AluisioASG has joined #nixos
justan0theruser has joined #nixos
coot has quit [Ping timeout: 264 seconds]
disasm has quit [Client Quit]
disasm has joined #nixos
disasm has quit [Client Quit]
drewr has quit [Remote host closed the connection]
disasm has joined #nixos
FRidh has joined #nixos
__Sander__ has joined #nixos
disasm has quit [Client Quit]
disasm has joined #nixos
asuryawanshi has joined #nixos
Shell has left #nixos ["Be back later..."]
zzamboni has joined #nixos
asuryawanshi has quit [Ping timeout: 264 seconds]
nuncanada has joined #nixos
<hyper_ch> who uses znc on nixos?
<hyper_ch> I seem to have znc installed just nicely but no idea how to start it
Synthetica has joined #nixos
drewr has joined #nixos
spietz has joined #nixos
<LnL> use services.znc.*
<ottidmes> LnL: BTW, about your ?, if you check the minutes, you can see the pattern
<LnL> totally didn't notice that :p
jrolfs has joined #nixos
<gchristensen> oh me either!
<hyper_ch> LnL: ps aux | grep znc shows nothting
knupfer has quit [Ping timeout: 255 seconds]
<gchristensen> I wonder if it is socket-activated
jrolfs has quit [Ping timeout: 240 seconds]
<hyper_ch> if I just run "znc" it says command not found
reinzelmann has quit [Quit: Leaving]
<gchristensen> that isn't so uncommon, enabling a service doesn't mean it is added to your PATH
<hyper_ch> but I do have a znc binary in /nix/store/xxxx-znc-1.6.5/bin/znc
<LnL> don't think it's exposed in systemPackages, nothing useful for your shell
<hyper_ch> so, how to setup znc properly then?
<LnL> services that don't have a client binary shouldn't do that IMHO
<LnL> I still have to convert my znc setup to nix, I configured it way back and am scared to touch it now :)
<hyper_ch> LnL: don't be scared... just do it
<hyper_ch> and then tell me how :)
<gchristensen> hyper_ch: you should probably check the logs for the znc service
<hyper_ch> gchristensen: ah, good idea Mar 12 15:50:28 servi znc-start[11549]: Module [bouncedcc] does not support module type [Network].
pie_ has joined #nixos
xcmw has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
knupfer has joined #nixos
<ottidmes> I got a question about builders and substituters, right now it seems to just always error whenever one of them is down (e.g. my desktop is a substituter, but it is not always on), is it possible to just ignore it when it is not accessible, I do not mind a warning, but exiting with an error is kind of annoying for this use case (I keep having to edit the nix.conf)
<hyper_ch> it starts now :)
ixxie has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
stphrolland has joined #nixos
<infinisil> LnL: I could share my just overhauled znc config, most of it in nix
ryanartecona has joined #nixos
<infinisil> But there's loads of znc modules that store their config in the data directory, and you need to configure it by sending irc messages to special nicks..
<infinisil> I'd like for nix to handle those too, but I haven't looked into how I could achieve this
<hyper_ch> LnL: infinisil: this starts znc but I can't test it since router port forwarding isn't activated yet https://paste.simplylinux.ch/view/raw/25d58e61
ma27 has quit [Ping timeout: 245 seconds]
blankhart has joined #nixos
spietz has quit [Quit: WeeChat 1.9.1]
<infinisil> But you still need to run commands to set it up fully
<infinisil> At least SSL requires you to get some certificate (can be from let's encrypt)
blankhart has quit [Read error: Connection reset by peer]
<LnL> yeah, the modules is the important part
<hyper_ch> I set config folder to /home/hyper/.znc so modules should be able to write to there
blankhart has joined #nixos
spietz has joined #nixos
<hyper_ch> -rw------- 1 hyper users 2288 Mar 12 15:58 znc.conf
<hyper_ch> drwxr-xr-x 2 hyper users 3 Mar 12 15:58 configs
<rauno> Hey
<rauno> Is there a way to upgrade NICs driver under nixos?
<rauno> Or the upgrades come only with new kernel update ?
<stphrolland> a few months ago, while I was discussing about login screen applications for nixos, and choosing differently from the default SLiM application, someone on this irc forum told me he wrote one in haskell, but I cannot find who or the github repo in question it in my notes. As far as I remember the author told me that there was no GUI yet. Does someone know who is it ? I would like to have a look at the
<stphrolland> code.
<hyper_ch> grep -i "login" nixos.log
<infinisil> stphrolland: That was probably taktoa[c] with https://github.com/taktoa/hsdm
<infinisil> Last commit 2 years ago, sooo..
MercurialAlchemi has quit [Ping timeout: 265 seconds]
spietz has quit [Client Quit]
<hyper_ch> pretty stable then :)
spietz has joined #nixos
<infinisil> Oh and clever seems to have contributed too
<stphrolland> infinisil: many thanks, that is it, I remember the name hsdm now. Nevermind if not active, I need a seed, a start.
<infinisil> stphrolland: You creating one yourself?
GlennS has joined #nixos
slabity[m] has joined #nixos
ixxie has joined #nixos
<stphrolland> infinisil: trying to understand linux better, trying do something useful in OpenGL for learning... and let's see. Maybe improving the existing or making my own. No plans, just code, learn and see.
detran has quit [Ping timeout: 276 seconds]
<stphrolland> that could be the opportunity
betaboon has quit [Quit: WeeChat 1.9.1]
<Yaniel> a X11 window manager won't be very useful when learning OpenGL
<Yaniel> a compositor maybe
detran has joined #nixos
<Yaniel> although I'd rather go for something like a game or level viewer
<{^_^}> [nixpkgs] @fpletz merged pull request #36394 → screen-message: clean up icon-theme.cache →
<{^_^}> [nixpkgs] @fpletz pushed 3 commits to master:
<{^_^}> → 488f88a7 by @rycee: screen-message: clean up icon-theme.cache
<{^_^}> → 85f8b401 by @rycee: screen-message: fix homepage
<{^_^}> → bd23e4ea by @fpletz: Merge pull request #36394 from rycee/fix/sm/icon-cache
ma27 has joined #nixos
<stphrolland> I was first aiming at learning Wayland, but what I read at the time was heavily based on the OpenGL implementation (if I remember correctly...) so I have to learn OpenGL so as to learn Wayland. And as a personnal taste I'd like a sexy loginscreen, and thought I could achieve that with OpenGL. If not with cairo. But too many things I have not learnt yet, so I have to start somewhere :-)
Ivanych has joined #nixos
<infinisil> stphrolland: Yeah, I don't think a display manager is the best way to learn opengl. One of the hardest part of a display manager is not the GUI afaik
<Yaniel> well the problem with a X11 window manager is that they don't really have to draw anything
<Yaniel> many choose to draw window decorations, though
<roconnor> Upgraded unstable this weekend, plasma-desktop seems to be dead. Any cache I need to clear out?
<musicmatze[m]> Can someone please have a look at https://github.com/NixOS/nixpkgs/pull/32710 ? This is a update we should include in nixos 18.03, too... I'm here and ready to squash the PR, just put a short notice in the PR!
ixxie has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
reber has joined #nixos
<reber> hi all
<roconnor> konsle, okular, krunner, kate and plasmashell all had a general protection fault.
<roconnor> in libQt5XcbQpa.so.5.10.0
<reber> error: opening lock file '/nix/var/nix/profiles/per-user/myuser/profile.lock': Permission denied <-- any idea ?
Jackneilll has quit [Read error: Connection reset by peer]
Sonarpulse has joined #nixos
freeman42x]NixOS has joined #nixos
jrolfs has joined #nixos
Jackneill has joined #nixos
asuryawanshi has joined #nixos
Poca has joined #nixos
Fare has quit [Ping timeout: 256 seconds]
<{^_^}> [nixpkgs] @sveitser opened pull request #36863 → vim-solidity: init at 2018-02-11 → https://git.io/vxfqM
jrolfs has quit [Ping timeout: 255 seconds]
__foo__ has joined #nixos
<octe> how do i find out which package owns a file in the store?
<aminechikhaoui> octe: the store path should have the package name in theory
<octe> ah.. for /nix/store/1bd41waiqgfn8whnxy3yinmp23dyvkdk-firefox-bin-60.0a1/bin/firefox the relevant name was "firefox-bin"
<octe> thanks
<octe> i tried firefox
detran has quit [Ping timeout: 256 seconds]
__foo__ has quit [Client Quit]
<roconnor> org.kde.ActivityManager: Cannot mix incompatible Qt library (version 0x50a00) with this library (version 0x50a01)
<{^_^}> [nixpkgs] @nlewo merged pull request #36861 → go-langserver: init at unstable-2018-03-05 → https://git.io/vxvbl
<{^_^}> [nixpkgs] @nlewo pushed 2 commits to master: https://git.io/vxfm8
<{^_^}> → 79f4548a by @johnchildren: go-langserver: init at unstable-2018-03-05
<{^_^}> → 4be7395b by @nlewo: Merge pull request #36861 from johnchildren/init/go-langserver-unstable
<infinisil> octe: what for?
<mjacob> hi! we want to try nix for our development environment (we got annoyed by different people having different versions of system packages etc.). one developer insists on having everything inside a (s)chroot. is there an easy way of creating a new chroot environment containing only nix?
<niksnut> yes, nix --store /my/chroot ...
<mjacob> niksnut: but how can we bootstrap nix then? the starting point is an operating system with no nix at all
<{^_^}> [nixpkgs] @peterhoeg opened pull request #36864 → nixos ddclient: support multiple domains and run via systemd timer → https://git.io/vxfmh
<niksnut> hm yeah, that's not possible atm
<niksnut> would be a nice feature for the installation script
<{^_^}> [nixpkgs] @peterhoeg opened pull request #36865 → nixos mosquitto: run in foreground so we get logging data → https://git.io/vxfYE
<mjacob> niksnut: what installation script exactly?
oida has quit [Remote host closed the connection]
<mjacob> niksnut: if you tell me how to get started, i might try to add this feature
oida has joined #nixos
<rawtaz> mjacob: sorry but dont you just boot an installation or liveCD or similar?
<mjacob> rawtaz: well, we want nix, but not nixos
frankpf has joined #nixos
loonquawl has joined #nixos
leat has quit [Ping timeout: 256 seconds]
<rawtaz> hmm ok
marusich has quit [Quit: Leaving]
humanoyd has quit [Quit: WeeChat 2.0.1]
<{^_^}> [nixpkgs] @7c6f434c pushed to master « tptp: 7.0.0 -> 7.1.0 »: https://git.io/vxfOp
<clever> niksnut: that is something i wrote last week
<lewo> mjacob: are you sure you want/need to install Nix in a chroot? (FYI, you only need a root account to do mkdir /nix/; chown USER /nix/)
<roconnor> I don't see how it is possible I have references to mixed versions of libQt. There are no old references in my system.
<{^_^}> [nixpkgs] @Ericson2314 opened pull request #36866 → clangUnwrapped: Removed unused abstraction → https://git.io/vxf3Y
<clever> niksnut: this generates an arx bundle containing the closure of nix and a basic shell env, run sh on it to install, and then you have a full non-root mount namespace in ~/nix-install along with a script for entering the namespace
<clever> niksnut: main issue i ran into, is that nix doesnt have a way to serialize a set into a .nix file, so i cant generate a manifest.nix, which means `nix-env -i` wont work right
Poca has quit [Ping timeout: 268 seconds]
chaker has joined #nixos
<monotux> silly question, but I'd like to change a systemd-script (nftables) to have it run _after_ the network is up. how would you do it?
<clever> monotux: i think theres an after field in the systemd config on nixos, for each service
frankpf has quit [Ping timeout: 256 seconds]
<monotux> oh, I'll see what I can find in the documentation then
chaker has quit [Client Quit]
<Lisanna> `lib.unique [ { a = lib.unique; } { a = lib.unique; } ]` => [ {...} {...} ]
<Lisanna> !!!!!
loonquawl has quit [Quit: loonquawl]
knupfer has quit [Ping timeout: 260 seconds]
<Lisanna> `{ a = lib.unique; } == { a = lib.unique; }` => false
<Lisanna> wtf
<manveru> (a: a) == (a: a) #=> false
<Lisanna> x_x
frankpf has joined #nixos
<catern> is there a generic way in nixpkgs to express that two packages can't exist in the same closure?
<{^_^}> [nixpkgs] @rycee pushed commit from @volth to master « perlPackages.ClassAccessorFast: alias of perlPackages.ClassAccessor »: https://git.io/vxfGv
<catern> or I guess, that two packages are incompatible?
<catern> or anything like that in any way?
<{^_^}> [nixpkgs] @rycee closed pull request #36828 → perlPackages.ClassAccessorFast = perlPackages.ClassAccessor → https://git.io/vxe9A
<Lisanna> manveru has it always been this way? this seems... weird
<manveru> pretty sure you can't compare functions
<manveru> let x = (a: a); y = x; in x == y #=> false
loonquawl has joined #nixos
jrolfs has joined #nixos
roconnor has quit [Quit: Konversation terminated!]
tnks has left #nixos [#nixos]
<Lisanna> well that sucks
<Lisanna> ...can't print functions... can't compare functions... bleh
<Lisanna> (I'm sure there are perfectly valid technical reasons for why ^^)
<ocharles> Does anyone know what I can do about https://gist.github.com/ocharles/a4b29fd30b7105f6ad0f8c0520e789d6?
<ocharles> I've written some software that uses GL, but I don't seem to be able to run it
<ocharles> works fine on my laptop, but not this desktop
<ocharles> glxinfo works fine though...
frankpf has quit [Ping timeout: 256 seconds]
<Lisanna> catern what do you mean "incompatible"?
<ocharles> oh hey, removing that LD_LIBRARY_PATH works
<ocharles> I didn't even mean to put that there :)
<catern> Lisanna: two versions of the same library, perhaps
<catern> can't be loaded at the same time, so they indicate there's some programming error
<catern> if they're both in buildInputs/propagatedBuildInputs
<Lisanna> catern I mean, buildEnv will error if you ask it to add two packages that have conflicting files
jrolfs has quit [Ping timeout: 256 seconds]
<Lisanna> not sure how buildInputs handles that sort of thing
<Lisanna> presumably it will do a similar thing
roconnor has joined #nixos
<{^_^}> [nixpkgs] @Ericson2314 merged pull request #36866 → clangUnwrapped: Removed unused abstraction → https://git.io/vxf3Y
<{^_^}> [nixpkgs] @Ericson2314 pushed 2 commits to master: https://git.io/vxfZA
<{^_^}> → c65f0ddb by @Ericson2314: clangUnwrapped: Removed unused abstraction
<{^_^}> → 556eb0f3 by @Ericson2314: Merge pull request #36866 from obsidiansystems/no-clangUnwrapped
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
blankhart has quit [Ping timeout: 245 seconds]
<hyper_ch> LnL: the config that I posted seems to work... I can login to the znc web interface
zzamboni has quit [Quit: Leaving.]
zzamboni has joined #nixos
zzamboni has quit [Client Quit]
iyzsong has quit [Ping timeout: 268 seconds]
<roconnor> ugh, it is only one machine of mine that has qt problems. :/
leat has joined #nixos
<infinisil> Oh My God
<goibhniu> roconnor: is it the "incompatible versions" problem? Do you know how to fix it?
<roconnor> yes. no.
<goibhniu> the easiest thing is to ensure all qt packages are in you systemPackages
blankhart has joined #nixos
<roconnor> I beleive they are.
<goibhniu> even things that you've built with nix-build can cause problems
spietz has quit [Quit: WeeChat 1.9.1]
<goibhniu> if you know the offending version of qt, you can query the store with `nix-store -q --referrers /nix/store/*-qt*-5.9.2` <- your version
jrolfs has joined #nixos
<roconnor> well i have old system profiles that refer to qt 5.10.0
<roconnor> let's check roots
jtojnar has quit [Remote host closed the connection]
<roconnor> interesting. it does claim I have something in my most recent personal profile.
<{^_^}> [nixpkgs] @Mic92 merged pull request #32710 → lumail: 2.9 -> 3.1 → https://git.io/vbrlu
<{^_^}> [nixpkgs] @Mic92 pushed 4 commits to master: https://git.io/vxfCR
<{^_^}> → 29c33ba5 by @matthiasbeyer: lumail: 2.9 -> 3.1
<{^_^}> → bb8e1c45 by @matthiasbeyer: lumail: Fix package definition
<{^_^}> → 13e95f33 by @matthiasbeyer: lumail: Use lua5.1
jrolfs has quit [Ping timeout: 256 seconds]
humanoyd has joined #nixos
<roconnor> goibhniu: um nix-store -q --tree says I have qt-base-5.10.0-bin directly referenced in my most recent profile. nix-env disagrees.
<roconnor> I'm tempted to just rollback my profile.
<roconnor> but I don't see how this is possible.
siva_ has quit [Quit: Page closed]
siva has quit [Quit: Page closed]
<clever> roconnor: one min
<clever> roconnor: something in your profile had qt-base in the propagated-user-env-packages
<clever> which makes nix-env silently act as if you had directly nix-env -i'd qt-base
Poca has joined #nixos
lexokay has joined #nixos
<lexokay> dudes,
Itkovian has joined #nixos
ryanartecona has quit [Quit: ryanartecona]
<roconnor> clever: any idea which one?
<lexokay> whatcha gonna do if source url domain became unavailable in nix-expression from nixpkgs ? Your cache buildmachine and users will be unable to built it and get the app?
<clever> roconnor: check the entire nix-store -q --tree ~/.nix-profile/ to see what depends on that version
<lexokay> *for particular package
xcmw has joined #nixos
<{^_^}> [nixpkgs] @bkchr opened pull request #36868 → kde-frameworks: 5.43 -> 5.44 → https://git.io/vxfWl
<{^_^}> [nixpkgs] @Mic92 pushed 3 commits to release-18.03: https://git.io/vxfW8
<{^_^}> → a2150996 by @matthiasbeyer: lumail: 2.9 -> 3.1
<{^_^}> → 556857bc by @matthiasbeyer: lumail: Fix package definition
<{^_^}> → c568f4d2 by @matthiasbeyer: lumail: Use lua5.1
<roconnor> clever: got it! Apparently I have bitcoin-qt installed in my profile, instead of bitcoind
<roconnor> goibhniu: whew. Thanks for helping me here.
<roconnor> It really was a wayward user-env package.
<hyper_ch> LnL: now I got znc also running with letsencrypt
<roconnor> Well, I still need to reboot and try it.
<roconnor> but it seems promising.
<goibhniu> roconnor: great that you got it working! It's a tricky problem :/
<roconnor> lexokay: isn't this a problem for all source-based package managers?
<{^_^}> [nixpkgs] @bkchr opened pull request #36869 → kde-applications: 17.12.2 -> 17.12.3 → https://git.io/vxflf
<{^_^}> [nixpkgs] @bkchr opened pull request #36870 → plasma5: 5.12.2 -> 5.12.3 → https://git.io/vxflq
<catern> lexokay: there is a tarballs.nixos.org mirror of all the tarballs used by nixpkgs
<lexokay> roconnor, well, I just think there must be distribution solution, like distro repo server that clones all sources of all packages and keeps them. that would be protection from such failure where upstream remove source or abandon domain name
<roconnor> catern: surely only those build by hydra.
<catern> roconnor: yes
<roconnor> not quite the same as everything in nixpkgs ;)
<catern> nitpicks, it's close enough
<catern> lexokay: it doesn't have an HTML interface for end-user usage
Itkovian has quit [Read error: Connection reset by peer]
<lexokay> catern, well fuck that in that case
<catern> ????
<lexokay> how could be useful it then?
<catern> do you know the difference between HTML and HTTP?
<lexokay> ML is for markup language, and TP is for text protocol
<catern> yes
dan_b has quit [Ping timeout: 260 seconds]
<catern> there is no markup language interface
<catern> just the protocol for getting files
<catern> which Nix fetchurl uses
<catern> under the hood
<catern> if you want to use tarballs.nixos.org yourself, look at the source of Nix fetchurl and do the same thing
roberth has quit [Ping timeout: 245 seconds]
<roconnor> goibhniu: unbelievably this didn't fix the problem.
Fare has joined #nixos
<lexokay> I'm just thinking about caching (source and binary) solution for abandonware case, let me bring to the table real world example https://github.com/NixOS/nixpkgs/blob/1dcd022f01b251b1656f349dcf749c0890de2799/pkgs/applications/audio/caudec/default.nix#L34
<roconnor> goibhniu: wait, somehow i didn't apply this fix.
<lexokay> the upstream url AFAIK unavailable, and domain probably changed
sanscoeur has joined #nixos
Itkovian has joined #nixos
<lexokay> atm I could install nixos.caudec package from binary cache, but I doubt it could be built from the source now (prove me wrong)
<roconnor> ah there was an error in my nix-env command
<Mic92> Does somebody has an example project for nixos vm tests + travis ci?
fiddlerwoaroof has joined #nixos
<goibhniu> lexokay: do we really want abandoned software in nixpkgs?
<jluttine> i'm using spacemacs for haskell programming. some error checker complains about an import that the module could not be found. however, it is installed in the environment and if i launch haskell repl in spacemacs, it loads the module just fine. any ideas what could be wrong?
<roconnor> goibhniu: whew okay it really works now!
robstr has joined #nixos
<srhb> lexokay: Kind of ruins the example, but... nix-build -E 'with import <nixpkgs> {}; fetchurl { url = "http://caudec.net/downloads/caudec-1.7.5.tar.gz"; sha256 = "5d1f5ab3286bb748bd29cbf45df2ad2faf5ed86070f90deccf71c60be832f3d5"; }'
<lexokay> goibhniu you maybe (or someone else) don't want. But I do want support ability to run proprietary abandonware in linux
la_putin has joined #nixos
<srhb> I'm assuming that's a mistake though :P
<la_putin> when zsh itself executes it loads /usr/lib/zsh/5.4.2/zsh/newuser.so then executes setup_ and finish_ right?
<goibhniu> roconnor: \o/
<roconnor> nixpkgs doesn't have to be the only repository of nix packages.
<roconnor> this is why channels exist.
<monotux> https://hastebin.com/rinexicemo.pl <-- how do I make sure that before and wants lists are _not_ merged with the default definitions? I'd like them to be empty
davidlt_ has joined #nixos
<lexokay> srhb hmm, thank you
<catern> roconnor: does anyone actually use channels to provide multiple repositories of nix packages
<catern> why would anyone do that rather than provide overlays
<roconnor> no, but it would be awesome.
<srhb> lexokay: You may want to google tarballs.nixos.org also, for information on how/why this works.
xcmw has quit [Ping timeout: 276 seconds]
<srhb> (In most cases, anyway, not this specifically)
<roconnor> why do you mean by an overlay?
<catern> nixpkgs supports overlays
davidlt has quit [Ping timeout: 240 seconds]
<goibhniu> lexokay: ah, I doubt we can keep proprietary code on tarballs.nixos.org
<monotux> anyone? I guess that it's a pretty common issue :/
<srhb> monotux: I forget how lists merge, but usually only the highest priority definitions are merged for attribute sets, perhaps it's the same?
<srhb> monotux: (In which case, lib.mkForce [])
<lexokay> FOSS tarballs would be fine, if only they would have end-user browser fronted UI
<roconnor> oh, overlays the idiom I introduced into nixpkgs-haskell from Dave Cunningham. Yay it is nixpkgs wide.
<roconnor> catern: yes, you'd probably use an overlay, but distribute binaries through a new channel.
<monotux> srhb: ah, I'm trying to do this in my system configuration if it matters
<{^_^}> [nixpkgs] @vbgl opened pull request #36871 → clingo: fix URL → https://git.io/vxfRT
<la_putin> join kde
<srhb> monotux: It doesn't. Not if they merge that way anyway. :)
hyper_ch2 has joined #nixos
robstr_ has joined #nixos
<monotux> hm lib seems to be out of scope, I guess I have to load it somehow
<srhb> monotux: Add it to your configuration.nix function.
<hyper_ch2> LnL: so, I have now working znc config :)
<hyper_ch2> shall I repaste?
<srhb> monotux: That is, the { config, ... }: bit -> { config, pkgs, ... }:
<srhb> Er
<srhb> config, lib, ...
sigmundv has quit [Ping timeout: 245 seconds]
<monotux> srhb: oooh thanks!
<srhb> I'm afraid listOf str doesn't merge that way though.
<srhb> but try :)
<srhb> Oh, it does.
<monotux> it seems to have worked!
hyper_ch2 has quit [Client Quit]
hyper_ch has quit [Quit: ZNC - http://znc.sourceforge.net]
<srhb> monotux: So essentially, each value has a priority attached and only the highest priority values (of the same priority) end up in the final list.
<monotux> thanks for the help, this has been bugging me for some time
hyper_ch has joined #nixos
<srhb> monotux: mkForce is basically just mkOverride (some high priority) value
Fare has quit [Ping timeout: 245 seconds]
hyper_ch has quit [Client Quit]
<Baughn> So I want to test the impact of an stdenv change... What's recommended?
<Baughn> I could quite possibly rebuild all of nixpkgs, but it'd take a week.
<clever> Baughn: what exactly are you changing?
hyper_ch has joined #nixos
<Baughn> I'd like to run the build phases in subshells.
hyper_ch has quit [Client Quit]
<clever> Baughn: ive found that some builds set bash vars in each phase, and expect it to spread to the next phase
hyper_ch has joined #nixos
jrolfs has joined #nixos
<clever> so i would expect that to break a lot
<Baughn> Ahh. Yes, that's one of the things I wanted to check.
<Baughn> Never mind, then.
<manveru> what's the benefit of subshells? making sure the cwd doesn't change?
<clever> manveru: pushd and popd can manage that without subshells
<Baughn> Isolation in general. A more predictable environment.
ryanartecona has joined #nixos
<manveru> if anything, i would like to have each phase in a derivation :)
<manveru> but that'd be almost impossible...
<clever> manveru: at one point, the chrome derivation had its unpack&patch phase in a dedicated derivation
<hlolli> is there no way to use `nvm` node version manager on nixos? I have multiple cases where I need to jump between versions, even within major versions of node. But I believe the problem is libstdc++6.so not being found.
<srhb> manveru: You're touching upon some of the "build-caching" things that have been explored by some people already
<clever> manveru: it took hydra 12 hours to transfer the unpacked source from a build slave back to the master
winem_ has quit [Ping timeout: 265 seconds]
<clever> manveru: and hydra doesnt schedule things to slaves that have given paths, so it took another 12 hours to push it back out
<manveru> hlolli: usually you use nix-shell for that
<gchristensen> clever: maybe fixable with that prefer-local-build thing?
coot_ has quit [Quit: coot_]
<clever> gchristensen: hydra cant do local builds
<hlolli> manveru: yes, I'm most likely tripping on beginner pitfalls. But as an easy question, how would I run node 9.8.0 now on nixos?
jrolfs has quit [Ping timeout: 264 seconds]
<manveru> i think the latest in nixpkgs is 9.7, so you'd make your own version
<hlolli> 9.7 should be fine, but I added nodejs nixos-rebuild switch, and I had version 6x.
<clever> nodejs nodejs-4_x nodejs-6_x nodejs-8_x nodejs-9_x
<Baughn> What version is NixOS are you running?
<clever> hlolli: add one of these 5 keys to your configuration.nix
<hlolli> nixos-version
<hlolli> 17.09.3146.084445b8f38 (Hummingbird)
gchristensen changed the topic of #nixos to: Share the output of nix-shell -p nix-info --run nix-info to help us help you. || https://nixos.org || Latest NixOS: https://nixos.org/nixos/download.html || Latest Nix: https://nixos.org/nix/download.html || Logs: https://botbot.me/freenode/nixos/ || #nixos-dev, #nix-darwin, #nixos-aarch64, #nixos-chat
<hlolli> I tried nodejs-9_x and it didn't find it, so old nixos could explain it?
<clever> hlolli: yeah, that could explain it
<infinisil> That's a long topic gchristensen!
<cransom> heh. hydra was super chatty over the weekend :(
<gchristensen> infinisil: its been this long forever =)
<Baughn> hlolli: That was released quite recently, wasn't it? Are you sure you want to use 9.x?
<hlolli> so how does one update nixos, I tried nixos-channel --update... is that the way to do that?
hyper_ch has quit [Changing host]
hyper_ch has joined #nixos
<Baughn> You'd have to switch to nixos-unstable.
<hlolli> node 9.x has atomics and sharedarraybuffer, which Im playing with
<Baughn> Or wait another few weeks for the next NixOS release.
<clever> hlolli: what channel does `nix-channel --list` say your on?
<infinisil> Ah right, and you can do /topic to view it, always forget about that (otherwise it only shows at the top of my client)
<hlolli> clever: it's empty hmm..
<clever> hlolli: as root
hyper_ch has left #nixos ["Konversation terminated!"]
<hlolli> here it is: nixos https://nixos.org/channels/nixos-17.09
<roconnor> pierron: Are you Nicolas?
<pierron> roconnor: yes
hyper_ch has joined #nixos
<hyper_ch> LnL: I successfully migrated znc to nixos :)
<roconnor> pierron: I'm watching your nix-con 2017 talk of overlays. It's great that it is come so far. I thought you might be interested in the early history of overlays.
<infinisil> hyper_ch: Nice
<hlolli> symphorien: that's a FAQ, nice, I guess I'll be asking more FAQ, but the right question is sometimes tricky to determine :)
Itkovian has quit [Ping timeout: 240 seconds]
lexokay has quit [Quit: Page closed]
<infinisil> hyper_ch: yeah you posted it before too
<hyper_ch> LnL: infinisil: channels with just # don't need to have it in the channel list, channels with two ## like ##rtorrent need 1 # like #rtorrent
<symphorien> Don't be afraid to ask, even if it is in the faq
<Myrl-saki> Uhhh
<hyper_ch> infinisil: nah, it's changed :)
<infinisil> hyper_ch: yeah i know
<pierron> roconnor: early history?
rauno has quit [Remote host closed the connection]
<mjacob> lewo: the other developer has it's development environment in a chroot anyway. what he did now was to install a normal debian in the chroot (schroot can do it automatically) and use the official nix installer script
<Myrl-saki> I have a 4-core system and I nix-built -j5 for a Haskell project.
<Myrl-saki> Err
<Myrl-saki> -j20
<roconnor> pierron: It started in 2014 when I was trying to understand how _override worked in nix because it seemed impossible. I wrote up http://r6.ca/blog/20140422T142911Z.html after understanding it.
[0x4A6F] has joined #nixos
<Myrl-saki> Wait, let me check my load statistics.
<hyper_ch> infinisil: so what problems did you have with znc?
<Baughn> I wonder what would happen if I run the nix installer work steam-run...
<infinisil> hyper_ch: lots of modules have stateful configs
<roconnor> pierron: I was discussing this with Dave Cunningham at the time and he came back to me with an "object-oriented" version of this override methodology.
<hyper_ch> infinisil: don't they write it into the znc config folder that you provide?
<roconnor> pierron: I packaged it up in a small library called oop.nix and introduced it @ https://nixos.org/nix-dev/2014-May/013272.html
<hyper_ch> infinisil: from what I see only znc.conf does get regenerated
<roconnor> pierron: the link to oop.nix there isn't availabe, but you can find the original library today @ http://r6.ca/oop.nix
<infinisil> hyper_ch: Only if you have mutable = true;
<infinisil> it only gets written there if mutable = true; i mean
<hyper_ch> infinisil: I did set config to be in /home/hyper/.znc
<hyper_ch> well, works for me now
<pierron> roconnor: do you know if this is what inspired Mathnerd?
<infinisil> hyper_ch: Yeah but it won't persist
<hyper_ch> I wouldn't know what module would be needed
<pierron> roconnor: or peti for doing the haskell packages?
<hyper_ch> I'm a very basic znc user
<roconnor> pierron: this didn't go far, then Peter Simons was redoing the haskell packages. I suggested using oop.nix: https://nixos.org/nix-dev/2014-December/015197.html
<roconnor> at first he rejected it, because it was unortodox @ https://nixos.org/nix-dev/2014-December/015281.html
<hlolli> symphorien: thanks, it worked, nice to have unstable in let binding and refer to is specifically!
<infinisil> hyper_ch: Ah i confused it a bit there
<infinisil> there are settings in the global config file
<hyper_ch> infinisil: :)
<infinisil> but there are also module settings that are stored deep down in the directory structure
<infinisil> those you can't set with nix
<infinisil> (with the current znc module)
<hyper_ch> well, I don't need that probably
<mjacob> lewo: but there's actually a point in keeping it separate. last time i installed nix on my machine it ended up mixing my system's PATH with nix's PATH, which was very annoying
<infinisil> hyper_ch: I thought so too in the beginning but it's actually really useful, some modules are nice
__Sander__ has quit [Quit: Konversation terminated!]
<pierron> roconnor: Nice to be able to trace that down :)
<pierron> roconnor: and good preliminary work!
jensens has quit [Ping timeout: 268 seconds]
<pierron> roconnor: now we have to rewrite Nixpkgs, to use a single fix-point :P
<infinisil> pierron: +++++
<infinisil> I approve
<roconnor> eventually peter accepted it @ https://nixos.org/nix-dev/2015-January/015587.html
<Myrl-saki> I have a 4C4T i5 and an SSD, should I nix-build -j8 on a Haskell build? Often times, on -j4, when I tab to the terminal, it's in the configure phase, so I don't think it's CPU bound?
jrolfs has joined #nixos
<pierron> infinisil: That's the first time I got such a quick approval on such a bold statement.
<hyper_ch> infinisil: I've used znc for a long time... I just need basic bnc function
<roconnor> pierron: it seems that slowly this idea of Dave's has been working it's way through the nixpkgs ecosystem. I don't know how to trace it from then to today because I haven't been paying close enough attention.
<infinisil> Myrl-saki: haskell dependency trees often have some tight spots where lots of things are waiting for a big subtree to finish
<clever> Myrl-saki: the dependency tree doesnt always allow it to do things in parallel
JosW has joined #nixos
<roconnor> I didn't realize it was already so far reaching
<Myrl-saki> clever: I'm more concerned about the configure phase taking longer than the build phase.
<pierron> roconnor: Nixpkgs rewrite to use the extension mechanism is from 2015, IIRC, after the failed attempt of Mathnerd to get it accepted.
<Myrl-saki> Is the configure phase CPU extensive?
ambro718 has joined #nixos
<infinisil> pierron: Hmm, wait what do you mean by that exactly though? I'm actually not thinking of a single fixed point, but using submodules (and therefore more fixed points) for all sorts of stuff in nixpkgs, packages, functions, etc.
<Myrl-saki> s/extensive/intensive/
<clever> Myrl-saki: i think it depends a lot on the package
<Myrl-saki> I may have had too much coffee right now.
spietz has joined #nixos
<Myrl-saki> clever: What affects the configure phase?
<ambro718> What do I need to do to make nixops work on nixos-unstable which uses Nix 2.0? I get this error in deploy: error: attribute 'nixosVersion' missing, at ...
<Myrl-saki> clever: I mean the CPU usage of the configure phase
<pierron> infinisil: I mean a single fix-point to have a single syntax for writting/extending packages, and no more override* function
<clever> Myrl-saki: for autoconf based projects, there is a lot of useless checking of things like the size of an int, which the stdenv could be caching
<srhb> ambro718: Build a newer version or shim in the missing modules yourself.
<infinisil> pierron: I'd like to use the module system to pass arguments to functions, to configure packages, configs for packages
<clever> Myrl-saki: for haskell projects, there isnt much in the configure phase
<pierron> infinisil: multiple fix-point would be a nightmare.
<ambro718> srhb: which version?
<Myrl-saki> clever: Right, but it's taking some time, around 2-5 seconds.
<Myrl-saki> Wait, let me run this on -j1 to make sure.
<srhb> ambro718: (at least)
<infinisil> pierron: Hmm, couldn't the module system allow for doing this override business?
<Myrl-saki> Oh wait, I think it's setupPhase that's taking a long time.
<ambro718> okay I'll try nixops master, thanks
<pierron> infinisil: the module system is powerful, but costly.
<pierron> infinisil: It has a big memory impact for keeping track of each option.
j6c667f has joined #nixos
<pierron> infinisil: you will probably not want it for handling every attribute in Nixpkgs.
<{^_^}> [nixpkgs] @7c6f434c merged pull request #36871 → clingo: fix URL → https://git.io/vxfRT
<{^_^}> [nixpkgs] @7c6f434c pushed 3 commits to master: https://git.io/vxfg4
<{^_^}> → eedc1732 by @vbgl: clingo: fix URL
<infinisil> pierron: Yeah this might be a problem..
<{^_^}> → 903f4dd3 by @vbgl: clingo: fix build on non-sandboxed darwin
<{^_^}> → fb5815ba by @7c6f434c: Merge pull request #36871 from vbgl/clingo-fix-url
jrolfs has quit [Ping timeout: 268 seconds]
<clever> Myrl-saki: ah, one problem ive noticed, nixpkgs wants to compile Setup.hs, and the linking can be rather slow
<Baughn> Is 18.03 going to include Nix 2.0? I notice it's still marked as unstable.
<pierron> infinisil: what I proposed is SOS, which implies a single fix-point, and a single syntax for override, and the same syntax for packages definitions.
hotfuzz has joined #nixos
<infinisil> pierron: And this performance problem is still there for even a small amount of included modules?
<Myrl-saki> clever: Is there a point of improvement on that?
<srhb> Baughn: Yes, and it's released already, despite the funny numbering
<clever> Myrl-saki: potentially, we can just runhaskell Setup.hs
<Myrl-saki> clever: I'm not familiar with the Haskell build system, much more with Haskell+Nix.
<pierron> infinisil: yes, this overhead is per-option.
<srhb> Baughn: (I think the numbering is mostly a hydra artifact, really...)
<infinisil> pierron: Hmm.. Okay I had some pretty crazy idea that could solve that
<Myrl-saki> clever: Thanks.
aarvar has joined #nixos
hotfuzz_ has quit [Ping timeout: 264 seconds]
<{^_^}> [nixpkgs] @phryneas opened pull request #36872 → gitkraken: 3.3.4 -> 3.4.0 → https://git.io/vxf2U
<clever> Myrl-saki: compileBuildDriverPhase is what compiles it
<pierron> infinisil: Before you attempt any rewrite of the module system, please look carefully at what it does ;)
<infinisil> pierron: And this idea is to build an interpreter in nix for an almost nix-like DSL. This DSL could be turned into a "compiled" form of nix code.
<pierron> infinisil: many tried …
<pierron> infinisil: you mean a JIT ?
<infinisil> pierron: I am fairly familiar with the module system :)
<Myrl-saki> clever: runHaskell is interpretation though, right?
<pierron> infinisil: are you sure? :P
<infinisil> pierron: Nah not a jit
<mjacob> i have a problem with "nix-channel --update": "warning: unable to download 'https://nixos.org/channels/nixpkgs-unstable': Problem with the SSL CA cert (path? access rights?) (77); retrying in 283 ms"
<Baughn> Why in *Nix*?
<clever> Myrl-saki: yes, but 99% of the time, Setup.hs is 2 lines: import Distribution.Simple ; main = defaultMain
<j6c667f> pierron: and this performance problem is still there for even a small amount of included modules?
<Myrl-saki> clever: Right.
<clever> Myrl-saki: i suspect (but havent confirmed) that the defaultMain is already compiled and will run fast even in runhaskell
<Baughn> mjacob: Are you by chance on OSX?
<Myrl-saki> clever: OTOH, what is Setup.hs?
<mjacob> Baughn: no, arch linux actually
<clever> Myrl-saki: its a wrapper around cabal, that allows a package to extend cabal
<Baughn> mjacob: Ah. They're usually the ones with the cert errors. Well... :X
<Myrl-saki> clever: Ah.
<Baughn> mjacob: When was the last time you updated?
<clever> Myrl-saki: the 2 line default i gave above compiles into a binary that is basically identical to cabal, so you can just `./Setup configure` or `./Setup repl`
<Myrl-saki> Also
<Myrl-saki> clever: Ahh
<infinisil> pierron: Kinda hard to explain what I mean, don't have a lot of time right now either
<mjacob> Baughn: on this specific machine, i just installed it
<Myrl-saki> clever: That makes more sense.
j6c667f has quit []
<mjacob> Baughn: curl '<that url>' work BTW
<clever> Myrl-saki: but a custom Setup.hs could do extra stuff in the configure
<roconnor> pierron: BTW, I disagree with using super.writeText over self.writeText. IMHO, you should basically always use self.foo because you want to access the late-bound version of the package so that later overalys can override. Ony when overriding a package is super approprate.
<pierron> infinisil: maybe later ;)
<infinisil> pierron: But in addition to being able to "compile" an almost nix-like DSL, it allows for typing nix too!
<Baughn> mjacob: The Nix tools generally try to be as isolated as at all possible. It won't use the system CA store.
<pierron> roconnor: the reason for that is the security-update branch.
<Baughn> mjacob: The error is odd. This is probably the time when I'd pull out `strace -ffe open`.
<roconnor> pierron: But, based on the comments at the end of your talk, it seems to be an issue of great debate.
<infinisil> pierron: And by typing i mean actual typing
<infinisil> without any additional nix syntax
<Baughn> Which may or may not work these days.
<Myrl-saki> clever: What things can it be used for? Getting hardware info to optimize the binary?
<pierron> roconnor: which gives a new semantic to the self:super: names, which are more in line with the way packages are built.
<roconnor> oh
<clever> Myrl-saki: i think things like native dependencies that cabal doesnt understand
civodul has quit [Quit: ERC (IRC client for Emacs 25.3.1)]
<la_putin> is there anywhere i can view detailed patch info for packages, like what needed to be changed in order to be compatible with nix os
<Myrl-saki> clever: Oh
<clever> Myrl-saki: ive seen one package use Setup.hs to pre-download and build a c++ project from source, so the ffi code could link into it
<clever> "your package manager is crap, let me do that for you"
<Myrl-saki> Also, I actually found a nice use-case for Nix.
<clever> "oh, your package manager turned off the network"
<{^_^}> [nixpkgs] @rycee pushed commit from @ryantm to master « bibutils: 5.6 -> 6.2 »: https://git.io/vxf2F
<ambro718> I can't get nixops master working, I changed the nixopsUnstable derivation in nixpkgs to use fetchgit, but it first gives me some version error then later fails with a mysterious error that also involves the version. https://paste.ubuntu.com/p/5GbdHBSMpX/
<pierron> infinisil: sounds like a JIT to me.
<{^_^}> [nixpkgs] @rycee closed pull request #36474 → bibutils: 5.6 -> 6.2 → https://git.io/vANab
<mjacob> Baughn: doesn't try to open any certs
<clever> ambro718: an issue is already open on nixops, and i think Baughn commented on it
<Myrl-saki> I have 2 "packages"(they should really be one, but it's better to keep them separate.) I can make a "meta-package" that compiles the both of them and then places the results in a single directory.
<Myrl-saki> I used to use a build.sh, now I can just use a default.nix.
jrolfs has joined #nixos
<pierron> roconnor: If you have not followed the security update PR, I highly recommend you from reading the introduction of it.
<Baughn> mjacob: Yes, figures. That trace obviously doesn't include the curl invocation. I figure it's being run as a nixbld user.
<Myrl-saki> It was a cabal+stack+bash build, now it's just cabal+nix.
<roconnor> pierron: link?
<Baughn> So, via RPC rather than forking
<infinisil> pierron: short explanation: building an AST in nix with functions that look like the real thing, checking that ast with a type system, then running a nix generator over that to produce faster code without all type checks
<srhb> ambro718: fwiw the quick build of grabbing the nixops master branch and just doing nix-build release.nix -A build.x86_64-linux
<infinisil> ANd yeah it can be used as an interpreter too
<srhb> ... should work
<infinisil> gotta go now
<Baughn> mjacob: Could you try `/nix/store/*-curl-*-bin*/bin/curl https://nixos.org/` ?
<ambro718> I am not doing that, want it to be defined in nixpkgs
<srhb> ambro718: OK :)
spietz has quit [Quit: WeeChat 1.9.1]
<Baughn> mjacob: Uhm, hopefully that doesn't catch more than one curl installation. If it does, pick one at random.
<{^_^}> [nixpkgs] @jfrankenau opened pull request #36873 → bcal: 1.7 -> 1.8 → https://git.io/vxfaR
<mjacob> Baughn: that's what's in /nix/store: https://paste.pound-python.org/show/l3mIMZELynMPHtsN8Vh3/
<mjacob> Baughn: is there a way to enable debug logging?
rogue_koder has joined #nixos
<ambro718> I found this post about my issue https://nixos.org/nix-dev/2016-June/020882.html I still don't know how to fix it
<ambro718> "build using release.nix"
<{^_^}> [nixpkgs] @rycee pushed commit from @pukkamustard to master « flyway: refactor supported platforms »: https://git.io/vxfaD
<{^_^}> [nixpkgs] @volth reopened pull request #36147 → [WIP] perl-packages.nix updated to actual CPAN versions [staging] → https://git.io/vA1rS
<{^_^}> [nixpkgs] @rycee closed pull request #36432 → flyway: refactor supported platforms → https://git.io/vAbgf
<ambro718> this is my nix expression: https://paste.ubuntu.com/p/qGCfzPbgrN/
<Baughn> mjacob: Not sure. Honestly, I'm not sure what the state of your system is. That listing was way too short, so we're probably talking about a bootstrap scenario...
stphrolland has quit [Quit: leaving]
jrolfs has quit [Ping timeout: 264 seconds]
<Baughn> mjacob: Did the Nix installer finish successfully? Either way, it might be an idea to delete /nix and reinstall.
<Baughn> Though if the standard Nix installer now includes build users (and I think it does) then there's extra cleanup to be done. Which may or may not matter.
<mjacob> Baughn: yes, i tried that several times. i have another idea, let me try this...
<srhb> ambro718: That expression is not made to just build using the git sources, but a preprocessed tarball that hydra produced.
<robstr_> hey, my current haskell/stack configuration looks like this https://gist.github.com/rsoeldner/b6f3afa4a6650e78b2ee5e0e33bcaf8d but because of the current stack bug I want to use stack from unstable channel, is there an easy way ?
<ambro718> srhb: I get that, but I don't understand how to make it work
<srhb> ambro718: Essentially you should reproduce what the release.nix file does, but in-tree.
<Baughn> ambro718: The current workaround to make nixops work is to checkout git and run it from dev-shell.
<Baughn> ambro718: It sucks.
<{^_^}> Channel nixos-17.09-small advanced to https://github.com/NixOS/nixpkgs/commit/c7f415b5225 (from 6 hours ago, history: https://channels.nix.gsc.io/nixos-17.09-small)
<Baughn> I'm not sure why fixing that one is taking so long, honestly.
<srhb> Baughn: It's just a convoluted way of building it.
<srhb> Hydra builds the source tarball which the in-tree expression can build
<srhb> But the hydra build hasn't been operational since 2017 sometime.
[0x4A6F]1 has joined #nixos
<Baughn> srhb: Sure, point is the version in nixpkgs doesn't work with nixpkgs.
<srhb> So no tarball
<srhb> Baughn: Yes, I'm explaining why it's troublesome :)
Fare has joined #nixos
<{^_^}> [nixpkgs] @lukeadams opened pull request #36874 → sambaMaster: 4.8.0_2018-01-25 -> 4.8.0_2018-03-09 → https://git.io/vxfV6
<srhb> Personally I just have an overlay callPackage the release.nix with the appropriate attributes.
<srhb> No need for a dev shell.
[0x4A6F] has quit [Ping timeout: 264 seconds]
[0x4A6F]1 is now known as [0x4A6F]
<srhb> ie. self: super: { nixops-git = (self.callPackage /home/sarah/src/nixops/release.nix {}).build.${builtins.currentSystem}; }
<srhb> I would much prefer if the in-tree expression just worked with the source tree, but alas...
<Baughn> Solution: Fix it, then add a test.
<Baughn> Seems like I may need to do so myself. :P
<Baughn> I also want nixops to use `nix build` instead of `nix-build`, so..
<srhb> Baughn: How come?
* srhb hasn't really found nix build very useful yet
<Baughn> Doesn't spam the console, shows build progress.
<srhb> :)
<ambro718> https://github.com/NixOS/nixops/blob/master/release.nix wants to import nixpkgs, cant get that to work, want to pass nixpkgs directly
<ambro718> is there a way to get this to work without patching it
<Baughn> srhb: http://ix.io/Wb0 <- Besides, I'm just using NixOps to simplify this thing.
<Baughn> As nice as the `nix` command is, I might drop nixops instead.
<{^_^}> [nixpkgs] @Mic92 merged pull request #36409 → bundler: 1.14.6 -> 1.16.1 → https://git.io/vAFou
<{^_^}> [nixpkgs] @Mic92 pushed 2 commits to master: https://git.io/vxfwu
<{^_^}> → 0aa95d73 by @Mic92: bundler: 1.14.6 -> 1.16.1
<{^_^}> → c3ccaede by @Mic92: Merge pull request #36409 from Mic92/bundler
<srhb> Baughn: I mostly use something other than nixops too.
<srhb> Baughn: Or well, a stripped down version of it really.
<srhb> (A stupid loop over nix-copy-closure and ssh foo .../activate)
<{^_^}> [nixpkgs] @roosemberth opened pull request #36875 → processing: 2.2.1 -> 3.3.6 → https://git.io/vxfwH
pkill9 has quit [Ping timeout: 276 seconds]
<roconnor> pierron: some day i need to have a deep dive with you on this topic. AFAIK replace-runtime-dependencies is the best we can possibly do without recompling the universe. I don't understand how your security propoal isn't the same as simply adding a security-updates overlay to the end of the package set which ends up recompiling the universe in case bash or gcc has a security update.
jrolfs has joined #nixos
<ambro718> what's the easiest way to make a derivation that just runs a shell script and does no post-processing? Currently having issues because it wants srcs?
<srhb> runCommand?
<Baughn> runCommand.
<pierron> roconnor: for that we would have to disintguish "self" into dynamically linked packages and statically linked packages.
fk_ has quit [Ping timeout: 260 seconds]
<ambro718> thanks
<pierron> roconnor: but this is doable.
<pierron> roconnor: the way it works is by being able to extract dynamically linked packages from the list of inputs.
spietz has joined #nixos
<{^_^}> Channel nixpkgs-17.09-darwin advanced to https://github.com/NixOS/nixpkgs/commit/c7f415b5225 (from 6 hours ago, history: https://channels.nix.gsc.io/nixpkgs-17.09-darwin)
<pierron> roconnor: thus you can diff the inputs of the patched-version with the non-patched version.
jrolfs has quit [Ping timeout: 256 seconds]
<pierron> roconnor: if there is no differences in the input/recipes, you get the same output.
szicari has quit [Read error: Connection reset by peer]
szicari_ has joined #nixos
<roconnor> pierron: your goal is to do run-time replacements of dynamic libs and rebuild staticly-linked dependencies?
<{^_^}> [nixpkgs] @LnL7 merged pull request #36833 → petsc: fix darwin build → https://git.io/vxexC
<{^_^}> [nixpkgs] @LnL7 pushed 2 commits to master: https://git.io/vxfry
<{^_^}> → edef3444 by @knedlsepp: petsc: fix darwin build
<{^_^}> → a431a65a by @LnL7: Merge pull request #36833 from knedlsepp/fix-petsc-on-darwin
<pierron> roconnor: if there is any differences then you get a patching package which replace theinputs by the patched inputs.
<pierron> roconnor: the goal is to avoid the compilation of packages to know how to fix them.
<pierron> roconnor: thus a security update should only recompile the minimal set of packages which depend on it.
<pierron> roconnor: and all the patching happens locally.
<roconnor> patching means replace-runtime-dependencies?
<pierron> roconnor: a security fix, will recompile all the statically linked packages.
<mjacob> Baughn: apparently, the nix post-install script of my distribution's package wasn't run properly
<mjacob> Baughn: anyway, it works now; thank you for your help and time!
<Baughn> mjacob: Ahh. What did you do instead?
<pierron> roconnor: yes, except that the list of "runtime dependencies" is computed at "recipe-time"
<pierron> roconnor: before having compiled anything.
<roconnor> And you believe you can differentiate between statically linked dependencies and dynamically linked dependencies? Automatically?
<pierron> roconnor: Thus, giving thr ability to only share the delta, and not the result of all the patched packages
ma27 has quit [Ping timeout: 255 seconds]
<pierron> roconnor: not automatically, but this can easily be verified at the end of the compilation.
<pierron> roconnor: this can be verified automatically
<mjacob> Baughn: i'm not sure. basically i tried reinstalling it a few more times ;)
<roconnor> I'm skeptical, but I guess I need to look into this more carefully. Thanks for all your work!
<roconnor> (I was pretty skeptical of _override back in 2014 too ;)
<roconnor> (turns out it was awesome)
robstr_ has quit [Quit: Page closed]
tmaekawa has joined #nixos
<jluttine> haskell package ghc-mod has been broken for quite some time in unstable
<jluttine> oh, there's an issue about it already :+1:
otulp has joined #nixos
<pierron> roconnor: I tried with someone else in the London meetup, and we converge to the same solution.
tmaekawa has quit [Client Quit]
<roconnor> fascinating
<aristid> pierron: you're nbp! always confusing when people change their nicks :D
<pierron> roconnor: I guess the biggest blocker for most people is to consider that Nixpkgs could be a single function, and get them to understand how packages inside this function behave differently when provided different inputs.
<pierron> aristid: yes, I wish I could change my nick on freenode …
<pierron> aristid: but someone else hold it at the moment.
<aristid> pierron: oh you're still nbp when that is possible :D
jrolfs has joined #nixos
<pierron> aristid: on irc.mozilla.org, yes
<aristid> i could swear you used to use that nick here too. anyways, don't mind me please, continue discussing actually relevant things :)
<pierron> aristid: I will answer, I do highlight it.
<pierron> (which is sometimes painful when people are pasting sha …)
babariviere has joined #nixos
pkill9 has joined #nixos
woffs has quit [Quit: Gateway shutdown]
bkchr has quit [Quit: Konversation terminated!]
Tucky has quit [Quit: WeeChat 2.0.1]
woffs has joined #nixos
jrolfs has quit [Ping timeout: 256 seconds]
Mateon3 has joined #nixos
<{^_^}> [nixpkgs] @LnL7 pushed commit from @knedlsepp to release-18.03 « petsc: fix darwin build »: https://git.io/vxfie
Mateon1 has quit [Ping timeout: 264 seconds]
ma27 has joined #nixos
Mateon3 is now known as Mateon1
<ambro718> so I got nixops master to install - https://paste.ubuntu.com/p/HnzrYbXmss/
<ambro718> now it doesn't even run :( ImportError: No module named libvirt
mrkgnao has quit [Ping timeout: 260 seconds]
jrolfs has joined #nixos
<ambro718> ok fixed it by adding dependency...
<Baughn> Ha! Finally got `nix build` to build multiple machine configs at once.
reinzelmann has joined #nixos
<{^_^}> [nixpkgs] @LnL7 pushed 2 commits to master: https://git.io/vxfPf
<{^_^}> → 74b81ea9 by @knedlsepp: pythonPackages.shapely: fix darwin build
<{^_^}> → 9055cb81 by @LnL7: Merge pull request #36837 from knedlsepp/fix-pythonPackages.shapely-on-darwin
<{^_^}> [nixpkgs] @LnL7 merged pull request #36837 → pythonPackages.shapely: fix darwin build → https://git.io/vxejl
<Baughn> Could use a bit of refactoring, but http://ix.io/Wc6
<Baughn> `nix build -f ~/dev/nix-system '(import ./personal.nix).all-systems`
<{^_^}> [nixpkgs] @LnL7 pushed commit from @knedlsepp to release-18.03 « pythonPackages.shapely: fix darwin build »: https://git.io/vxfPm
Supersonic has quit [Quit: Bye]
Supersonic has joined #nixos
<{^_^}> [nixpkgs] @markuskowa closed pull request #36836 → beegfs: introduce patches for kernel 4.14 → https://git.io/vxepb
<{^_^}> [nixpkgs] @roosemberth closed pull request #36875 → processing: 2.2.1 -> 3.3.6 → https://git.io/vxfwH
<{^_^}> [nixpkgs] @LnL7 merged pull request #36873 → bcal: 1.7 -> 1.8 → https://git.io/vxfaR
<{^_^}> [nixpkgs] @LnL7 pushed 2 commits to master: https://git.io/vxfXv
<{^_^}> → 59cc47d8 by @jfrankenau: bcal: 1.7 -> 1.8
<{^_^}> → a6bf2b21 by @LnL7: Merge pull request #36873 from jfrankenau/update-bcal
<{^_^}> [nixpkgs] @LnL7 pushed commit from @jfrankenau to release-18.03 « bcal: 1.7 -> 1.8 »: https://git.io/vxfXt
Guanin has quit [Remote host closed the connection]
<{^_^}> [nixpkgs] @Mic92 pushed to master « Revert "bundler: 1.14.6 -> 1.16.1" »: https://git.io/vxfXn
Guanin has joined #nixos
<Baughn> `nix copy --from $(readlink result/madoka) --to ssh://madoka`
<Baughn> `error: creating directory '/nix/store/s8jcmg4g1090j6hzp1yh33njkaizf7z1-nixos-system-madoka-18.03pre-git/nix': Read-only file system`
<Baughn> Aww...
jrolfs has quit [Ping timeout: 264 seconds]
raynold has joined #nixos
hakujin has quit [Quit: WeeChat 2.0.1]
Itkovian has joined #nixos
John882 has joined #nixos
<clever> Baughn: skip the --from
<clever> Baughn: its trying to create a whole new store in result/madoka/nix/store to act as the source
<Baughn> Ah.
jrolfs has joined #nixos
<clever> a heap of undocumented examples
hlolli has quit [Quit: Leaving]
rihards has joined #nixos
<Baughn> Meanwhile I figured out how to run the NixOS tests against a particular machine config. Which is nice.
babariviere has quit [Quit: WeeChat 2.0]
<{^_^}> [nixpkgs] @vcunat pushed 2 commits to master: https://git.io/vxfDl
<{^_^}> → a758ae44 by @tilpner: fcgiwrap: don't error on implicit fallthrough
<{^_^}> → b0b1de89 by @vcunat: Merge #36807: fcgiwrap: avoid failure on warning
jrolfs has quit [Ping timeout: 264 seconds]
<{^_^}> [nixpkgs] @vcunat merged pull request #36807 → fcgiwrap: don't error on implicit fallthrough → https://git.io/vxe8D
reinzelmann has quit [Quit: Leaving]
<{^_^}> [nixpkgs] @vcunat pushed 2 commits to master: https://git.io/vxfyJ
<{^_^}> → f3809c6f by @Ma27: libgroove: fix build
<{^_^}> → 8a74c75e by @vcunat: Merge #36626: libgroove: fix build
xcmw has joined #nixos
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<{^_^}> [nixpkgs] @vcunat pushed 2 commits to release-18.03: https://git.io/vxfyo
<{^_^}> → 0092cf65 by @vcunat: Merge #36807: fcgiwrap: avoid failure on warning
<{^_^}> → bb885854 by @vcunat: Merge #36626: libgroove: fix build
<{^_^}> [nixpkgs] @vcunat merged pull request #36626 → libgroove: fix build → https://git.io/vApmF
jrolfs has joined #nixos
humanoyd has quit [Quit: WeeChat 2.0.1]
<{^_^}> [nixpkgs] @vcunat pushed 2 commits to master: https://git.io/vxfSI
<{^_^}> → 468c5f4e by @xeji: darling-dmg: fix build with gcc7
<{^_^}> → 1d968d18 by @vcunat: Merge #36686: darling-dmg: fix build with gcc7
<{^_^}> [nixpkgs] @vcunat pushed to release-18.03 « Merge #36686: darling-dmg: fix build with gcc7 »: https://git.io/vxfSW
<Myrl-saki> Wow, I actually broke Nix. And I don't know how.
<Myrl-saki> My ghcjs is compiling something, and god knows what the hell it's trying to compile.
<sphalerite> Myrl-saki: strace to the rescue?
<{^_^}> [nixpkgs] @vcunat closed pull request #36766 → fcgiwrap: fix build with gcc7 → https://git.io/vAj2t
<sphalerite> and procfs
<sphalerite> (or lsof if you prefer)
jrolfs has quit [Ping timeout: 256 seconds]
<Myrl-saki> sphalerite: Wait, I think I know what's happening. My ./. has build artifacts, and I think the timestamp messed with the rebuild detection.
<sphalerite> oh yes, that's always fun
<sphalerite> lib.cleanSource :)
<sphalerite> although that probably won't catch the specific artifacts in this case
mounty_ has joined #nixos
<{^_^}> [nixpkgs] @LnL7 pushed to master « xarchiver: fix darwion build »: https://git.io/vxfSX
<Myrl-saki> sphalerite: I was right. Sadly and happily, Nix wasn't broken.
<{^_^}> [nixpkgs] @vcunat merged pull request #36686 → darling-dmg: fix build with gcc7 → https://git.io/vAhc7
<{^_^}> [nixpkgs] @LnL7 pushed to release-18.03 « xarchiver: fix darwion build »: https://git.io/vxfSQ
<{^_^}> [nixpkgs] @vcunat pushed 2 commits to master: https://git.io/vxfS5
<{^_^}> → 95e88ad8 by Timo Kaufmann: bip: Fix build
<{^_^}> → 8eca6358 by @vcunat: Merge #36764: bip: fixup build with gcc7
jrolfs has joined #nixos
<{^_^}> [nixpkgs] @vcunat pushed to release-18.03 « Merge #36764: bip: fixup build with gcc7 »: https://git.io/vxf9U
<infinisil> Alright I've had a pretty weird problem with monitors for a while
<Myrl-saki> sphalerite: For future reference, what does lib.cleanSource do?
<infinisil> With only monitor A plugged in everything works
<infinisil> With monitor A and B plugged in and both of them active (through xrandr) everything works
hlolli has joined #nixos
<sphalerite> Myrl-saki: filters out .git and other VCS directories, vim swapfiles,*.o, *.so, and result symlinks
<infinisil> BUT with monitor A and B plugged in and /only/ A active, my X renders everything super delayed
amir has quit [Read error: Connection reset by peer]
<{^_^}> [nixpkgs] @LnL7 merged pull request #36764 → bip: Fix build → https://git.io/vAjgR
<infinisil> It's unusable. Although clicks get registerd immediately, what it displays is at like 0.1 FPS
<sphalerite> Myrl-saki: oh and *~. Goal being to minimise duplication and space consumed by files that are irrelevant to the build
<Myrl-saki> sphalerite: Ah. :P
<sphalerite> Myrl-saki: you can use filterSource with your own filter, possibly using lib.cleanSourceFilter as well, to adapt it to your use case though
<infinisil> It's an nvidia graphics card, and I am using the nvidia driver
amir has joined #nixos
<infinisil> Anybody have a clue?
<Myrl-saki> sphalerite: That's also what I was thinking. How about haskellPackages having its own .cleanSource?
<sphalerite> infinisil: does it work with nouveau? If so, complain to nvidia :p
<hlolli> I'm getting unable to open display ":0.0" after connecting with networkmanager via wpa_supplicant and dhclient. I wonder if this is connected to my /etc/hosts but I have networking.hostname="localhost"; in my config.nix, am I missing something?
<infinisil> sphalerite: Oh yeah I'll try that
<alexteves> is there a nix-darwin equivalent for non-nixos linux?
Fare has quit [Ping timeout: 240 seconds]
<sphalerite> alexteves: closest I know of is home-manager
<mpickering> Has anyone ever proposed extending meta.platforms to indicate whether a package is either known to not work on a system or just hasn't been tested?
<mpickering> I feel like I spend a reasonable amount of time trying to work this out
<sphalerite> hlolli: what's giving thi serror?
<hlolli> sphalerite: any program that needs xserver, firefox, xrandr etc..
<hlolli> also HexChat, but I decided to open it before connecting to the internet :P
<alexteves> sphalerite: thanks, sounds like a start
jrolfs has quit [Ping timeout: 256 seconds]
<sphalerite> hlolli: that is very odd! Not sure what to make of it…
<sphalerite> hlolli: oh actually, is the hostname still "localhost" after connecting? Check the output of hostname
<hlolli> sphalerite: you mean echo $HOSTNAME ? Well it's empty
<infinisil> mpickering: you mean like a per-architecture broken attribute?
<sphalerite> hlolli: no I mean `hostname`
<sphalerite> infinisil: no, that's what meta.platforms is already
<infinisil> Hmm right..
<hlolli> spahlerite: the bin hostname exists, I run it and it's empty
<mpickering> Ofborg should test this from now on for new packages
<gchristensen> whats that?
<sphalerite> hlolli: and before connecting I presume it outputs localhost
<mpickering> That the meta.platforms attribute it set correctly.
<mpickering> *is
<hlolli> I'd need to restart, be back in 1 min..
aborsu has joined #nixos
<mpickering> The changes look good from yesterday to indicate different failure modes, thanks gchristensen
MP2E has joined #nixos
hlolli has quit [Remote host closed the connection]
<gchristensen> what is it that should be tested?
<gchristensen> oh, you're saying it already dose?
<gchristensen> does*
<mpickering> a lot of packages have "platforms.linux" when they also build on darwin
<mpickering> because they didn't test on darwin
Itkovian has joined #nixos
<gchristensen> ah
hlolli has joined #nixos
<hlolli> sphalerite: yes that fits, hostname reuturn value is empty after starting the internet
<LnL> ofborg is already helping with that a lot
<hlolli> but is localhost before
<{^_^}> [nixpkgs] @dywedir opened pull request #36877 → hplip: 3.17.10 -> 3.17.11 → https://git.io/vxfQG
<sphalerite> hlolli: that does sound like a bug. I'm guessing it's handling it the same as if hostname were set to null, which means "localhost until a hostname is assigned via DHCP"
<sphalerite> hlolli: a twofold bug, in fact; Firstly networking.hostName = "localhost"; not working as expected, and secondly X applications breaking on the hostname change that DHCP causes
<hlolli> sphalerite: hehe, that's nasty. It happens at 'sudo dhclient wlp2s0`
jrolfs has joined #nixos
<sphalerite> hlolli: as a workaround you should be able to do `sudo hostname localhost`, but it's definitely worth reporting the issue at https://github.com/nixos/nixpkgs/issues
Fare has joined #nixos
<hlolli> sphalerite: thanks for the workaround, nice! Yes, will open a ticket!
<roconnor> Is there a recommended article for reading about the new nix 2.0 and how to take advantage of it?
<infinisil> sphalerite: the right way to make it use nouveau is `services.xserver.videoDrivers = mkForce [ "nouveau" ];`, right?
<infinisil> It doesn't work, at all, using nouveau doesn't even make it start the display-manager
<infinisil> Oh wait, maybe it uses the other display, which i need to turn on, hold on..
Ariakenom has joined #nixos
<infinisil> Nope, both screens are black with nouveau..
<infinisil> And I thought it's supposed to be a failsafe fallback..
jrolfs has quit [Ping timeout: 260 seconds]
xcmw has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<sphalerite> infinisil: just don't set the nvidia option
<hlolli> infinisil: did you ever get `[XORG] (EE) /dev/dri/card0: failed to set DRM interface version 1.4: Permission denied` when trying to optirun, just reading about your sruggles with nvidia.
<sphalerite> roconnor: not really, but Eelco's nixcon talk https://www.youtube.com/watch?v=XVIKScU7Uf4 gives a good overview of new features
<infinisil> hlolli: optirun?
<infinisil> I will check the logs though
<hlolli> infinsil: ok nevermind, it's the bumblebee problem, very common, had the same in the past on Fedora.
<roconnor> sphalerite: ty
<{^_^}> [nixpkgs] @LnL7 pushed to master « v8: fix clang build »: https://git.io/vxf51
ryanartecona has quit [Quit: ryanartecona]
hamishmack has quit [Quit: hamishmack]
<infinisil> Alright gonna try without setting any videoDriver (so using the default ones)
<{^_^}> [nixpkgs] @LnL7 pushed to release-18.03 « v8: fix clang build »: https://git.io/vxf55
<sphalerite> probably get the same results though, since it'll use nouveau probably
<infinisil> Yup.. some flickering even!
<infinisil> but then black
jrolfs has joined #nixos
<{^_^}> [nixpkgs] @LnL7 pushed to master « sslscan: fix clang build »: https://git.io/vxfds
amir has quit [Read error: Connection reset by peer]
<{^_^}> [nixpkgs] @LnL7 pushed to release-18.03 « sslscan: fix clang build »: https://git.io/vxfdC
amir has joined #nixos
jrolfs has quit [Ping timeout: 264 seconds]
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
alexteves has quit [Ping timeout: 264 seconds]
szicari_ has quit [Quit: szicari_]
<ottidmes> infinisil: vesa should work, but your goal is not to just have a working screen, but to have your GPU working properly, then I would have had some suggestions if it was about an AMD card
szicari has joined #nixos
<paraseba> did nix-serve stop working recently? I'm getting 'invalid character in Base64 string at /nix/.../nix-serve.psgi line 42' since the last update. It seems to fail trying to base64 encode a signature
<paraseba> maybe there is some more robust alternative to nix-serve?
<{^_^}> [nixpkgs] @LnL7 pushed to master « libpar2: fix clang build »: https://git.io/vxfFn
ambro718 has quit [Quit: Konversation terminated!]
<aristid> i've got a remote build error, probably related to nix 2.0: error: cannot add path '/nix/store/bqnf8g8b2vgin4hk0y6qiwy7ribc1y5i-linux-config-4.14.26' because it lacks a valid signature
<aristid> anybody knows about this error?
<{^_^}> [nixpkgs] @LnL7 pushed to release-18.03 « libpar2: fix clang build »: https://git.io/vxfFu
ryanartecona has joined #nixos
Lears has joined #nixos
blankhart has quit [Ping timeout: 264 seconds]
<ottidmes> aristid: I assume you are running Nix 2.0
jrolfs has joined #nixos
hlolli has quit [Remote host closed the connection]
AWSCUBER has joined #nixos
<ottidmes> aristid: And I assume you got the error when using a substituter or some other mechanism to get a store path from a remote store?
<aristid> ottidmes: i use nix.distributedBuilds nixos option
<ottidmes> aristid: Alright, then you will have to add the secret key of the nix store on the builder to secret-key-files in the nix.conf of the builder
AWSCUBER has left #nixos [#nixos]
<ottidmes> aristid: You can generate one with nix-store if you do not already have a key pair
<{^_^}> [nixpkgs] @LnL7 pushed to master « libcdio: fix darwin build »: https://git.io/vxfbB
<sphalerite> strange, I don't seem to remember that being necessary for distributed builds for me
<ottidmes> aristid: The public key will need to be added to the nix.conf (trusted-public-keys) of the machine using the builder
[0x4A6F] has quit [Ping timeout: 264 seconds]
<lejonet> Hmm, is there anyway to force nixops to recreate a user, if you've changed it a little bit?
<{^_^}> [nixpkgs] @fpletz closed pull request #35460 → imlib2: 1.4.10 -> 1.5.0 → https://git.io/vAVDf
<aristid> ottidmes: oh, so the old symmetric key is no longer enough?
<sphalerite> lejonet: what have you changed exactly? Maybe setting mutableUsers to false will do what you want
[0x4A6F] has joined #nixos
Itkovian has joined #nixos
<{^_^}> [nixpkgs] @fpletz merged pull request #36598 → imlib2: 1.4.10 -> 1.5.0 → https://git.io/vAxHs
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to staging: https://git.io/vxfb1
<{^_^}> → 08165b23 by @ryantm: imlib2: 1.4.10 -> 1.5.0
<{^_^}> → e9961bf9 by @fpletz: Merge pull request #36598 from ryantm/auto-update/imlib2
<lejonet> sphalerite: I've mainly changed its home variable, and set a hashedPassword
mounty_ has quit [Ping timeout: 268 seconds]
<ottidmes> aristid: The one in /etc/nix/signing-key.sec (by default at least)? Yes, support for it has been removed: https://github.com/NixOS/nix/commit/f435f8247553656774dd1b2c88e9de5d59cab203
<{^_^}> [nixpkgs] @LnL7 pushed to release-18.03 « libcdio: fix darwin build »: https://git.io/vxfby
<lejonet> sphalerite: I'll try that and see what happens
<aristid> ottidmes: uh, how do i tell nixos-rebuild to NOT do remote builds so i can fix this? :D
orivej has joined #nixos
<sphalerite> lejonet: for home setting mutableUsers to false should do it. For the password, https://github.com/NixOS/nixpkgs/issues/36493 ; I'm not sure what the status of this is exactly, if it's expected but poorly documented behaviour or a bug.
civodul has joined #nixos
<ottidmes> aristid: sudo cp /etc/nix/nix.conf{,~} && sudo mv /etc/nix/nix.conf{~,} && sudo nano /etc/nix/nix.conf && systemctl restart nix-daemon
<wilornel> garbas: Hi garbas
<lejonet> sphalerite: its most likely mutableUsers = false; I want anyway, as the main source of users, except system users and 1 local debug user, is going to be LDAP
<ottidmes> aristid: Once you have setup the secret key option, secret-key-files, in the builder, all new paths will automatically be signed with these keys, but existing store paths will not, so you have to do this initial signing yourself: nix sign-paths --all --key-file <secret-key-file>
<roconnor> do you folks hardware.pulseaudio.enable = true?
jrolfs has quit [Ping timeout: 264 seconds]
<sphalerite> aristid: or you could build the system locally before running nixos-rebuild like `nix build -f '<nixpkgs/nixos>' config.system.build.toplevel`
<sphalerite> roconnor: I do
<sphalerite> aristid: err plus --builders ''
<garbas> wilornel!
<aristid> sphalerite: error: you are not privileged to build derivations
jrolfs has joined #nixos
<ottidmes> aristid: Note that you have to add the public key for cache.nixos.org to trusted-public-keys as well, it has this by default, but if you set it yourself, the default will be overwritten, so you have to add this explicitly
coot_ has joined #nixos
<sphalerite> aristid: as root then I guess
<aristid> sphalerite: wtf
<aristid> i always used to be build to nix-build as any user
<sphalerite> aristid: the issue is the --builders '' option
<aristid> sphalerite: huh?
<infinisil> Oh my god, /boot full..
<infinisil> Never going to use a mere 100MB for /boot ever again
rihards has quit [Quit: rihards]
<roconnor> infinisil: don't worry it happens all the time to me.
<roconnor> you need to delete some system profiles.
<sphalerite> aristid: if it weren't for that you could do it as an unprivileged user, but then it would try to use the remote builders. But since you can pass arbitrary builders using --builders, it can only be overridden by trusted users
<aristid> infinisil: on nixos, you need to give more space to anything nixos-related
<infinisil> roconnor: yeah
<lejonet> infinisil: I once set 32MB for /boot on a system... I could barely fit 2 kernels xD
<aristid> sphalerite: so i should make my main user trusted
<infinisil> I have 1GB /boot on one machine
hamishmack has joined #nixos
<infinisil> Which is plenty
<{^_^}> [nixpkgs] @markuskowa opened pull request #36879 → beegfs: 6.17 -> 6.18 → https://git.io/vxfNH
<aristid> sphalerite: why doesn't nixos-rebuild take the --builders option?
<sphalerite> aristid: yeah, although I'm guessing you will usually want to use the remote builders so it won't be necessary most of the time
<ottidmes> infinisil: 512MB is too, I have only had it full once, that was after constantly rebuilding and testing initrd's without garbage collecting them
<infinisil> Um, I can't seem to free any space on /boot
<roconnor> infinisil: I usually move kernels to /tmp
<sphalerite> aristid: it should! Not sure if this is an implementation bug or a todo, but it would definitely be worth opening an issue if there isn't already one
<roconnor> and then do a nixos-rebuild boot
<infinisil> I ran `sudo nix-collect-garbage --delete-older-than 3d`, and `sudo /run/current-system/bin/switch-to-configuration boot`
<roconnor> Granted this isn't robustly safe.
<sphalerite> yeah you'll probably need to remove some manually
<infinisil> But the second command gives me cannot copy /nix/store/9v8lwhclnvbwbykx66spx20d3mphr4ph-initrd/initrd to /boot/kernels/9v8lwhclnvbwbykx66spx20d3mphr4ph-initrd-initrd.tmp
<ottidmes> aristid: Why not just edit nix.conf like I suggested, remove the builders option, done, I have used it often to temporarily disable substituter because of different kind of errors
<{^_^}> [nixpkgs] @fpletz pushed 0 commits to staging-17.03: https://git.io/vxfAL
<roconnor> infinisil: collecting garbage isn't relevent. You need to remove old profiles.
<infinisil> roconnor: But shouldn't delete-older-than take care of that?
<infinisil> It worked in the past
<sphalerite> aristid: what you should be able to do is pass --option builders '' to nixos-rebuild
<roconnor> oh does it?
Guest98 has joined #nixos
<sphalerite> roconnor: yes, nix-collect-garbage is a bit of a misnomer.
<sphalerite> it removes old generations of *all* profiles on the system
<sphalerite> if you pass -d or --delete-older-than
<roconnor> oh neat.
<roconnor> I mean dangerous, but also neat :D
<sphalerite> I prefer manually deleting gens with nix-env --delete-* and then gcing with nix-store --gc
Neo-- has quit [Ping timeout: 252 seconds]
<sphalerite> gchristensen: any chance of getting the logs link in the channel topic switched to samueldr's?
<ottidmes> I just make sure I have a working system, and then just gc everything, but only when my free space drops to low (which is shown in my taskbar, so I can prevent out of space errors before they happen)
<gchristensen> link?
<samueldr> I can make it respond to another domain name if it's getting more official
<sphalerite> gchristensen: https://logs.nix.samueldr.com/nixos/
gchristensen changed the topic of #nixos to: Share the output of nix-shell -p nix-info --run nix-info to help us help you. || 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
<sphalerite> samueldr: congrats on your place in the topic ;)
<samueldr> yay!
<samueldr> though, it's a tiny bit concerning if botbot is feeling sick
<samueldr> I mean, it's not a nixos-exclusive tool
jrolfs has quit [Ping timeout: 256 seconds]
<sphalerite> yeah a bunch of people have been complaining about it in #lincolnloop
<sphalerite> so far it looks like it's just dying
<LnL> hey
<samueldr> my bot is open for any nix and nixos-related channel that wants logs
<infinisil> sphalerite: is there a difference between nix-collect-garbage and nix-store --gc?
<gchristensen> hey, LnL
<{^_^}> [nixpkgs] @oxij opened pull request #36881 → [Request For Testing] pkgs: libpulseaudio -> libcardiacarrest; nixos: override when running a daemon → https://git.io/vxfAj
<sphalerite> infinisil: nix-collect-garbage is just a thin wrapper around it that also has gen-deletion
szicari has quit [Quit: szicari]
<sphalerite> nix-collect-garbage(1): The command nix-collect-garbage is mostly an alias of nix-store --gc, that is, it deletes all unreachable paths in the Nix store to clean up your system. However, it provides two additional options: […]
<roconnor> infinisil: nix-store --gc cannot delete profiles
jrolfs has joined #nixos
<infinisil> Ah, should've just read the manpage
<ottidmes> samueldr: Awesome, your logger has sane support for searching! The builtin search of the browser worked very poorly if not at all with the botbot.me logs. And it has a good search itself it seems, awesome!
roberth has joined #nixos
<sphalerite> ottidmes: botbot has a working search as well
<samueldr> for me it was hit or miss, especially when coming from a google result
<samueldr> though, this isn't my software, I'm using this: https://github.com/whitequark/irclogger
<infinisil> Alright I finally solved my problem by deleting individual profiles with `sudo nix-env -d /nix/var/nix/profiles/system --delete-generations <numbers here>`
<{^_^}> [nixpkgs] @fpletz merged pull request #36844 → spideroak: 6.1.9 -> 7.0.1 → https://git.io/vxvky
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxfxG
<{^_^}> → 47c7ad14 by @lloydhazlett: spideroak: 6.1.9 -> 7.0.1
<{^_^}> → 0d703ec3 by @fpletz: Merge pull request #36844 from lloydhazlett/spideroak-7.0.1
<ottidmes> sphalerite: It never worked properly for me, really frustrating if you find exactly what your are searching for, but still have to search manually for the right location
<sphalerite> but yes I like the UI!
<infinisil> Oh and then a garbage collection
<roconnor> infinisil: you can also do --delete-generations 3d
<manveru> samueldr: you need some logs for the history?
<infinisil> roconnor: that didn't work for some reason
<roconnor> you don't need to do garbage collection. :)
<sphalerite> infinisil: TIL nix-env -d (as opposed to the long form --delete-generations) exists
<samueldr> manveru: I decided that going further than 2017 wasn't as helpful as I thought, but if it's becoming official, I may want to look into that
<infinisil> And I needed to remove some kernels manually after than, then sudo /run/current-system/bin/switch-to-configuration boot
<roconnor> nixos-rebuild boot will work
<samueldr> though, do note that this exists: https://nixos.org/irc/logs/
<infinisil> sphalerite: It does? I can't find anything in the manpage
<sphalerite> the bootloader scripts could probably do a better job of handling out-of-space conditions.
<samueldr> I think it may be appropriate to go both ways, log into the same / similar format, and import those
<roconnor> I suppose switch-to-configuration boot might work too.
<manveru> samueldr: ah cool, i have them since about 2014 :)
<infinisil> roconnor: Are you coconnor too?
<roconnor> no
<samueldr> so I'd be missing 20160722~~2017
<infinisil> roconnor: heh weird
<samueldr> up to 2016-11-22 looks like
<samueldr> not much!
<sphalerite> infinisil: oh right I misread your command, you put -d /nix/var/nix/profiles/system (I'm guessing you meant -p /nix/var/nix/profiles/system)
<infinisil> Alright ZFS question: I have a mirror, but everytime I reboot, one of the disks is offline and I need to manually online it, and then it resilvers it. What do i need to do so that both disks are available at boot?
<infinisil> sphalerite: Ah yeah, mistype
<samueldr> sphalerite might like the fact that it works without javascript :)
<aristid> sphalerite: lol, now i need to find a way to trigger a remote build to test if my fixed setup works. the only problem: most stuff is in binary cache :D
knupfer has joined #nixos
<sphalerite> samueldr: well I run firefox in a fedora chroot on my chromebook now :p
<sphalerite> aristid: override it! :D
<aristid> sphalerite: i... think chromiumBeta works for my purpose :D
<sphalerite> aristid: or go with something nonfree
<samueldr> hmm, the dillo output is... artistic, but readable
jrolfs has quit [Ping timeout: 264 seconds]
vidbina has quit [Ping timeout: 268 seconds]
jrolfs has joined #nixos
<sphalerite> yeah
blankhart has joined #nixos
<aristid> sphalerite: it... it's building locally *SAD*
noam has quit [Read error: Connection reset by peer]
noam has joined #nixos
<sphalerite> infinisil: that's weird, does it say why it's offline?
<ottidmes> aristid: Is it not potentially building remote, so what if you set max-jobs to 0, will it then build remotely, would at least be a test
<sphalerite> ottidmes: it usually prefers the remote builders over building locally so at least *something* should build remotely
<{^_^}> [nixpkgs] @fpletz merged pull request #36571 → acme: Fix test → https://git.io/vAxWT
<{^_^}> [nixpkgs] @fpletz pushed 3 commits to master: https://git.io/vxfhU
<{^_^}> → bd355808 by @srhb: nixos/tests/acme: go compat update, unvendor pkcs11
<{^_^}> → df3706c4 by @srhb: nixos/tests/acme: use mail-test-srv tls certs from source
<{^_^}> → 255e2751 by @fpletz: Merge pull request #36571 from srhb/acme-test-fix
<infinisil> sphalerite: Pool degraded, one of the devices is offline, something like that
<sphalerite> aristid: which command are you using to build and how are you determining that it's building locally?
<aristid> sphalerite: "nix build", and htop on server and laptop
<sphalerite> hm :/ and what does your nix.conf look like?
<infinisil> Oh wait, the problem is gone
reber has quit [Remote host closed the connection]
szicari has joined #nixos
jrolfs has quit [Ping timeout: 256 seconds]
<aristid> sphalerite: lol, stupid error: i had still commented out the nix.distributedBuilds = true;
* aristid gets the badge of "that's really stupid"
<sphalerite> hm. Do we really need that option at all? Seems a bit redundant
<rawtaz> you can have mine, you dont need to waste a new one
robstr_ has joined #nixos
[0x4A6F] has quit [Ping timeout: 264 seconds]
<aristid> sphalerite: historical reasons i guess?
Synthetica has quit [Quit: Connection closed for inactivity]
<{^_^}> [nixpkgs] @fpletz merged pull request #36818 → wireguard: 0.0.20180218 -> 0.0.20180304 → https://git.io/vxewe
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxfhX
<{^_^}> → a0cc592c by @jfrankenau: wireguard: 0.0.20180218 -> 0.0.20180304
<{^_^}> → ce23a71e by @fpletz: Merge pull request #36818 from jfrankenau/update-wireguard
magnetophon has quit [Remote host closed the connection]
dan_b has joined #nixos
<aristid> i guess i need keys in both directions
<aristid> because it needs to upload the inputs to the server
<sphalerite> I don't see what sort of historical reasons could be behind that — since specifying remote builders has always been necessary for distributed builds to work
<{^_^}> [nixpkgs] @fpletz pushed 4 commits to release-18.03: https://git.io/vxfhQ
<{^_^}> → 9dbc7639 by @srhb: nixos/tests/acme: use mail-test-srv tls certs from source
<{^_^}> → 665c75ec by @srhb: nixos/tests/acme: go compat update, unvendor pkcs11
<{^_^}> → 60d6d53e by @lloydhazlett: spideroak: 6.1.9 -> 7.0.1
<sphalerite> aristid: or just log in as a trusted user on the server
magnetophon has joined #nixos
<sphalerite> more reliable, since the outputs on the client side might not be signed by a key trusted on othe remote builder
sigmundv has joined #nixos
[0x4A6F] has joined #nixos
<robstr_> After playing arround with my configurations, I receive `Configuring ghc-mod-5.8.0.0... Setup: Encountered missing dependencies: hlint >=2.0.8 && <2.1` BUT this worked before, I did a complete `nixos-rebuild switch` Has someone an idea ?
<Myrl-saki> clever: Ping (no need to pong if you're not here rn)
<aristid> sphalerite: that seems to work
<robstr_> sphalerite: lol... thank you
<aristid> heh "warning: dumping very large path (> 256 MiB); this may run out of memory"
[0x4A6F]1 has joined #nixos
<Fare> I don't understand what I need to do to make ssh-substituter-host work
[0x4A6F] has quit [Ping timeout: 268 seconds]
[0x4A6F]1 is now known as [0x4A6F]
<Fare> it tells me imported archive of â/nix/store/ha76pwcs24b0lislbhw41pn3nddlwii5-gerbil-unstable-2018-03-10â lacks a signature at /run/current-system/sw/bin/nix-copy-closure line 106.
jrolfs has joined #nixos
Neo-- has joined #nixos
raynold has quit [Quit: Connection closed for inactivity]
<sphalerite> Fare: probably same problem as aristid was having — you need to use a trusted user on the remote end
<sphalerite> wait, ssh-substituter-host?
<Fare> what's a "trusted user" ? One who can sudo without password?
<sphalerite> I think that's superseded by builders now
<sphalerite> no, one in trusted-users in nix.conf
djahandarie has left #nixos [#nixos]
<aristid> Fare: you probably switched to nix 2.0 very recently, too? :D
<aristid> sphalerite: i do feel a bit uneasy making "nixbuildfarm" a "trusted user" :D
mounty_ has joined #nixos
<aristid> but i guess if i wanted to avoid that, i'd need to set up hydra
drewc has joined #nixos
<roconnor> aristid: when you guys are done let me know what I need to do to set-up my remote builds for nix-2.0 :D
<Fare> Looks like I'm still using 1.11.16
<{^_^}> [nixpkgs] @fpletz merged pull request #32992 → nixos-container: Fix `destroy` terminating before it's done. Fixes #32545 → https://git.io/vbHoj
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJe1
<{^_^}> → 5d83988c by @nh2: nixos-container: Fix `destroy` terminating before it's done. Fixes #32545.
<{^_^}> → 5461e72c by @fpletz: Merge pull request #32992 from nh2/fix-32545-blocking-container-destroy
<aristid> roconnor: oh it works now :)
<aristid> roconnor: i guess i'll make a public gist :)
<roconnor> <3
<LnL> roconnor: what platform?
<roconnor> x86
<{^_^}> [nixpkgs] @fpletz pushed to master « nixos/tests/containers-tmpfs: fix test name »: https://git.io/vxJe5
blankhart has quit [Ping timeout: 240 seconds]
<LnL> aristid: build-remote requires a trusted user because builds are substituted both ways
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to release-18.03: https://git.io/vxJep
<{^_^}> → a3d87f1e by @fpletz: nixos/tests/containers-tmpfs: fix test name
<{^_^}> → 0c077427 by @nh2: nixos-container: Fix `destroy` terminating before it's done. Fixes #32545.
<LnL> roconnor: the other part of the dash :p nixos, linux + nix, darwin, ...
<roconnor> linux
<aristid> LnL: yeah i know, i just feel uneasy about it :)
<{^_^}> [nixpkgs] @fpletz merged pull request #36879 → beegfs: 6.17 -> 6.18 → https://git.io/vxfNH
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJvU
<{^_^}> → 11d40d53 by @markuskowa: beegfs: 6.17 -> 6.18
<{^_^}> → 9715e278 by @fpletz: Merge pull request #36879 from markuskowa/beegfs-6.18
robstr has quit [Remote host closed the connection]
<aristid> roconnor: uh, my config is obviously very nixos, so if you use plain nix-on-linux you might need to adapt it :)
robstr has joined #nixos
<{^_^}> [nixpkgs] @fpletz pushed commit from @markuskowa to release-18.03 « beegfs: 6.17 -> 6.18 »: https://git.io/vxJvZ
<roconnor> oh I'm using nixos
<LnL> roconnor: daemon or single user install
<LnL> oh
<aristid> roconnor: then enjoy!
<{^_^}> [nixpkgs] @fpletz merged pull request #36872 → gitkraken: 3.3.4 -> 3.4.0 → https://git.io/vxf2U
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJvC
<{^_^}> → e11f0aee by @phryneas: gitkraken: 3.3.4 -> 3.4.0
<{^_^}> → 62aa04ce by @fpletz: Merge pull request #36872 from phryneas/gitkraken_3.4.0
Fare has quit [Ping timeout: 256 seconds]
<aristid> also i guess everybody else can tell me what i did poorly in this config :D
takeda has quit [Quit: leaving]
<aristid> while fpletz merges ALL THE PRs
<LnL> aristid: roconnor: don't forget programs.ssh.knownHosts
<aristid> LnL: why?
szicari has quit [Quit: szicari]
<LnL> that's the most common pitfall people run into when setting up distributed builds
<aristid> LnL: because it's less secure without this option?
<roconnor> aristid: did you have to generate keys with SSL or something?
<roconnor> not ssh-keys, I mean binary signing keys.
robstr has quit [Ping timeout: 240 seconds]
<LnL> no, /root/.known_hosts also works but you need one of those otherwise it won't work
asuryawanshi has quit [Ping timeout: 255 seconds]
<{^_^}> [nixpkgs] @fpletz merged pull request #36877 → hplip: 3.17.10 -> 3.17.11 → https://git.io/vxfQG
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJfL
<{^_^}> → be6d9e1e by @dywedir: hplip: 3.17.10 -> 3.17.11
<{^_^}> → fc2ac9ac by @fpletz: Merge pull request #36877 from dywedir/hplip
<aristid> roconnor: not with ssl
robstr_ has quit [Ping timeout: 260 seconds]
<aristid> roconnor: nix-store --generate-binary-cache-key
<roconnor> I have /etc/nix/signing-key.pub and *.sec from before.
<roconnor> oh, that command seems new
<wilornel> garbas: I have some question about ib-controller and ib-tws. Are you online?
<aristid> roconnor: the old keys are not supported anymore
<aristid> i wish i had a closet to put my server in, so i don't need to hear it :D
<{^_^}> [nixpkgs] @fpletz merged pull request #36863 → vim-solidity: init at 2018-02-11 → https://git.io/vxfqM
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJfy
<{^_^}> → 1beda3fc by @sveitser: vim-solidity: init at 2018-02-11
<{^_^}> → 07f908f5 by @fpletz: Merge pull request #36863 from sveitser/vim-solidity
<sphalerite> !doesanyoneuse
<{^_^}> Don't ask if anyone uses software xyz, just ask your specific question. This helps solve your problem quicker, and allows people who haven't used xyz but may still be able to help you to do so.
<sphalerite> wilornel: ^
<roconnor> I'll delete these old keys now so I don't confuse myself.
<sphalerite> you might need very garbas-specific knowledge here, idk, but maybe you should just ask rather than needing 3 roundtrips to get to the actual question :D
Neo-- has quit [Ping timeout: 256 seconds]
<{^_^}> [nixpkgs] @fpletz merged pull request #36832 → wxhexeditor: 0.22 -> 0.24 → https://git.io/vxex3
<{^_^}> [nixpkgs] @fpletz pushed commit from @symphorien to master « wxhexeditor: 0.22 -> 0.24 (#36832) »: https://git.io/vxJJ4
<neonfuz> anyone know what package normally has the c development manpages?
<neonfuz> like for the various headers (IE "man string.h")
<sphalerite> neonfuz: man-pages
<sphalerite> neonfuz: although I don't think there's a manpage per header, but for related functions instead
<sphalerite> so man strlen,strcpy,etc
<neonfuz> k
<aristid> roconnor: probably a good idea :D
MichaelRaskin has joined #nixos
tkral has quit [Quit: Ping timeout (120 seconds)]
tkral has joined #nixos
John882 has quit [Ping timeout: 264 seconds]
mounty has quit [Remote host closed the connection]
mounty has joined #nixos
bsima has quit [Ping timeout: 260 seconds]
hiroshi has quit [Read error: Connection reset by peer]
jrolfs has quit [Ping timeout: 240 seconds]
<{^_^}> [nixpkgs] @Infinisil opened pull request #36882 → [WIP] nixos/zfs: Don't force import by default → https://git.io/vxJUW
disasm has quit [Ping timeout: 260 seconds]
badi has quit [Ping timeout: 260 seconds]
wladz has quit [Ping timeout: 260 seconds]
ryantm has quit [Ping timeout: 260 seconds]
neonfuz has quit [Ping timeout: 260 seconds]
lsyoyom has quit [Ping timeout: 260 seconds]
mfiano has quit [Ping timeout: 260 seconds]
nahamu has quit [Ping timeout: 260 seconds]
hiratara has quit [Ping timeout: 276 seconds]
wladz has joined #nixos
Myrl-saki has quit [Ping timeout: 260 seconds]
tippenein has quit [Ping timeout: 260 seconds]
ryantm has joined #nixos
bsima has joined #nixos
coot_ has quit [Quit: coot_]
Myrl-saki has joined #nixos
tippenein has joined #nixos
<{^_^}> [nixpkgs] @fpletz merged pull request #33643 → pcstat: init at (shortRev 91a7346) → https://git.io/vNt7F
<{^_^}> [nixpkgs] @fpletz pushed commit from @AmineChikhaoui to master « pcstat: init at 2017-05-28 (#33643) »: https://git.io/vxJUE
<infinisil> Everybody with a ZFS setup please check out the above PR (https://github.com/NixOS/nixpkgs/pull/36882), it's important that this gets tested properly, it affects all ZFS users
neonfuz has joined #nixos
nahamu has joined #nixos
disasm has joined #nixos
hiratara has joined #nixos
badi has joined #nixos
pjan has quit [Ping timeout: 240 seconds]
Fare has joined #nixos
jrolfs has joined #nixos
<aristid> infinisil: if i already have forceImportRoot = false; and forceImportAll = false; ... does this mean i have tested your PR? :D
<infinisil> aristid: Indeed! Drop a comment indicating that :)
mfiano has joined #nixos
pjan has joined #nixos
<aristid> infinisil: done
hiroshi has joined #nixos
<sphalerite> infinisil: maybe poke the mailing list as well
blankhart has joined #nixos
hlolli has joined #nixos
lsyoyom has joined #nixos
<hlolli> what's the syntax for loading an unstable dependency into the nix-shell, my intuition is `nix-shell -p alsaLib -p mpg123 -p <nixos-unstable>.node-9_x` but that wont cut the mustard
<infinisil> sphalerite: good idea
<sphalerite> brb rebooting to try it :)
<{^_^}> [nixpkgs] @fpletz pushed 4 commits to release-18.03: https://git.io/vxJTo
<{^_^}> → d6986819 by @phryneas: gitkraken: 3.3.4 -> 3.4.0
<{^_^}> → b9cfe0ce by @dywedir: hplip: 3.17.10 -> 3.17.11
<{^_^}> → 0cc00017 by @symphorien: wxhexeditor: 0.22 -> 0.24 (#36832)
aborsu has quit [Quit: aborsu]
<hlolli> btw (unrelated to my unstable dep question above) I got nvidia working on my laptop, first time in years I've seen nivida propirotary driver running on linux. Two pitfalls: 1 the Card bus in lspci is hex but should be given to nvidia config as decimal. And a Screen section needs to be defined with identifier used by Card section. Just shareing success story :)
<aristid> hlolli: hmm, setting up the nvidia card in my desktop was far easier
<sphalerite> hlolli: you shouldn't usually need to pass any options manually at all
<hlolli> well, it wouldn't find the device, then didnt find any available screens. The speed of which I was recompiling retesting the bumblebee compilation is insanely fast compared to when I was struggling with this in Fedora days.
FareTower has joined #nixos
Fare has quit [Ping timeout: 264 seconds]
<sphalerite> weird, but yay nixos!
<{^_^}> [nixpkgs] @abbradar pushed 4 commits to master: https://git.io/vxJkl
<{^_^}> → 2dfbef27 by @abbradar: dfhack: fix checksum
<{^_^}> → 5e9e3079 by @abbradar: xgboost: disable on AArch64
<{^_^}> → 5ed883a4 by @abbradar: warsow: 1.03 -> 2.1.2
boomshroom has joined #nixos
<boomshroom> Hello!
<boomshroom> I don't have much time, but I'm trying to offload a rust build to my desktop and the ssh connection is failing because `serialised integer 7161674624452356180 is too large for type 'j'`. What is that about and where did that number come from?
JosW has quit [Quit: Konversation terminated!]
<{^_^}> [nixpkgs] @abbradar pushed to master « warsow: add missing patch »: https://git.io/vxJkQ
FareTower has quit [Ping timeout: 246 seconds]
<{^_^}> [nixpkgs] @abbradar pushed 4 commits to release-18.03: https://git.io/vxJkp
<{^_^}> → 8f7a7dfb by @abbradar: xgboost: disable on AArch64
<{^_^}> → 99901537 by @abbradar: dfhack: fix checksum
<{^_^}> → 0f558b07 by @abbradar: warsow: 1.03 -> 2.1.2
<boomshroom> Also, the nix daemon keeps running out of memory.
erasmas has quit [Remote host closed the connection]
<infinisil> Oh no, my mail account isn't properly set up on my mac, can't send mails :/
<infinisil> Yes I managed to live for all my time on linux without sending mails, imagine that
[0x4A6F] has quit [Ping timeout: 264 seconds]
Gohla has quit [Ping timeout: 256 seconds]
<ottidmes> boomshroom: I got that error as well
<ottidmes> boomshroom: The fix for me was that I was using a ssh user without a proper shell configured, so nologin was used to pass the command to, and apparantly that results into a response that Nix daemon then fails to parse
[0x4A6F] has joined #nixos
<infinisil> Alright I sent an email to the mailing list
<infinisil> Anybody else I should ping for ZFS related things?
Gohla has joined #nixos
<sphalerite> between the mailing list and the pings on the github issue I think it should be good :)
<sphalerite> oh maybe clever too (cleverca22)
xcmw has joined #nixos
xcmw has quit [Client Quit]
<infinisil> sphalerite: Do you know if people still get pinged when i edit a comment and add their username?
Guest98 has quit [Remote host closed the connection]
<{^_^}> [nixpkgs] @nioncode opened pull request #36884 → p4v: init at 2017.3.1601999 → https://git.io/vxJLP
<mjacob> nix-env tries to write to some global profile directory, which my user is not allowed to. how can i create a user-specific profile?
civodul has quit [Quit: ERC (IRC client for Emacs 25.3.1)]
<sphalerite> infinisil: yeah they do I think
hiratara has quit [Quit: ZNC - http://znc.in]
jrolfs has quit [Ping timeout: 265 seconds]
raynold has joined #nixos
jrolfs has joined #nixos
boomshroom has quit [Quit: WeeChat 2.0]
spietz has quit [Quit: WeeChat 1.9.1]
Lisanna has quit [Quit: Lisanna]
infinisil has quit [Quit: Configuring ZNC, sorry for the join/quits!]
<{^_^}> [nixpkgs] @the-kenny pushed commit from @Ma27 to release-18.03 « pythonPackages.mt-940: fix build (#36784) »: https://git.io/vxJtD
<{^_^}> [nixpkgs] @globin merged pull request #36523 → grafana: 5.0.0 -> 5.0.1 → https://git.io/vAA3u
<{^_^}> [nixpkgs] @globin pushed commit from @WilliButz to master « grafana: 5.0.0 -> 5.0.1 »: https://git.io/vxJtQ
knupfer has quit [Ping timeout: 264 seconds]
simukis has quit [Ping timeout: 240 seconds]
infinisil has joined #nixos
davidlt_ is now known as davidlt
jrolfs has quit [Ping timeout: 256 seconds]
[0x4A6F] has quit [Ping timeout: 240 seconds]
FareTower has joined #nixos
jrolfs has joined #nixos
loonquawl has quit [Remote host closed the connection]
rabidfurby has joined #nixos
Ariakenom has quit [Read error: Connection reset by peer]
hlolli has quit [Quit: Leaving]
<{^_^}> [nixpkgs] @Ericson2314 opened pull request #36885 → treewide: Prepare for #34444 → https://git.io/vxJm0
<{^_^}> [nixpkgs] @veprbl opened pull request #36886 → doc: mention symlinkJoin in multiple-outputs section → https://git.io/vxJmE
Itkovian has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
rabidfurby has quit [Quit: The Lounge - https://thelounge.github.io]
<{^_^}> [nixpkgs] @volth opened pull request #36888 → xfce: do not use old names → https://git.io/vxJYk
Tobba has joined #nixos
<{^_^}> [nixpkgs] @svanderburg pushed to master « nodePackages: regenerate with node2nix 1.5.3 »: https://git.io/vxJYO
jrolfs has quit [Ping timeout: 260 seconds]
<{^_^}> [nixpkgs] @Ericson2314 merged pull request #36885 → treewide: Prepare for #34444 → https://git.io/vxJm0
<{^_^}> [nixpkgs] @Ericson2314 pushed 9 commits to master: https://git.io/vxJYW
<{^_^}> → dde80d70 by @Ericson2314: qtbase: Fix x86_64-conditional logic
<{^_^}> → eeb8419c by @Ericson2314: ceph: Fix --with-file-aio logic for new meta.platforms and cross
<{^_^}> → 79d8353b by @Ericson2314: treewide: Make `shouldUsePackages` copypasta use meta.available
snajpa- has left #nixos [#nixos]
<{^_^}> [nixpkgs] @svanderburg pushed to release-18.03 « nodePackages: regenerate with node2nix 1.5.3 »: https://git.io/vxJOk
hlolli has joined #nixos
<hlolli> after adding pulseaudio override, I can't rebuild, despite deleteing that override and rolling back (via nix-env --rollback ..). Error is: error: attribute ‘buildPerlPackage’ missing
<hlolli> my attemt to rollback was nix-env --rollback -p /nix/var/nix/profiles/system`
<{^_^}> [nixpkgs] @fpletz merged pull request #36630 → cpp-gsl: treat sign-conversion as warning → https://git.io/vApnA
<{^_^}> [nixpkgs] @fpletz pushed commit from @yuriaisaka to master « cpp-gsl: treat sign-conversion as warning (#36630) »: https://git.io/vxJOl
<mpickering> anyone familiar familiar with qt5 packaging (and darwin) know why "macextras" isn't an attribute in scope when "libsForqt5" is called?
<mpickering> or rather "qtmacextras"
<{^_^}> [nixpkgs] @fpletz merged pull request #36654 → jfbview: mark as broken (upstream issue) → https://git.io/vApVF
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJOz
<{^_^}> → a4f53eae by @xeji: jfbview: mark as broken (upstream issue)
<{^_^}> → cede1032 by @fpletz: Merge pull request #36654 from xeji/jfbview-broken
<{^_^}> [nixpkgs] @svanderburg pushed to release-17.09 « nodePackages: regenerate with node2nix 1.5.3 »: https://git.io/vxJOV
<makefu> hlolli: normally you can use nixos-rebuild switch --rollback but your command should have the same effect. can you show the override you wanted to do?
<{^_^}> [nixpkgs] @fpletz merged pull request #36666 → codeblocks: 16.01 -> 17.12, fix build → https://git.io/vApNO
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJOo
<{^_^}> → 58b7f17a by @xeji: codeblocks: 16.01 -> 17.12, fix build
<{^_^}> → 3c18afa0 by @fpletz: Merge pull request #36666 from xeji/codeblocks
blonkhart has joined #nixos
<hlolli> makefu: thanks, probably what I was looking for, I believe the only thing I did for this generation was this override
<hlolli> hardware.pulseaudio.package = pkgs.pulseaudio.override {
<hlolli> useSystemd = true;
<hlolli> };
<hlolli> systemd = pkgs.systemd;
<{^_^}> [nixpkgs] @fpletz merged pull request #36694 → dislocker: 0.6.1 -> 0.7.1 → https://git.io/vAhuk
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJOD
<{^_^}> → 2d704315 by @xeji: dislocker: 0.6.1 -> 0.7.1
<{^_^}> → 6d6d93c8 by @fpletz: Merge pull request #36694 from xeji/dislocker
<makefu> hlolli: your override also seems to be redundant as useSystemd is true by default and systemd normally uses the correct package
<{^_^}> [nixpkgs] @fpletz merged pull request #36773 → pythonPackages.asana: fix build → https://git.io/vAj9m
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJOp
<{^_^}> → 17e1795d by @Ma27: pythonPackages.asana: fix build
<{^_^}> → c94af976 by @fpletz: Merge pull request #36773 from Ma27/fix-python-asana
<hlolli> c/p from https://github.com/NixOS/nixos/issues/59 but yeh I should doubt more what I read online :)
freeman42x]NixOS has quit [Ping timeout: 268 seconds]
<makefu> by default pulseaudioLight is being used but stils still uses systemd :)
<makefu> that was an issue from 2012, 5 years ago :D
ryanartecona has quit [Quit: ryanartecona]
<makefu> the wiki is a bit more up-to-date https://nixos.wiki/wiki/PulseAudio
<hlolli> yeh, and I this wasn't even the reason for the bug I was trying to fix. The switch --rollback will keep the error appearing, despite how many rollbacks I do.
<{^_^}> [nixpkgs] @fpletz merged pull request #36776 → pythonPackages.gplaycli: 0.1.2 -> 3.21 → https://git.io/vAjQy
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJ3W
<{^_^}> → 4feaaa7f by @Ma27: pythonPackages.gplaycli: 0.1.2 -> 3.21
<{^_^}> → a1af032f by @fpletz: Merge pull request #36776 from Ma27/fix-gplaycli
hlolli has quit [Remote host closed the connection]
<{^_^}> [nixpkgs] @volth opened pull request #36889 → treewide: s/xlibs/xorg/g → https://git.io/vxJ30
hlolli has joined #nixos
pkill9 has quit [Ping timeout: 256 seconds]
<{^_^}> [nixpkgs] @fpletz merged pull request #36817 → vxl: fix build by using gcc6 → https://git.io/vxeVj
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJ3E
<{^_^}> → 0367a623 by @xeji: vxl: fix build by using gcc6
<{^_^}> → 73f8157a by @fpletz: Merge pull request #36817 from xeji/p/vxl
<{^_^}> [nixpkgs] @fpletz merged pull request #36820 → torque: fix build with gcc7 → https://git.io/vxewo
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJ3u
<{^_^}> → db2b3067 by @xeji: torque: fix build w/gcc7
<{^_^}> → 6a5cb56a by @fpletz: Merge pull request #36820 from xeji/p/torque
<hlolli> another thing I did besides this override was running `nix-env -i` thinking that it was sore of source of information, there's a clear error no idea if connected
<hlolli> error: Package ‘Agda-2.5.3’ in /nix/store/cbq8y13iv06hxgn3vl7qh8w3xnrs0s46-nixos-17.09.3146.084445b8f38/nixos/pkgs/development/haskell-modules/hackage-packages.nix:752 is marked as broken, refusing to evaluate.
<{^_^}> [nixpkgs] @bjornfor pushed commit from @ryantm to release-18.03 « sdcc: 3.6.0 -> 3.7.0 »: https://git.io/vxJ3g
<hlolli> but I'm not even useing haskell
<manveru> hlolli: use `nix-env -iA`
<hlolli> manveru: ok I see.
hiratara has joined #nixos
<hlolli> But this is my trace from `nixos-rebuild switch --show-trace` https://pastebin.com/i6kGQf4P I could open ticket, no stress actually
<{^_^}> [nixpkgs] @fpletz merged pull request #36814 → ufraw: fix build → https://git.io/vxeac
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJsf
<{^_^}> → 817cd00b by @xeji: ufraw: fix build by using gcc6
<{^_^}> → efd71020 by @fpletz: Merge pull request #36814 from xeji/p/ufraw
<hlolli> ah sorry for the noise, I fixed it :)
<hlolli> I did not realize the, what I should have realized, that my bad nix skills were the source of it
<hlolli> nixpkgs.config.packageOverrides = pkgs: {
<hlolli> pkgs.nodejs-9_x = import <nixos-unstable>.nodejs-9_x;
<hlolli> }; I somehow expected those kind of experiment to produce syntax errors or something.
fiddlerwoaroof has quit [Ping timeout: 256 seconds]
fiddlerwoaroof has joined #nixos
<{^_^}> [nixpkgs] @fpletz pushed to master « nodePackages: fix evaluation »: https://git.io/vxJsi
<{^_^}> [nixpkgs] @fpletz pushed 11 commits to release-18.03: https://git.io/vxJsd
<{^_^}> → 442da8b8 by @WilliButz: grafana: 5.0.0 -> 5.0.1
<{^_^}> → 050ad5dd by @yuriaisaka: cpp-gsl: treat sign-conversion as warning (#36630)
<{^_^}> → 2062aef4 by @xeji: jfbview: mark as broken (upstream issue)
<{^_^}> [nixpkgs] @fpletz merged pull request #36835 → bitcoin-unlimited: fix build with boost 1.66 → https://git.io/vxepV
<{^_^}> [nixpkgs] @fpletz pushed 2 commits to master: https://git.io/vxJGJ
<{^_^}> → 876420f0 by @YorikSar: bitcoin-unlimited: fix build with boost 1.66
<{^_^}> → c26f9600 by @fpletz: Merge pull request #36835 from YorikSar/fix-bitcoin-unlimited
<infinisil> hlolli: Well it's not a syntax error, but it will give an error when evaluated
<infinisil> So if you're not getting an error then it's not evaluating it
pkill9 has joined #nixos
<hlolli> ok, I was just somehow trying to get the unstable nodejs-9_x into my env so that I could make a shell with it via `nix-shell -p nodejs-9_x` would be any variable for that matter. Just need to read the manual better :)
<{^_^}> [nixpkgs] @fpletz pushed to release-17.09 « nodePackages: fix evaluation »: https://git.io/vxJGa
orivej has quit [Ping timeout: 256 seconds]