* gchristensen
aggressively, fruitlessly reloads the dell order page on his next laptop
davidak has joined #nixos
<contrapumpkin>
hmm libsystemd-id128
<contrapumpkin>
where do I find that?
<cstrahan>
Would be nice if there was a standard, and actually used, way for ruby gems to configure such things prior to installation. Gotta do hacks like pretend you have a c extension, and then write constants out and such.
<manveru>
contrapumpkin: then you need to add it to your package buildInputs
estewei has quit [(Quit: Quit)]
<ndowens08>
LnL: problem is that it doesnt notify of updates it seems
<manveru>
or propagatedBuildInputs
<cstrahan>
I'd bet that just about anything that uses ffi or fiddle will require manual patching.
<contrapumpkin>
manveru: yeah although even propagatedBuildInputs doesn't help when it's just a service I want to run
<ndowens08>
if there was a way to do something like debian does, like when you package there, you create a watch file that tells their system to check url for versions higher than in repo and notify
<contrapumpkin>
manveru, cstrahan: hmm, there's nothing in my cwd when I'm in my patchPhase
<contrapumpkin>
oh, are you building it in $out or something?
<contrapumpkin>
hmm, that doesn't seem to be it
<contrapumpkin>
ah, the source only seems to appear by buildPhase
<digitalmentat>
domenkozar, I pushed the changes that address MichaelRaskin's and edolstra's concerns (I think)
<contrapumpkin>
cstrahan, manveru: so how do I hook the gem build to patch it?
ryanartecona has quit [(Quit: ryanartecona)]
<contrapumpkin>
it doesn't seem to call any hooks
<contrapumpkin>
well, prePatch works but the source isn't there yet
Unode has quit [(Quit: Off it goes)]
phreedom has quit [(Ping timeout: 240 seconds)]
<ndowens08>
this is something i can work with :)
<cstrahan>
contrapumpkin: off the top of my head, first thing you'll need to do is set dontBuild=false -- normally we don't unpack the source and build a new .gem.
<contrapumpkin>
on the individual gem?
<cstrahan>
I'm on a train, on my phone, so not the most helpful at the moment.
<cstrahan>
Yeah.
mudri has quit [(Quit: WeeChat 1.7)]
Unode has joined #nixos
<contrapumpkin>
sweet, that seemed to work, thanks!
<ndowens08>
wow, uscan is already in our repo
siddharthist has joined #nixos
siddharthist has quit [(Client Quit)]
phreedom has joined #nixos
grantwu has joined #nixos
<grantwu>
How do I upgrade my packages?
<grantwu>
The manual seems to have instructions for NixOS but not Nix on some other distro.
<ndowens08>
nix-env -u after nix-channel --update
<grantwu>
Okay, thanks!
grantwu has left #nixos ["WeeChat 1.5"]
<contrapumpkin>
boo, I can't run VM tests from inside VMWare
<contrapumpkin>
seems to be some weird nested virtualization kvm unfriendliness
* contrapumpkin
pouts
<mguentner>
contrapumpkin: you can however kvm in your kvm (yo dawg!) (if it works)
<contrapumpkin>
my host isn't linux
<mguentner>
contrapumpkin: are you suggesting that your host does not threat you well? :)
<eacameron>
clever: Is that in the context of a derivation?
<clever>
eacameron: yeah, and i believe the fixup phase runs it on everything in $out/bin/ for you
<clever>
eacameron: line 30 will turn #!/usr/bin/env bash into an absolute /nix/store/foo-bash-1.2.3/bin/bash
<eacameron>
clever: Ah well that's very good to know. TBH though I'm not in a derivation. Just writing a handy script for myself.
<erlandsona>
So that's pretty sweet! Didn't know that was a thing!
<clever>
eacameron: ah, then Nafai's idea will still work, and it will just remain impure
<eacameron>
clever: Great. But I'm very glad I don't have to modify the script for a derivation...cuz it might happen.
<erlandsona>
Anyone know how to get lightdm.background to work with a jpg? I think it's not working because the image is in my users Downloads folder. But I'm not really sure where else I would put it and chmod 777 the_file didn't work either?
<erlandsona>
Rather does it have to be a specific size / format?
<clever>
eacameron: this is how patchShebangs hooks itself into the fixup phase, so you could just manualy run it on $out to apply the same fix under runCommand
<eacameron>
clever: I see. But might as well just use buildPhase
<alphor>
how do I set a broadcast address for an interface? there's nothing in nixos-option and if i don't set it, it (predictably) causes routing issues. setting up networking via extraCommands works as expected.
<clever>
alphor: i just configure networking.interfaces.<foo> = { ipAddress = "1.2.3.4"; prefixLength = 24; }; and it works fine
<alphor>
huh I didn't realize it was ipAddress and not address. Does nixos-option query the running system or docs?
<clever>
alphor: the running system, which will also include the strings the docs where made from
davidak has quit [(Quit: Leaving.)]
<clever>
alphor: but i think it has trouble with that type of option, where the name is part of the variable
<clever>
alphor: the configuration.nix man page also has every single option in it
Rizy has quit [(Quit: Rizy)]
takle has quit [(Remote host closed the connection)]
<alphor>
clever: setting ipAddress gives me an option doesn't exist error. nixos-version is 16.09.1698.2da8a5d (Flounder)
<clever>
what exactly did you put in the config?
<Ralith>
`networking.interfaces.<iface>.ip4` takes a list of { address = "..."; prefixLength = ...; }
<Ralith>
and there's the obvious ip6 version too
<clever>
and there is an ipAddress option that is a sibling of ip4
rjsalts has quit [(Quit: Coyote finally caught me)]
rjsalts has joined #nixos
fhoffmeyer2 has quit [(Ping timeout: 240 seconds)]
fhoffmeyer has joined #nixos
takle has joined #nixos
takle has quit [(Ping timeout: 258 seconds)]
MP2E has joined #nixos
systemfault has quit [(Quit: Bye!)]
dramforever has quit [(Quit: Leaving)]
takle has joined #nixos
mkoenig has quit [(Ping timeout: 260 seconds)]
excelangue has quit [(Ping timeout: 240 seconds)]
takle has quit [(Ping timeout: 276 seconds)]
mkoenig has joined #nixos
fhoffmeyer2 has joined #nixos
fhoffmeyer has quit [(Read error: Connection reset by peer)]
takle has joined #nixos
systemfault has joined #nixos
systemfault has quit [(Client Quit)]
takle has quit [(Ping timeout: 245 seconds)]
takle has joined #nixos
<NixOS_GitHub>
[nixpkgs] globin force-pushed openssl-1.1 from 105f535 to 3997eb9: https://git.io/vDOWO
<NixOS_GitHub>
nixpkgs/openssl-1.1 b8496c4 Robin Gloster: openssl: default to openssl_1_1_0
<NixOS_GitHub>
nixpkgs/openssl-1.1 b31a1c4 Robin Gloster: socat: 1.7.3.1 -> 1.7.3.2...
<NixOS_GitHub>
nixpkgs/openssl-1.1 8e4ba87 Robin Gloster: afflib: 3.7.6 -> 3.7.15
excelangue has joined #nixos
endformationage has quit [(Quit: WeeChat 1.6)]
takle has quit [(Ping timeout: 240 seconds)]
MercurialAlchemi has joined #nixos
olesk has joined #nixos
takle has joined #nixos
takle has quit [(Ping timeout: 252 seconds)]
dramforever has joined #nixos
takle has joined #nixos
olesk has quit [(Ping timeout: 256 seconds)]
tempest has quit [(Quit: WeeChat 1.5)]
<NixOS_GitHub>
[nixpkgs] globin force-pushed openssl-1.1 from 3997eb9 to 1e4a0ad: https://git.io/vDOWO
<NixOS_GitHub>
nixpkgs/openssl-1.1 73d21e0 Robin Gloster: podofo: 0.9.4 -> 0.9.5
<NixOS_GitHub>
nixpkgs/openssl-1.1 21d2ed9 Robin Gloster: certbot: 0.9.3 -> 0.11.1
<NixOS_GitHub>
nixpkgs/openssl-1.1 1e4a0ad Robin Gloster: pin some packages to openssl 1.0.2 for now
takle has quit [(Ping timeout: 276 seconds)]
takle has joined #nixos
<NixOS_GitHub>
[nixpkgs] peterhoeg opened pull request #22775: tox clients and libraries updates (master...u/qtox) https://git.io/vD61Z
takle has quit [(Ping timeout: 276 seconds)]
katyucha has quit [(Killed (verne.freenode.net (Nickname regained by services)))]
katyucha1 is now known as katyucha
katyucha1 has joined #nixos
reinzelmann has joined #nixos
ToxicFrog has quit [(Ping timeout: 276 seconds)]
takle has joined #nixos
olesk has joined #nixos
<NixOS_GitHub>
[nixpkgs] globin force-pushed openssl-1.1 from 1e4a0ad to 48446d1: https://git.io/vDOWO
<NixOS_GitHub>
nixpkgs/openssl-1.1 48446d1 Robin Gloster: pin some packages to openssl 1.0.2 for now
takle has quit [(Ping timeout: 260 seconds)]
takle has joined #nixos
takle has quit [(Ping timeout: 240 seconds)]
ToxicFrog has joined #nixos
MichaelRaskin has quit [(Remote host closed the connection)]
<NixOS_GitHub>
[nixpkgs] nicknovitski opened pull request #22776: vim_configurable: Add packPath option to vimrcConfig (master...vim-configurable-packpath) https://git.io/vD6Ds
johann__ has joined #nixos
goibhniu has joined #nixos
takle has joined #nixos
jgertm_ has joined #nixos
jgertm is now known as Guest54751
jgertm_ is now known as jgertm
goibhniu has quit [(Ping timeout: 260 seconds)]
justbeingglad has joined #nixos
takle has quit [(Ping timeout: 240 seconds)]
<NixOS_GitHub>
[nixpkgs] FRidh force-pushed python-wip from db48687 to b313a7b: https://git.io/vzaOS
<NixOS_GitHub>
nixpkgs/python-wip 7aac64d Frederik Rietdijk: pythonPackages.urllib3: 1.12 -> 1.20
<NixOS_GitHub>
nixpkgs/python-wip dd609ec Frederik Rietdijk: pythonPackages.llvmlite: 0.14.0 -> 0.15.0
<NixOS_GitHub>
nixpkgs/python-wip 6cf6826 Frederik Rietdijk: pythonPackages.numba: 0.29.0 -> 0.30.1
pi3r has joined #nixos
paperd has quit [(Quit: WeeChat 1.0.1)]
hsushil10 has joined #nixos
pi3r has quit [(Read error: Connection reset by peer)]
pi3r has joined #nixos
goibhniu has joined #nixos
<the-kenny>
Oh well, I suppose many people disagree that github is obsolete, based on the activity there
<the-kenny>
Github is one of these services that will only get replaced with something better over a long long time. Like sourceforge
* the-kenny
still wishes for a fully decentralized git hosting service with decentralized Pull Requests and Code Reviews
justbeingglad has left #nixos []
ToxicFrog has quit [(Ping timeout: 240 seconds)]
pi3r has quit [(Ping timeout: 260 seconds)]
fre has quit [(Ping timeout: 240 seconds)]
ToxicFrog has joined #nixos
takle has joined #nixos
defel has joined #nixos
dramforever has quit [(Ping timeout: 240 seconds)]
Itkovian has joined #nixos
takle has quit [(Ping timeout: 264 seconds)]
blahdodo has quit [(Ping timeout: 255 seconds)]
mkoenig has quit [(Ping timeout: 258 seconds)]
pi3r has joined #nixos
mkoenig has joined #nixos
ToxicFrog has quit [(Ping timeout: 255 seconds)]
roundhouse has joined #nixos
ToxicFrog has joined #nixos
nschoe has joined #nixos
<roundhouse>
hi guys, is the tor-browser package currently broken? I haven't used it in a while but right now with both 16.09 and unstable I get "Unable to start Tor. The Tor executable is missing" error upon trying to start tor-browser.
civodul has joined #nixos
jgertm has quit [(Ping timeout: 258 seconds)]
blahdodo has joined #nixos
takle has joined #nixos
<NixOS_GitHub>
[nixpkgs] FRidh pushed 6 new commits to staging: https://git.io/vD69A
<NixOS_GitHub>
nixpkgs/staging dc2262b Frederik Rietdijk: pythonPackages.urllib3: 1.12 -> 1.20
<NixOS_GitHub>
nixpkgs/staging 9ac7533 Frederik Rietdijk: pythonPackages.llvmlite: 0.14.0 -> 0.15.0
<NixOS_GitHub>
nixpkgs/staging 8a47306 Frederik Rietdijk: pythonPackages.numba: 0.29.0 -> 0.30.1
FRidh has joined #nixos
takle has quit [(Read error: Connection reset by peer)]
takle has joined #nixos
<NixOS_GitHub>
[nixpkgs] puffnfresh opened pull request #22777: keybase-gui: init at 20170209.17b641d (master...package/keybase-gui) https://git.io/vD6Hn
ToxicFrog has quit [(Ping timeout: 276 seconds)]
panostimos has joined #nixos
takle has quit [(Ping timeout: 264 seconds)]
takle has joined #nixos
ToxicFrog has joined #nixos
<NixOS_GitHub>
[nixpkgs] cillianderoiste pushed 1 new commit to master: https://git.io/vD6Hi
<NixOS_GitHub>
nixpkgs/master 868d30f Cillian de Roiste: openconnect: add support for stoken...
<Ralith>
roundhouse: that certainly sounds broken
takle has quit [(Ping timeout: 276 seconds)]
fritschy has joined #nixos
blahdodo has quit [(Ping timeout: 255 seconds)]
<roundhouse>
ralith, thanks. If anyone could reproduce the bug, I'll file a bug report.
<Ralith>
I'd check the maintainer and see if they're on IRC or otherwise easily pingable
<NixOS_GitHub>
[nixpkgs] edolstra pushed 1 new commit to master: https://git.io/vD67k
<NixOS_GitHub>
nixpkgs/master 6a472cf Eelco Dolstra: opensubdiv: Remove cudatoolkit from the closure...
derjohn_mob has quit [(Ping timeout: 258 seconds)]
<roundhouse>
ralith, how can I find the maintainer of a package?
<Ralith>
should be listed in the package source
<Ralith>
there's probably some nix incantation to query it too
<NixOS_GitHub>
[nixpkgs] globin force-pushed openssl-1.1 from 48446d1 to 68751cc: https://git.io/vDOWO
<NixOS_GitHub>
nixpkgs/openssl-1.1 a736266 Robin Gloster: twinkle: remove
<NixOS_GitHub>
nixpkgs/openssl-1.1 7a22ee5 Robin Gloster: sflphone: remove
<NixOS_GitHub>
nixpkgs/openssl-1.1 3911a48 Robin Gloster: freeswitch: 1.6.9 -> 1.6.15
olesk has quit [(Ping timeout: 240 seconds)]
<roundhouse>
hmm, it says "aintainers = with maintainers; [ offline matejc doublec thoughtpolice joachifm ];" Do these map canonically to IRC names?
takle has joined #nixos
<roundhouse>
joachifm, is the tor-browser package currently broken? I haven't used it in a while but right now with both 16.09 and unstable I get "Unable to start Tor. The Tor executable is missing" error upon trying to start tor-browser.
<Ralith>
not canonically, but odds are generally good
<roundhouse>
ralith, I messaged joachifm, which also exists as a irc name. Thanks for the help
takle has quit [(Ping timeout: 276 seconds)]
<NixOS_GitHub>
[nixpkgs] globin force-pushed openssl-1.1 from 68751cc to fe05939: https://git.io/vDOWO
<NixOS_GitHub>
nixpkgs/openssl-1.1 d7d0e2e Robin Gloster: aws-sdk-cpp: add patch to build with openssl 1.1.0
<NixOS_GitHub>
nixpkgs/openssl-1.1 fe05939 Robin Gloster: pin some packages to openssl 1.0.2 for now
<NixOS_GitHub>
[nixpkgs] FRidh opened pull request #22778: Python changelog - do not merge (master...changelog) https://git.io/vD65J
takle has joined #nixos
takle has quit [(Ping timeout: 252 seconds)]
blahdodo has joined #nixos
mortum5 has joined #nixos
takle has joined #nixos
olesk has joined #nixos
derjohn_mob has joined #nixos
cmercier has quit [(Remote host closed the connection)]
cmercier has joined #nixos
takle has quit [(Remote host closed the connection)]
jensens has joined #nixos
takle has joined #nixos
hsushil10 has quit [(Quit: -a- IRC for Android 2.1.34)]
fre has joined #nixos
jgertm has joined #nixos
ThatDocsLady has joined #nixos
thc202 has joined #nixos
eacameron has quit [(Remote host closed the connection)]
<jophish>
Hmm, I think I've discovered something *super* weird
<jophish>
is anyone able to enter a nix-shell in a directory in /run/user/1000/something
mortum5 has quit [(Ping timeout: 240 seconds)]
<jophish>
because I get the error: error: opening file ‘/run/user/default.nix’: No such file or directory
contrapumpkin has quit [(Ping timeout: 260 seconds)]
<jophish>
but, nix-shell default.nix works
<jophish>
hmm, even 'nix-shell .' works
<jophish>
so it seems as though nix-shell's working directory identification isn't working very well
<roundhouse>
works for me (/run/user/1001/dconf)
<jophish>
how strange
cmercier has quit [(Read error: No route to host)]
<gchristensen>
Mic92: should these wireguard PRs be backported?
blahdodo has joined #nixos
<Mic92>
gchristensen: we could treat it like dropbox and backport every version as soon as it gets released. Current version in stable is incompatible with unstable.
<Mic92>
There was only that one incompatible release back in dezember
<gchristensen>
that is very annoying
<Mic92>
After that only userspace api was violated, which is not that much of a problem, because both kernel module and tools gets released at the same time.
<gchristensen>
ah
<gchristensen>
I asked him for release notes
<Mic92>
gchristensen: I know his opinion
<Mic92>
upgrade or die
<gchristensen>
I know
<gchristensen>
unfortunately for him, we have rules
Shoue has quit [(Ping timeout: 255 seconds)]
<gchristensen>
also he can't just be an asshole
<ikwildrpepper>
gchristensen: who is 'he'?
<gchristensen>
ikwildrpepper: the author of wireguard is incredibly high-and-mighty
<NixOS_GitHub>
[nixpkgs] edolstra pushed 1 new commit to master: https://git.io/vDitU
<NixOS_GitHub>
nixpkgs/master 1a9cf94 Eelco Dolstra: Revert "opensubdiv: Remove cudatoolkit from the closure"...
teto has quit [(Ping timeout: 245 seconds)]
<LnL>
gchristensen: lol
<fpletz>
gchristensen: why can't he an asshole? also, the software is clearly marked as unstable/experimental
<gchristensen>
well ok he _can_ be an asshole
<riptawr>
jophish: this works for derivations but what about the global conf and global env vars? builtins.getEnv is there but there is no setEnv
<gchristensen>
but I doesn't make me inclined to look at his PRs
<riptawr>
i am thinking about environment.profiles and profileRElativeEnvVars, but not sure i configured it correctly, since it seems to not be working after a switch
<gchristensen>
not that I'm being a gatekeeper... but that being nicer would be nice.
teto has joined #nixos
<gchristensen>
fpletz: where is it marked unstable / experimental?
<riptawr>
omg, just found environment.variables, nvm
<riptawr>
...and it build etc-profile into the store, seems to be it
<fpletz>
gchristensen: well, the version number 0.0.x :) but I don't think that comment was rude - but then again, I'm not a native speaker
<NixOS_GitHub>
[nixops] rbvermaa pushed 1 new commit to master: https://git.io/vDitN
<NixOS_GitHub>
nixops/master fcc6352 Rob Vermaas: Update releasenotes for 1.5
<fpletz>
gchristensen: I've met him at fosdem and had some private conversions in irc… I think he's a good guy :)
<gchristensen>
I believe you
<ikwildrpepper>
he's just clear about his support policy :)
<gchristensen>
I find his approach to be incredibly rude, and I wouldn't be impressed if anyone behaved like that.
<gchristensen>
hmm should we be more clearly denoting wireguard as experimental?
<fpletz>
the thing is that wireguard is not finished yet - we should ask ourselves if we should even support it in a stable distribution (like those rc kernels that are available in stable but never updated \o/)
<gchristensen>
indeed...
<NixOS_GitHub>
[nixops] nh2 opened pull request #604: Atomically set permissions on nixops state file creation. (See #594) (master...statefile-atomic-create-permissions) https://git.io/vDimq
<gchristensen>
we could rename it wireguard-experimental
<LnL>
yeah, perhaps we should prune some unstable packages like that from the release
<gchristensen>
they seem to just be commit logs though
<NixOS_GitHub>
[nixpkgs] 7c6f434c pushed 1 new commit to master: https://git.io/vDimH
<NixOS_GitHub>
nixpkgs/master 3508b48 Michael Raskin: haskellPackages: fgl, fgl-arbitrary: jailbreak (w.r.t. hspec for tests)
<fpletz>
or a stable meta-flag that outputs a warning message if you're on a release branch and trying to install that package? :)
<mbrgm>
hmmm... is there a way for predictable interface naming to trigger before network-online.target?
<gchristensen>
before we discuss that too far -- what should we do about the ancient wireguard package we have in 16.09?
<mbrgm>
because I'm trying to package a service, which relies on a configured device name, but the nic is renamed from eth0 to enp0s3 way after network-online.target
<GlennS>
Anyone know if it's possible to use antiquotation (variable substitution) when using builtins.readFile (converts file to string)?
<mbrgm>
so the service will fail on boot because it cannot find the device
hlavaty has quit [(Remote host closed the connection)]
<dtz>
GlennS: yes it is, I'm using that currently
zx2c4 has joined #nixos
<zx2c4>
hey what's grahamc's handle?
<GlennS>
dtz: does it happen by default? If so, I guess I've just made some stupid synax error.
<gchristensen>
zx2c4: gchristensen
<zx2c4>
gchristensen: hey there you are
<zx2c4>
:)
<zx2c4>
I saw your thumbs up on that PR but i wasnt sure what it meant so i thought it'd be easier to track you down here to talk instead
<sphalerite>
I don't see how zx2c4 is being an asshole, I thought this would be some jwz-level stuff :P
<gchristensen>
hi there (I just spilled a bunch of coffee beans, back in a couple minutes! :)
<GlennS>
ah, I see
<zx2c4>
hey fpletz
<zx2c4>
sphalerite: um what
<GlennS>
I mean actually inside the text in the file
<GlennS>
rather than on the file name
<gchristensen>
zx2c4: I was just commenting on how I find you to be fairly coarse
<sphalerite>
uuh... speak of the devil and he shall appear? x)
<GlennS>
but thanks
<dtz>
OH haha I'd imagine that doesn't work by default no, but you can probably do the substitutions before or after reading it explicitly
<gchristensen>
zx2c4: especially since we're same team on being OSS maintainers
<gchristensen>
zx2c4: so we're just discussing how to handle wireguard for stable releases
<zx2c4>
gchristensen: regarding https://github.com/NixOS/nixpkgs/pull/22786 , does that thumbs up mean "this pull request will be merged and then after I will backport" or does it mean "this pull request will be merged and then after you should make pull requests for the backports" or does it mean "nothing is going to happen but i like everything that's been said"
<zx2c4>
or... something else?
<dtz>
yep, sorry for the misleading answer! :)
<zx2c4>
sphalerite: too bad i dont have the backlog. whats the skinny?
<sphalerite>
what gchristensen said
<zx2c4>
gchristensen: oh, okay
<zx2c4>
sorry, slow reading
<gchristensen>
zx2c4: great question, thank you for coming to IRC :)
<gchristensen>
zx2c4: I think for 16.09 we should absolutely backport the upgrade. 16.09 goes out of support in about 1.5 months.
<zx2c4>
wireguard simply isnt at the point where it makes sense to branch a certain snapshot and then "backport" particular fixes. it's very much a work in progress, albeit a fairly stable work in progress. it's pretty safe to just bump snapshots for all nix versions
<zx2c4>
gchristensen: ahh interesting about the support
<zx2c4>
yea probably best to bump before it's unsupported
<gchristensen>
zx2c4: for 17.03, we're trying to decide if we should remove it from the stable release (where users would have to add slihtly more code to use it)
<zx2c4>
well it _is_ pretty stable code, and is more and more widely used...
<fpletz>
we're just concered about another protocol change, I guess :)
<GlennS>
dtz: yeah, I think I can probably do it with builtins.replaceStrings
<zx2c4>
is there anything upstream can do to assist? for example, are those changelogs sufficient?
<gchristensen>
or rename it to wireguard-experimental (to reflect its upstream lack of historical support, it is a _stable_ release, and mandatory upgrades is a bit antithetical to that) or issue a warning
<NixOS_GitHub>
[nixpkgs] e-user opened pull request #22789: freeipa: init at 4.4.3 (master...pkgs/freeipa-4.4.3) https://git.io/vDiOa
<zx2c4>
the -experimental tag was actually removed from the upstream release tags a while back
<zx2c4>
after it "graduated" experimental status
<zx2c4>
(which is why it's now in debian sid)
<gchristensen>
yes i was removed from the tags, but not the release notes
<gchristensen>
which still explicitly calls it experimental, not dependably secure, and not applicable for CVEs
Shou has joined #nixos
<gchristensen>
(I read your release notes, thank you :))
<fpletz>
Profpatsch: reagarding pulseaudio zeroconf, looks right to me
<zx2c4>
that language actually changed slightly with the latest release notes
<zx2c4>
im curious:
pi3r has joined #nixos
<zx2c4>
how many extra steps would it be for stable users to add it if it's not in stable?
<zx2c4>
right now, if i understand correctly, the steps are:
<gchristensen>
it'd be a few extra lines in their configuration to import the unstable release packages
fre has quit [(Ping timeout: 276 seconds)]
<gchristensen>
though I'm not sure how it'd play with kernel modules
<gchristensen>
but importing unstable packages in to their stable system is a fairly common thing people do
Wizek has joined #nixos
mudri has joined #nixos
Wizek_ has joined #nixos
<zx2c4>
interesting
<fpletz>
gchristensen: note that this is in my opinion an anti-pattern because all other dependencies are fetched in their unstable version… and if you import a nixpkgs version in your configuration.nix you have to make sure to update it to current revisions to get updates
<gchristensen>
yeah, I agree... hmm
<fpletz>
zx2c4: we have a few other packages with the same problem and were thinking about a general solution but are not sure yet about the interface… but that should ideally only be one or two lines of code
<gchristensen>
zx2c4: do you have a timeline on becoming "stable" / issuing releases?
eacameron has joined #nixos
IITaudio has joined #nixos
<zx2c4>
gchristensen: sometime this year
<zx2c4>
things are already stabilizing pretty rapidly
<zx2c4>
im going to NDSS in a few weeks to discuss with academics
<zx2c4>
but i think things are mostly stablized
<zx2c4>
why not just backport each snapshot?
<IITaudio>
ehm.. how do I fix "current Nix store schema is version 10, but I only support 7"?
mizu_no_oto has joined #nixos
<gchristensen>
zx2c4: ^ to prevent this problem >.< :)
<zx2c4>
gchristensen: interesting
eacameron has quit [(Ping timeout: 240 seconds)]
mudri has quit [(Ping timeout: 268 seconds)]
<Profpatsch>
fpletz: Do you use it actively?
<Profpatsch>
Because it just doesn’t work here.
<Profpatsch>
Or do I have to enable avahi.publish.workstation?
ixxie has quit [(Remote host closed the connection)]
IITaudio has quit [(Remote host closed the connection)]
<SuprDewd>
regarding the "Move the Wiki!" milestone, it isn't clear to me where it's being moved. can someone elaborate ?
<fpletz>
Profpatsch: hrm, I can test later, we generally set PULSE_SERVER instead of pavucontrol (which gets the sinks from avahi)… there are some bugs with programs using alsa if per-stream remote sinks with pavucontrol are used
<fpletz>
Profpatsch: not in the office yet :)
dramforever has joined #nixos
<NixOS_GitHub>
[nixpkgs] rickynils pushed 2 new commits to master: https://git.io/vDiG8
<NixOS_GitHub>
nixpkgs/master 232639c Rickard Nilsson: haskellPackages: Remove jailbreak for xxhash and Glob (fixed upstream)
derjohn_mob has quit [(Ping timeout: 258 seconds)]
dramforever has quit [(Read error: Connection reset by peer)]
<mbrgm>
does someone know how to queue a systemd service _after_ udev has done the predictable network interface renaming?
<fpletz>
Profpatsch: I don't think those two commits broke anything though
FRidh has joined #nixos
<gchristensen>
zx2c4: what I'm thinking is updating the description of wireguard to indicate that it is a pre-prelease, is experimental and not to be depended upon for security, and then moving forward with regular backports to stable
cmercier has quit [(Remote host closed the connection)]
<gchristensen>
fpletz: ^
<zx2c4>
gchristensen: that works for me
<zx2c4>
so long as the package name doesnt change, i think thats a good plan
Shou has quit [(Ping timeout: 255 seconds)]
cmercier has joined #nixos
<zx2c4>
fpletz: we met at FOSDEM right?
riptawr has quit [(Ping timeout: 256 seconds)]
<gchristensen>
zx2c4: why not the package name?
<gchristensen>
(we'd put in an alias / notice about the rename)
<fpletz>
zx2c4: yeah, but speaking to you right after the talk wasn't a good idea %)
<zx2c4>
fpletz: haha
<zx2c4>
gchristensen: overly alarmist and not as smooth when it is no longer experimental
<NixOS_GitHub>
[nixops] rbvermaa pushed 2 new commits to master: https://git.io/vDiZs
<NixOS_GitHub>
nixops/master 8293ddf Niklas Hambüchen: Atomically set permissions on nixops state file creation. (See #594)...
<NixOS_GitHub>
nixops/master 3a8410e Rob Vermaas: Merge pull request #604 from nh2/statefile-atomic-create-permissions...
<gchristensen>
could be, though when it is stable we can put the alias / notice in again and it'd be pretty smooth. re alarmist: maybe so.
bennofs has joined #nixos
dramforever has joined #nixos
Shou has joined #nixos
uralbash has quit [(Ping timeout: 255 seconds)]
uralbash has joined #nixos
<Profpatsch>
fpletz: Do you have your setup online somewher?
<Profpatsch>
*er
<Profpatsch>
Our streaming machine is one gruesome hack (autologin of the user to get a pulse & avahi user-instance …)
<Profpatsch>
Thanks, Poettering
mguentner2 is now known as mguentner
<NixOS_GitHub>
[nixpkgs] FRidh pushed 1 new commit to master: https://git.io/vDicC
<NixOS_GitHub>
nixpkgs/master 51a338a Frederik Rietdijk: sabnzbd: 1.1.0 -> 1.2.0
<fpletz>
Profpatsch: we're using the systemd-wide daemon but the code is not public, I can probably paste you our config, one sec
Shou has quit [(Ping timeout: 240 seconds)]
<Profpatsch>
fpletz: Hm, how do you get avahi working then?
<NixOS_GitHub>
[nixpkgs] FRidh pushed 1 new commit to master: https://git.io/vDic1
<NixOS_GitHub>
nixpkgs/master 67bd8d5 Frederik Rietdijk: sabnzbd: fix dependencies
<Profpatsch>
The module doesn’t work for the system-wide pa afaik
<zx2c4>
if you integrate this into the nix configuration,
<zx2c4>
that'd be mega cool
<zx2c4>
you'd be the first distro to do something like that
<zx2c4>
"leak free VPN"
<gchristensen>
I was surprised to see NixOS so low in the install instructions, btw :innocent:
<gchristensen>
that would be cool
<fpletz>
and we could probably pretty easily put our networking setup for physical interfaces into a dedicated namespace… we have systemd after all! \o/
<fpletz>
zx2c4: I've put it on my todo-list :)
<zx2c4>
Cool
<zx2c4>
gchristensen: haha maybe ill move it up
<zx2c4>
i toiled in order to get yall an icon in font linux!
<gchristensen>
it is a great icon, I was super impressed :) can we somehow make that available to our users in a "press kit" or something?
<NixOS_GitHub>
[nixpkgs] dezgeg pushed 1 new commit to master: https://git.io/vDilK
<NixOS_GitHub>
nixpkgs/master c867981 Tuomas Tynkkynen: amoeba: init at 1.1...
FRidh has quit [(Remote host closed the connection)]
mudri has quit [(Ping timeout: 258 seconds)]
IITaudio has joined #nixos
takle has quit [(Remote host closed the connection)]
Unode has quit [(Read error: Connection reset by peer)]
Unode has joined #nixos
digitalmentat has joined #nixos
Dezgeg has quit [(Ping timeout: 264 seconds)]
Dezgeg has joined #nixos
<Rizy>
hi, i'm using nix in osx and try to follow the nixpill tutorial. When I want to build derivation on C files. it says "gcc-wrapper-5.4.0/bin/as: assembler (clang) not installed"
<Rizy>
i've tried install clang, but the error still there
<IITaudio>
guys extremly bad situation here: I moved the folder /nix/var/nix/db (trying to solve the wrong schema version) and now every command I run fails .. any idea?:)
<clever>
IITaudio: move it back, and then look into what changed inside it
<digitalmentat>
globin, are you around?
<clever>
IITaudio: what exactly is the schema error msg?
<IITaudio>
clever: the mv command fails command_not_found_handler:4: /run/current-system/sw/bin/command-not-found: bad interpreter: /nix/store/p7k810475bbk0v9wi418sz8jh3dzlpiz-perl-5.22.2/bin/p: no such file or directory
<mguentner>
clever: I assume nix 1.12 upgraded the schema and 1.11 fails
<clever>
IITaudio: ah, sounds like nix already started to mass-delete things in /nix/store, because the db said they shouldnt exist
<clever>
IITaudio: simplest fix is to just boot from an install media, and re-run nixos-install, that will rebuild everything from your existing configuration.nix
<IITaudio>
mguenter: yes, for some reason it was the 1.12 upgrade (that I can't remember to have done in any way) and using 1.12 itself didn't worked.
<globin>
digitalmentat: yep
<digitalmentat>
globin, (this is @ixmatus on Github)
<clever>
IITaudio: if nix hadnt mass-deleted things, this could have probably have been fixed
derjohn_mob has joined #nixos
<clever>
IITaudio: i have accidentaly installed my nixos with nix master before, and had to manualy downgrade the db
<digitalmentat>
I've been wanting to move away from the hand-rolled GCC invocation for a while but not quite sure how I should go about that
<digitalmentat>
is there a "with GCC" derivation wrapper to do what I want?
<digitalmentat>
stdenv.cc perhaps?
<IITaudio>
clever: clever ok I've got a shell (thanks to emacs) how to I manually downgrade it?
<clever>
digitalmentat: stdenv.mkDerivation will have gcc in $PATH
<clever>
IITaudio: i opened the nix source and looked at the differences, at the time, it was just 2 or 3 new columns, that the old version could safely ignore
<digitalmentat>
clever, yes I know beacuse it's working but I want to try and use an automatically produced GCC invocation instead of a hand-rolled one
<clever>
IITaudio: so i just manualy set the schema version back to a lower number, and it was done
<digitalmentat>
but we don't have a Makefile in there so I don't think the default builder will work
<IITaudio>
clever: is the db in sqlite3? 'cause I can't open it .-.
<digitalmentat>
clever, will look thanks for the tip
<clever>
IITaudio: how big is the file and what error does it give?
<digitalmentat>
clever, do you know if the compiler hardening flags are already set (as in, is gcc wrapped with them?)
<digitalmentat>
even with a manual invocation?
<clever>
digitalmentat: they are, stdenv always uses the wrapped gcc
<digitalmentat>
got it, okay, it was confusing to me but that's exactly what I wanted to know; thank you
roconnor__ has joined #nixos
<IITaudio>
clever: ok I think it was opened by another process, now I can open it..
systemfault has joined #nixos
ttuegel has joined #nixos
mortum5 has quit [(Quit: Leaving)]
Shou has quit [(Ping timeout: 256 seconds)]
<ttuegel>
How do I stop Nix from gzipping man pages in a package? The package's Makefile already gzips them, so they end up gzipped twice, and man doesn't like that.
excelangue has quit [(Ping timeout: 260 seconds)]
wofom has quit [(Read error: Connection reset by peer)]
<clever>
ttuegel: /home/clever/nixpkgs/pkgs/build-support/setup-hooks/compress-man-pages.sh: if [ ! -d "$dir/share/man" ]; then return; fi
IITaudio has quit [(Read error: Connection reset by peer)]
<clever>
ttuegel: dontGzipMan = true; in the derivation
<ttuegel>
clever: Thank you! I was looking in setup.sh; I couldn't find the right setup-hook.
<clever>
ttuegel: and going by line 9, it will ignore any files ending in .gz or .bz2, so id say the package is probably wrong in not using those extensions
<ttuegel>
clever: Then the setup hook is broken. The manpages are named xxx.1.gz.gz.
<clever>
ah
<ttuegel>
That's what man doesn't like :)
<NixOS_GitHub>
[nixpkgs] FRidh pushed 1 new commit to python3: https://git.io/vDiz5
<NixOS_GitHub>
nixpkgs/python3 d8e39b8 Frederik Rietdijk: Python 2.7: set higher priority...
<ttuegel>
Just guessing, but those globs probably need to be escaped.
excelangue has joined #nixos
<clever>
set -x would probably help
<clever>
postInstall = "set -x";
<clever>
then bash will spew out a trace of every command it runs during fixup
<ttuegel>
Aha!
olesk has quit [(Ping timeout: 255 seconds)]
systemfault has quit [(Quit: Bye!)]
<bennofs>
ttuegel: oh, if compressManPages runs in a directory that has a .gz file, then GLOBIGNORE will only contain that .gz file, no?
<clever>
yeah, thats a possibility
wizeman has quit [(Ping timeout: 240 seconds)]
<bennofs>
ttuegel: could you test if putting "" around that GLOBIGNORE fixes the issue?
gridaphobe has quit [(Ping timeout: 258 seconds)]
<ttuegel>
bennofs: Yeah, that's what I'm trying...
wizeman has joined #nixos
<clever>
i dont think "set -x" shows bash variables being set, that needs bash -v, which is harder to set in nix-build
<dweller>
hum, for some reason, when doing nix-env -i all, its trying to rebuild gnutls
mbrock has quit [(Ping timeout: 258 seconds)]
CodingWithClass has quit [(Ping timeout: 276 seconds)]
<dweller>
and failing because of missing netstat
BramD has quit [(Ping timeout: 258 seconds)]
monad_cat has quit [(Ping timeout: 258 seconds)]
pkogan has quit [(Ping timeout: 258 seconds)]
scode has quit [(Ping timeout: 258 seconds)]
ocharles has quit [(Ping timeout: 258 seconds)]
gridaphobe has joined #nixos
<dweller>
same with nixos-rebuild and im pretty sure i didnt touched any of gnutls inputs nor gnutls itself in my repo
scode has joined #nixos
BramD_ has joined #nixos
CodingWithClass has joined #nixos
pkogan has joined #nixos
Rotaerk has quit [(Quit: Leaving)]
monad_cat has joined #nixos
mbrock has joined #nixos
ocharles has joined #nixos
mekeor has joined #nixos
mekeor has left #nixos []
<joachifm>
roundhouse: torbrowser works for me; I'm thinking ~/.torbrowser4 may have become stale in your case
<roundhouse>
joachifm, that fixes it. Thanks alot. I'll close the issue
<ttuegel>
clever: bennofs: 'set -x' does show variables being set, and bash knows to implicitly quote whatever you set for GLOBIGNORE. So, GLOBIGNORE is being set correctly, but it still doesn't work. :-(
<globin>
digitalmentat: `hardeningEnable=pie gcc ...` should enable piw
<globin>
pie*
<globin>
digitalmentat: or exporting the env var prior to that
<digitalmentat>
globin, does hardeningEnable = [ "pie" ]; work?
sellout- has joined #nixos
<globin>
digitalmentat: there is a checksec derivation to check what is enabled on the binary
<globin>
digitalmentat: let me look at the code, but normally yes
<NixOS_GitHub>
[nixpkgs] ttuegel pushed 1 new commit to master: https://git.io/vDia7
<NixOS_GitHub>
nixpkgs/master d43738a Thomas Tuegel: notmuch: don't double-compress man pages
<digitalmentat>
globin, oh I was looking for just something like that (checksec)
iyzsong has quit [(Quit: bye)]
cpennington has joined #nixos
<NixOS_GitHub>
[nixpkgs] joachifm closed pull request #22510: Fix/ngspice add XSPICE and CIDER support (master...fix/ngspice-update) https://git.io/vD8Ov
<globin>
digitalmentat: should be fine :>
<digitalmentat>
thanks globin :)
<NixOS_GitHub>
[nixpkgs] lverns opened pull request #22794: mnemosyne: move generated locale files to correct location (master...mnemosyne-fix-locale) https://git.io/vDiwz
takle has joined #nixos
<NixOS_GitHub>
[nixpkgs] joachifm pushed 2 new commits to master: https://git.io/vDiw6
<NixOS_GitHub>
nixpkgs/master af09e72 Neil Mayhew: spideroak: enable it to appear in the application menus...
<NixOS_GitHub>
nixpkgs/master 943951f Joachim F: Merge pull request #22529 from neilmayhew/fix/spideroak...
<bennofs>
ttuegel: seems like *.gz does not match foo/xxx.gz, since it wouldn't be matched by echo *.gz either
<bennofs>
ttuegel: perhaps a better fix would be something like find -not -name "*.gz" ?
johann__ has quit [(Read error: Connection reset by peer)]
johann__ has joined #nixos
eacameron has joined #nixos
kthnnlg has joined #nixos
MercurialAlchemi has quit [(Ping timeout: 240 seconds)]
fre has joined #nixos
IITaudio has joined #nixos
<domenkozar>
very unscientific measurement comparison, but here's what happened once we switched on parallel building and ghc section splitting for ghc802:
<domenkozar>
haskell-src-exts builds in 240s instead of 850s
jgertm has quit [(Ping timeout: 240 seconds)]
<IITaudio>
clever: ok I went with the nixos-install from a live medium
<bennofs>
domenkozar: nice!
<kthnnlg>
Hi All, I would like to increase the permitted number of open fds a la `ulimit -n $x` for some $x of my choosing. Is there a configuration line that I can add to my configuration.nix file to achieve this effect? thanks
<bennofs>
domenkozar: and output size went down as well
<domenkozar>
bennofs: not for all packages
<domenkozar>
but yeah :)
<IITaudio>
clever: hopefully it will work ./
<bennofs>
domenkozar: still, 60MB output for haskell-src-exts? That's a lot
jgertm has joined #nixos
<avn>
bennofs: may be costs to split generated docs?
<mbrgm>
is there a way to indent an interpolated multiline string inside a multiline string?
bennofs has quit [(Quit: WeeChat 1.7)]
bennofs has joined #nixos
<domenkozar>
bennofs: did you do more work on stack2nix?
<bennofs>
domenkozar: for me, I first wanted to get stackage2nix to reproduce the stackage-builder's test results without any additional configuration (we do a lot of things in configuration-common.nix, which really would be unsuited for stack2nix)
davidak has joined #nixos
erasmas has joined #nixos
<domenkozar>
why unsuited, it's a nix package at the end :)
<bennofs>
domenkozar: that works now, I've got a package set that can be generated fully automatically from a stackage snapshot and does not need any manual overrides (test disables or similar) except the ones from configuration-nix in nixpkgs (configuration-nix only patches stuff like hardcoded paths, but does not modify the build options)
<bennofs>
i was planning to look at where to hook into stack next (maybe stack2nix will even be unrelated to stackage2nix, but together with stackage2nix we could make a binary cache to use with stack2nix)
<ikwildrpepper>
teh[m]: were you the one who implemented the digital ocean backend?
<domenkozar>
ikwildrpepper: yes
<ikwildrpepper>
domenkozar: have you used it?
<domenkozar>
nope
<domenkozar>
but someone else did
<ikwildrpepper>
domenkozar: as I keep getting 'error: No token provided. Please use a valid token'
<domenkozar>
ikwildrpepper: by following docs?
<clever>
domenkozar: one of the datacenters i'm at now is just horid, dedicated servers get a private ip via dhcp, VM's get no ip, and you must configure a static to get anything working right
<bennofs>
domenkozar: but i need to investigate first how stack deals with the hfsevents/hinotify on mac vs linux issue first
<clever>
domenkozar: i based my latest intall on the assumption that dhcp would give me a private + nat, only to discover that vm's behave differently
<domenkozar>
clever: sounds like user metadata on EC2 :)
<clever>
domenkozar: the vm cant even access metadata, it winds up on a 169.254 ip, with no access to anything
<bennofs>
domenkozar: i want stack2nix to be as equavilent to building with stack as possible. if it works with stack, it should work with stack2nix and the other way around
<domenkozar>
bennofs: I still don't understand how you plan to get it to work with IFD, since for Nix you need a hash for the git checkout and stack doesn't provide one
<clever>
domenkozar: but by chance, i caught its ipv6 link-local in tcpdump when sniffing the vm's mac
<clever>
and that got me in
<domenkozar>
clever: a classic :D
<bennofs>
domenkozar: hash for what git checkout?
<bennofs>
domenkozar: extra-deps?
<clever>
domenkozar: but what does it say about the datacenter, when 1 server can sniff broadcast traffic from another server?
<domenkozar>
no, location: type: git
<domenkozar>
using pkgs.fetchgit in nix, you need to provide sha256
<clever>
domenkozar: i think i could MITM any system in the building...
<bennofs>
domenkozar: yeah, i'm planning on letting stack fetch the git deps prior to generating the nix derivation
<bennofs>
domenkozar: since the way stack is implemented right now, it does that very early in the startup process
<bennofs>
domenkozar: so to nix, the git checkout would just be another local package source
<domenkozar>
bennofs: well that won't work with IFD
<domenkozar>
I mean you won't be able to autogenerate this in sandbox
<domenkozar>
since you'll have to run stack
<domenkozar>
which requires networking
<domenkozar>
it will work with Nix 1.12
<domenkozar>
which has builtins.fetchgit
digitalmentat has quit [(Quit: Leaving)]
<bennofs>
domenkozar: hmm, i don't really need IFD for stack2nix
<NixOS_GitHub>
[nixpkgs] dtzWill closed pull request #22138: WIP: Add support for building Swift packages with deps (master...feature/swift-build-support) https://git.io/vMxpL
<bennofs>
domenkozar: oh right. that comment was wrong, you're right, autogenerating in a sandbox is never going to work anyway
<bennofs>
domenkozar: i think the best way to go about this is that stack just encodes what it would do into nix expressions and then uses nix to perform the build
<domenkozar>
so there are two uses cases here
<domenkozar>
actually let me write to that issue so it's googlable
<ikwildrpepper>
domenkozar: yeah, as far as I can see, I am following the docs
excelangue has quit [(Ping timeout: 240 seconds)]
<domenkozar>
ikwildrpepper: best to email teh, he's super responsive there :)
excelangue has joined #nixos
<mbrgm>
I'm in the situation where I want to force a string options for a nixos module to be != "". is the preferred option an assertion, a type check or simply not providing a default value so the 'used but not defined' error kicks in?
<mbrgm>
option*
tm__ has quit [(Ping timeout: 252 seconds)]
<NixOS_GitHub>
[nixpkgs] basvandijk opened pull request #22797: haskell: support Haskell Program Coverage (HPC) (master...hpc) https://git.io/vDiPW
Shou has joined #nixos
tm__ has joined #nixos
<NixOS_GitHub>
[nixpkgs] edolstra pushed 3 new commits to master: https://git.io/vDiPa
<NixOS_GitHub>
nixpkgs/master 23fee8b Tuomas Tynkkynen: nixos-rebuild: Support passing e.g. '-j8'...
<NixOS_GitHub>
nixpkgs/master 2000f09 Tuomas Tynkkynen: nixos-rebuild: Don't build nixos-rebuild with --fast
IITaudio has quit [(Remote host closed the connection)]
bennofs1 has joined #nixos
IITaudio has joined #nixos
binaryphile has joined #nixos
asymmetric_ is now known as asymmetric
<fpletz>
is it just me or are github reactions broken? :/ "You can't perform that action at this time."
bennofs1 has quit [(Quit: WeeChat 1.6)]
<IITaudio>
clever: thanks for your help! It tooks hours but at the end nixos re-installed :)
<clever>
IITaudio: in theory, the worst that can happen there is that all of /nix/store is lost, but it can just rebuild it from the binary cache and config
<clever>
IITaudio: and maybe the nix-env -i profiles are gone, but those should be minor
<mbrgm>
is it ok to ask general systems questions here? or just nix* related ones?
digitalmentat has joined #nixos
<IITaudio>
clever: I had to delete manually all the nix/store because of lack of disk space; that's the reason why it took so long to reinstall it
<clever>
IITaudio: ah
MarcWebe1 has quit [(Quit: leaving)]
<clever>
IITaudio: on a list of more insane stuff i did yesterday, i attempted to switch a remote server from ubuntu to nixos, without any physical or console access
MarcWeber has joined #nixos
<IITaudio>
(also, being on a live system I used that time to increase the partition size a bit)
<asymmetric>
is there a guid from migrating to nix (ubuntu) to nixos?
<asymmetric>
*from nix to nixos
<asymmetric>
*guide
<gchristensen>
maybe just try that one again :)
<IITaudio>
clever: ehm.. with a partition available you could install nix on it, reinstall grub to point to this, reboot and you are done, right?
<clever>
IITaudio: there was ~1gig free on the hdd
binaryphile has quit [(Quit: Leaving)]
<IITaudio>
clever: (okay that's strange. nix-collect-garbage freed 78Gb. I run it yesterday)
stepcut has quit [(Remote host closed the connection)]
Itkovian has joined #nixos
<clever>
asymmetric: depends a lot on what kind of migration you want to do, simplest would be to just backup your ~/.nixpkgs/ folder, boot from the install iso, and follow the instructions on the site
<clever>
asymmetric: which will involve formating the hdd
<clever>
asymmetric: but if you have experience with linux, you could make a second rootfs and put nixos on that
<NixOS_GitHub>
[nixpkgs] benley pushed 2 new commits to master: https://git.io/vDiNj
<NixOS_GitHub>
nixpkgs/master a50b4d0 Vincent Demeester: docker: 1.13.0 -> 1.13.1...
<NixOS_GitHub>
nixpkgs/master b42f820 Benjamin Staffin: Merge pull request #22745 from vdemeester/docker_1_13_1...
cmercier has quit [(Ping timeout: 255 seconds)]
civodul has quit [(Remote host closed the connection)]
<NixOS_GitHub>
[nixpkgs] benley pushed 2 new commits to master: https://git.io/vDiAs
<NixOS_GitHub>
nixpkgs/master 40dc5e6 Benjamin Staffin: Merge pull request #22771 from izuk/bazel...
<asymmetric>
clever: thanks! your kexec project seems pretty cool too
<clever>
asymmetric: main issue with the kexec stuff, is that you must get everything perfect, if you miss anything, you wind up with http://i.imgur.com/s23nqih.jpg and no way to fix it
<clever>
but i did just have an idea on how to work around that somewhat
<asymmetric>
yeah, seems cool but not sure i'll go for that, since it's my work laptop we're talking about :)
<clever>
with a few more changes, i could embed the netboot kernel+initrd into the grub.cfg for nixos
<clever>
if you have physical access and can boot from cd/usb, its trivial to fix these kinds of problems
<clever>
the reason its an issue with kexec, is that its meant for machines you lack physical access to
<clever>
asymmetric: the server in that screenshot would take a 89 hour drive if i wanted to access it...
<asymmetric>
that's... far
<clever>
and i think i would have to cross 3 international borders
<asymmetric>
but trivial as in? how would you go about fixing the devserver issue in the img?
kthnnlg has quit [(Remote host closed the connection)]
<clever>
asymmetric: boot from the nixos iso, fix a single mistake in configuration.nix, and re-run nixos-install, and its fixed
<srhb>
Hm, remind me what I missed again. I have overrided withPrimus = true for steam, but I still get: /bin/sh: primusrun: command not found
digitalmentat has quit [(Quit: Leaving)]
<srhb>
when I launch a game that has its launch options set as primusrun %COMMAND%
<srhb>
I think I fixed this before...
dannyg has left #nixos []
<asymmetric>
but you'd have to know what the error in nix is i suppose
<asymmetric>
i'm not a proficient nix user yet
<asymmetric>
so i'm afraid i'd get lost
xwvvvvwx has joined #nixos
<clever>
asymmetric: yeah, main thing physical access gives is trial&error
<clever>
asymmetric: so you can just reboot and try again until it works
jgertm has quit [(Ping timeout: 255 seconds)]
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
ryanartecona has joined #nixos
ambro718 has joined #nixos
bennofs has joined #nixos
bennofs has quit [(Client Quit)]
Itkovian has joined #nixos
olesk has joined #nixos
bennofs has joined #nixos
pi3r has joined #nixos
Shou has quit [(Ping timeout: 268 seconds)]
georges-duperon has joined #nixos
<IITaudio>
clever: ok, added the table to the db and now it seems to be working. I wonder for how long
<IITaudio>
clever: first time I have system-problems on nioxs in more then 2 years
cpennington has quit [(Ping timeout: 255 seconds)]
<clever>
IITaudio: i was doing some pretty crazy things when i had caused my schema problem
<clever>
IITaudio: i wound up on a road trip, with half of a gentoo install (i was cleaning the hdd out with the move command), so it barely booted, but it had a year old copy of nix installed
<srhb>
I have this in my overrides: mySteam = self.steam.overrideAttrs (attrs: { withPrimus = true; });
<srhb>
But now that I look at it, it seems to be related to the chrootenv, not steam itself.
<clever>
IITaudio: i grabbed the storepath for a recent nix build off hydra using lynx (a text mode browser) and ran "nix-store -r /nix/store/foo-nix-1.2.3" to upgrade the nix, and used that to finish the install
<clever>
IITaudio: but my critical mistake, i was browsing nix master on hydra
<IITaudio>
clever: sometimes I do install packages from nixpkgs master with nix-env -f . -i package, but never thought to update nix that way
<clever>
IITaudio: the key here, is that the year old nix cant read the recent nixpkgs, so it cant compute the storepath for its own update
<clever>
IITaudio: but the binary cache stuff hasnt changed, and still works across a much wider version range
Gohla has quit [(Ping timeout: 255 seconds)]
<IITaudio>
clever: yeah I understand. Ok now just 4 hours late I can go back working. Thanks again for your precious help :)
kthnnlg has quit [(Remote host closed the connection)]
takle has quit [(Ping timeout: 260 seconds)]
xwvvvvwx has joined #nixos
danharaj has joined #nixos
cpennington has quit [(Ping timeout: 240 seconds)]
danielrf has joined #nixos
takle has joined #nixos
Shou has quit [(Ping timeout: 240 seconds)]
jensens has quit [(Ping timeout: 258 seconds)]
bennofs has quit [(Quit: WeeChat 1.7)]
<benley>
any real objection to merging #22767 at this point?
<benley>
(the useSandbox=true one)
<gchristensen>
benley: see eelco's comment on it
<benley>
I just read through the whole thread on 179 again
<benley>
it's not entirely clear to me if it's a big deal right now or not :-P
ttuegel has joined #nixos
<gchristensen>
same :)
<gchristensen>
we're very rapidly rising in # of PRs I think
<gchristensen>
worse, hard PRs :) not just easy merges
<NixOS_GitHub>
[nixpkgs] ttuegel pushed 1 new commit to master: https://git.io/vDPGa
<NixOS_GitHub>
nixpkgs/master ce85929 Thomas Tuegel: offlineimap: install man pages
derjohn_mob has quit [(Remote host closed the connection)]
cpennington has joined #nixos
<NixOS_GitHub>
[nixpkgs] jwiegley pushed 1 new commit to master: https://git.io/vDPG7
<NixOS_GitHub>
nixpkgs/master 3a0efcc John Wiegley: configuration-common: http-api-data is now at version 0.3.5
jensens has joined #nixos
<ttuegel>
gchristensen: Here is where easier testing helps, I think. To me, any PR that doesn't break evaluation and doesn't break any builds is "easy". The only hard part is reviewing them until they reach that state.
<gchristensen>
ttuegel: by the way, is there any way you can take over and own the removal of kde4? I really want it to happen, but don't think I'm skilled enough.
<ttuegel>
gchristensen: No problem! I probably won't have time to work on that until the weekend, but I'll make it happen.
<gchristensen>
<3 <3 <3! awesome :D
simendsjo has joined #nixos
<NixOS_GitHub>
[nixpkgs] ttuegel pushed 5 new commits to master: https://git.io/vDPnl
<NixOS_GitHub>
nixpkgs/master 5ebf37f Michael Alan Dorman: org-packages: 2017-02-13
<NixOS_GitHub>
nixpkgs/master 58b5c49 Michael Alan Dorman: elpa-packages: 2017-02-13
<NixOS_GitHub>
nixpkgs/master cd90c7f Michael Alan Dorman: melpa-stable-packages: 2017-02-13
jensens has quit [(Ping timeout: 260 seconds)]
davidak has joined #nixos
athan has joined #nixos
derjohn_mob has joined #nixos
<NixOS_GitHub>
[nixpkgs] ttuegel pushed 9 new commits to staging: https://git.io/vDPnb
<NixOS_GitHub>
nixpkgs/staging 71812be Frederik Rietdijk: kwooty: kde-apps url's have changed
<NixOS_GitHub>
nixpkgs/staging d375d7b Frederik Rietdijk: kmplayer: file not available via new mirror either
<NixOS_GitHub>
nixpkgs/staging 63bff03 Frederik Rietdijk: kronometer: 2.1.0 -> 2.1.3
Wizek has quit [(Ping timeout: 240 seconds)]
<NixOS_GitHub>
[nixpkgs] ttuegel closed pull request #22664: melpa: fixed hashes of frame-cmds, frame-fns and zoom-frm (release-16.09...fix-emacs-packages) https://git.io/vDVNw
Wizek has joined #nixos
<NixOS_GitHub>
[nixpkgs] grahamc pushed 2 new commits to master: https://git.io/vDPcn
<NixOS_GitHub>
nixpkgs/master d0a0867 Graham Christensen: nix-daemon: default useSandbox to true
<NixOS_GitHub>
nixpkgs/master 3be1388 Graham Christensen: Merge pull request #22767 from grahamc/sandbox-by-default...
<gchristensen>
benley: ^
derjohn_mob has quit [(Ping timeout: 268 seconds)]
<benley>
woot :-)
* gchristensen
cautiously hopes it doesn't get reverted
<pikajude>
oh boy
<pikajude>
oh boy!!
<NixOS_GitHub>
[nixpkgs] ttuegel closed pull request #21366: digikam: Add patch to fix compilation against Lensfun 0.3.2 (master...digikam_patch_lensfun_0.3.2) https://git.io/vDPcP
<benley>
It seems like a pretty sane default; as someone pointed out, you can always disable it if it affects your setup too much.
<benley>
I've been running with sandboxing enabled for a super long time :P
<ttuegel>
Also, we require PRs to be tested with it enabled, so...
Sonarpulse has joined #nixos
<benley>
is it turned on in the config that travis-ci runs?
<NixOS_GitHub>
[nixpkgs] ttuegel closed pull request #7277: Qt: build with -pipe for some amount of parallelism (master...slightly-faster-qt-build) https://git.io/veMxa
<pikajude>
performance? perish the thought
<ttuegel>
I'm surprised it affects performance; I thought Linux user namespaces were cheap. I haven't checked, though.
<dtzWill>
thanks for the ping, curious what the reason is--so we can work on fixing whatever the issue(s) are :).
<benley>
running them is cheap, but it seems there's a time penalty in setting them up
fhoffmeyer2 has quit [(Read error: Connection reset by peer)]
<gchristensen>
for 15 sweet minutes we had the sandbox defaulted to on :)
<NixOS_GitHub>
[nixpkgs] ttuegel closed pull request #18455: add .overrideDerivation and .overrideAttrs to packages created with callPackages (master...override) https://git.io/vi4GG
<ttuegel>
benley: That seems strange. Fork, unshare, bind-mount. I wonder what the bottleneck is.
<benley>
Read through the discussion on #179 if you haven't already.
jgertm has quit [(Ping timeout: 264 seconds)]
takle has quit [(Remote host closed the connection)]
<johnw>
peti: ping
t7 has quit [(Remote host closed the connection)]
<ttuegel>
Yeah, I don't understand why the whole store can't be bind mounted once. The package hashes are supposed to provide isolation.
<gchristensen>
heh
<ttuegel>
I understand why bindmounting every individual dependency is slow. :-)
xwvvvvwx has quit [(Ping timeout: 240 seconds)]
<gchristensen>
some packages go spelunking through the filesystem looking for what they want
fhoffmeyer has joined #nixos
<clever>
gchristensen: and some programs try to browse /nix/store at runtime, and then just lock up solid due to the sheer number of files
<clever>
i cant even use file->open in one of the text editors ive ran in the past
<ttuegel>
gchristensen: If the hashes aren't security, then we should just use user namespaces for everything and bind mount into /usr so we don't have to patch every frakking package!
<benley>
bindmounting the whole store seems like a reasonable compromise to have available
<gchristensen>
heh
<ttuegel>
clever: tab-completion into /nix/store in ZSH hard-locks my computer sometimes. :-\
<clever>
ttuegel: same on ZFS here
<benley>
clever: I accidentally opened /nix/store/.links in Nautilus one time. Bad idea.
<dtzWill>
lmao yes
<ttuegel>
Maybe /nix/store should be mounted as r-- instead of r-x in the global namespace on NixOS so you can't do that!
<benley>
that also seems fairly sane
<gchristensen>
the hashes shouldn't be considered security because it is depending on obscurity
<clever>
ttuegel: i have thought about that for keeping passwords in the store, but you can still recursively follow the references
<benley>
and you can still use nix-store --export iirc
<ttuegel>
clever: Right, I'm not thinking of it as security, just as insurance against locking your machine by doing 'cd /nix/store; ls'
<clever>
ttuegel: ah
<clever>
ttuegel: though nix-collect-garbage still does it, when it tries to stat every entry in .links
<ttuegel>
clever: I think that does so in the daemon namespace, though.
t7 has joined #nixos
<clever>
yeah
<clever>
but that doesnt stop it from crippling my machine
<ttuegel>
gchristensen: OK, then we should forget this whole /nix/store business and put every exe in a user namespace with it's own /usr.
<ToxicFrog>
ttuegel: considering how often I have to go rummaging around in /nix/store to debug things, that would make me spectacularly unhappy
<ToxicFrog>
Making it a-x, I mean
<ttuegel>
clever: Are you using an SSD? And do you have real-time discard enabled?
<clever>
ttuegel: the problem is the reads not writes, 2 SSD's in a mirror on ZFS
<gchristensen>
only if you're looking to the store for security
<clever>
ttuegel: when ZFS gets low on ram, it goes into a spinlock to wait for more to free up, and sometimes it spinlocks all 8 cores at once, so nothing can ever become free
<ttuegel>
clever: nix-collect-garbage locks my machines because of all the discards. But, switching to fstrim fixed that.
olesk has quit [(Ping timeout: 258 seconds)]
<ttuegel>
Eww.
stepcut has joined #nixos
<clever>
ttuegel: nix-collect-garbage with a negative --max-freed still locks the machine up
<ttuegel>
gchristensen: We are using "security" to mean different things. Imagine I said "purity" everywhere instead.
<clever>
ttuegel: this limits the GC to keep 20gig free and no more, but it still runs (with a negative limit) when i have over 20gig free
<clever>
ttuegel: and it still scans .links for stuff to junk, when nothing has changed
<gchristensen>
fair enough ttuegel, I stand by my pendantic interpretation though because having hard to find passwords in /nix/store is not a good solution
Wizek_ has quit [(Read error: Connection reset by peer)]
Wizek has quit [(Ping timeout: 240 seconds)]
<ttuegel>
gchristensen: No, but that doesn't have bearing on the question of whether to bind-mount the whole store or just individual paths for a sandboxed builder.
<gchristensen>
yep
<gchristensen>
makes sense
<benley>
would it be possible to bind-mount the whole store, but have the permissions be --x without changing the store permissions outside of the sandbox?
<Profpatsch>
It’s kind of funny that `fix` and `fix'` are in trivial.nix
<gchristensen>
right, Profpatsch?
<ttuegel>
benley: For just the top-level bind point at /nix/store? Sure.
xwvvvvwx has joined #nixos
<benley>
that would go a long way toward discouraging builds from rummaging around in the store, if that is even a concern
<benley>
without inconveniencing interactive users
[0x4A6F] has quit [(Ping timeout: 240 seconds)]
simendsjo has quit [(Quit: ERC (IRC client for Emacs 24.5.2))]
simendsjo has joined #nixos
<MarcWeber>
garbas: rinohtype has dependencies depending on rinohtype. Commenting #self.rinohtype in dependencies propagated list makes the build fail.
<MarcWeber>
Why are cyclic dependencies allowed - or how does pypi install work then?
takle has joined #nixos
ryanartecona has quit [(Quit: ryanartecona)]
simendsjo has quit [(Client Quit)]
simendsjo has joined #nixos
<MarcWeber>
Retrying having rinohtype not depend on the other libraries directly..
olesk has joined #nixos
simendsjo has quit [(Remote host closed the connection)]
takle has quit [(Ping timeout: 260 seconds)]
simendsjo has joined #nixos
ttuegel has quit [(Remote host closed the connection)]
Wizek has joined #nixos
derjohn_mob has joined #nixos
athan has quit [(Quit: leaving)]
DavidEGr1yson is now known as DavidEGrayson
zarel has joined #nixos
<zarel>
hello!
mudri has joined #nixos
rly_ has joined #nixos
phreedom has quit [(Ping timeout: 240 seconds)]
<pikajude>
hi
joelpet has quit [(Remote host closed the connection)]
olesk has quit [(Ping timeout: 240 seconds)]
zarel has quit [(Remote host closed the connection)]
danielrf has quit [(Read error: Connection reset by peer)]
bennofs has quit [(Ping timeout: 245 seconds)]
Sonarpulse has quit [(Ping timeout: 255 seconds)]
roundhouse1 has quit [(Remote host closed the connection)]
stukj has joined #nixos
olesk has joined #nixos
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
zarel has quit [(Ping timeout: 255 seconds)]
Itkovian has joined #nixos
<stukj>
hello. i'm using stumpwm in nixos-unstable. after the last upgrade (13/02/2017) I can not maximize emacs. i don't have problems with chromium or xterm
<stukj>
am I the only one with problems?
devoid has joined #nixos
<danharaj>
i've got linker errors on android and i think it's because i'm not including all the GHC libraries i need to include: http://lpaste.net/352522
simendsjo has quit [(Quit: ERC (IRC client for Emacs 24.5.2))]
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
Itkovian has joined #nixos
ablubla_ has joined #nixos
Itkovian has quit [(Client Quit)]
<ablubla_>
Hey, I want to customize a package in my config.nix by overriding the postInstall. However the post install uses variables from within the package, can I somehow use them ( ${super.luaPath} or something like that?)
Wizek_ has joined #nixos
<benley>
make sure you're using the new overrideAttrs function rather than overrideDerivation for that, and I think you can yes
<srhb>
benley: That's only in unstable, isn't it?
<benley>
yeah, unfortunately
<ablubla_>
:(
Itkovian has joined #nixos
<benley>
tbh I've found it more convenient to keep my own branch of nixpkgs and add commits to it when I want to patch things.
<ablubla_>
hm
<devoid>
Anyone understand the inner workings of nixos-container, specifically how the /nix/store is mounted in the container?
<devoid>
I've read the nixos-container Perl program and tried to understand systemd containers but I am missing something I think...
<ablubla_>
hm, wait, I use nix on arch, and I am on nixpgks-unstable so I should have overrideAttrs, right?
<devoid>
as there are no symlinks to the 'real' /nix/store inside the container's root.
<viric>
hm I can't find any good "block device in userspace"
olesk has quit [(Ping timeout: 260 seconds)]
<mbrgm>
how can I build a nix derivation with debug symbols?
stepcut has quit [(Remote host closed the connection)]
<mbrgm>
ah, just found it :-/
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
<devoid>
I am assuming that systemd-nspawn just mount's /nix/store -> /nix/store though.
<devoid>
Is there a way to export one of these containers to run via Docker?
<ablubla_>
benley: how would I use overrideAttrs to change the --config flag in this example? luaPath and luaCPath are the variables of the package I want to keep https://paste.xinu.at/300iHf/
<niksnut>
devoid: the container mounts /nix/store of the host read-only
DutchWolfie has joined #nixos
<devoid>
It looks like the "nixos-container update test" command runs this nix-env build http://pastebin.com/vB8Ug2R5 under the covers… is there a way to see what derivation that produces and enumerate all of the nix-store entries that are closed under it?
acarrico has quit [(Ping timeout: 256 seconds)]
<benley>
ablubla_: sorry I can't help more right now, gotta run
DutchWolfie has quit [(Quit: Konversation terminated!)]
<NixOS_GitHub>
[nixpkgs] pstn opened pull request #22804: Bumped up dehydrated to 0.4.0 (master...dehydrated) https://git.io/vDP9W
cpennington has quit [(Remote host closed the connection)]
digitalmentat has quit [(Ping timeout: 255 seconds)]
<devoid>
sorry, I haven't worked with nix on this level in a while and even then it was all cargo-cult… the big picture problem is going from configuration.nix -> Docker container (that can run anywhere)
stepcut has joined #nixos
hiratara has quit [(Ping timeout: 255 seconds)]
<devoid>
I have the root dir, now I just need to determine a) what the top-level container derivation is.
mjhoy has joined #nixos
<devoid>
and b) what files in /nix/store that refers to?
joelmo has joined #nixos
<devoid>
For A, I can do `nix-env -vvv` on the install to get the final drv file, but that's not particularly good. I know nix-build outputs the drv path but I don't know how to translate it.
<joelmo>
My systemd user timers stopped working in NixOS, from the log: backup.timer: Failed to enter waiting state: Operation not permitted
hiratara has joined #nixos
<joelmo>
Do anyone use user timers and experience this from unstable?
<devoid>
With the drv file, I can do `nix-store -qR /nix/store/{hash}-nixos-system-{etc}.drv` but that, in general, just lists more drv files.
kampfschlaefer has joined #nixos
FRidh has quit [(Remote host closed the connection)]
<devoid>
Previously I've used nix-store —export and to export then import these closures on different systems.
stepcut has quit [(Remote host closed the connection)]
<benley>
joelmo: are you familiar with pkgs.dockerTools in nixpkgs?
<devoid>
But here I just need to export as a TAR or enumerate the contents...
<benley>
oops that was supposed to be devoid
olesk has joined #nixos
<benley>
devoid: are you familiar with pkgs.dockerTools in nixpkgs?
<devoid>
benley, no, although I'm looking at a few examples now...
<devoid>
benley, is buildImage defined in pkgs/build-support/singularity-tools/default.nix ?
<NixOS_GitHub>
[nixpkgs] basvandijk opened pull request #22805: ghcWithHoogle: fix for ghcjs (master...fix-hoogle-ghcjs) https://git.io/vDPQd
Jackneill has quit [(Read error: Connection reset by peer)]
takle has joined #nixos
Jackneill has joined #nixos
<benley>
devoid: no, it should be in pkgs/build-support/docker/default.nix
<benley>
wow, I don't know what singularity-tools is
<devoid>
this is what frustrates me about nixpkgs, it is impossible to figure out where things are defined.
<benley>
devoid: the first place I look is often pkgs/top-level/all-packages.nix
<benley>
but yeah, I get frustrated by that sometimes too
<NixOS_GitHub>
[nixpkgs] bjornfor opened pull request #22806: Fixups after new security wrapper (master...fixups-after-new-security-wrapper) https://git.io/vDP5P
layus has joined #nixos
civodul has quit [(Ping timeout: 240 seconds)]
civodul has joined #nixos
ttuegel has joined #nixos
takle has joined #nixos
Jackneill_ has quit [(Read error: Connection reset by peer)]
Sonarpulse has joined #nixos
Shou has joined #nixos
mjhoy has quit [(Ping timeout: 260 seconds)]
endformationage has joined #nixos
takle has quit [(Ping timeout: 255 seconds)]
<devoid>
benley, building an example container now to see what its doing
<devoid>
I have a hard time reading and understanding these nix expressions
petar_ is now known as petar
<benley>
the language takes some getting used to, especially if you're not accustomed to functional languages
olesk has quit [(Ping timeout: 245 seconds)]
<NixOS_GitHub>
[nixpkgs] fpletz pushed 1 new commit to master: https://git.io/vDPNJ
<NixOS_GitHub>
nixpkgs/master c2a21d4 Tom Doggett: gcalcli: init at 3.4.0 (#22725)
<ttuegel>
Does anybody know which commit broke the NixOS manual, so I can revert it and get back to rebuilding? ;-)
<MarcWeber>
ttuegel: You can disable it: services.nixosManual.enable = false; # takes too much time to build
<ttuegel>
MarcWeber: Yes, but then I don't get the man pages.
takle has joined #nixos
<ttuegel>
I mean, obviously that's what I'm going to do! :-) But, ugh.
<MarcWeber>
What about your question -> how to get back to rebuilding?
<benley>
err, really? I thought all the manpages came with the nix derivation itself, not the nixosManual service
<MarcWeber>
Then you can use git bisect to find out which commit broke the manual and submit an issue / pull request.
<ttuegel>
benley: Not for nixos-rebuild and friends.
<devoid>
any way to figure out what nix is currently doing? I have a feeling it's compiling a bit more than I want for this container (10 minutes in on a 12 core system)
<pikajude>
not exactly
<pikajude>
you could build -j1 though
<pikajude>
or use ps
hiratara has joined #nixos
<devoid>
oh god, why is it building chromium from source
isHavvy has joined #nixos
mkoenig has joined #nixos
takle has quit [(Ping timeout: 258 seconds)]
ttuegel has quit [(Remote host closed the connection)]
stig_ has joined #nixos
modafinil_ has joined #nixos
joshie_ has joined #nixos
sellout-1 has joined #nixos
dhess` has joined #nixos
clever_ has joined #nixos
peterhoeg_ has joined #nixos
<LnL>
devoid: good luck
jgrasser_ has joined #nixos
markus1199 has joined #nixos
kriztw_ has joined #nixos
jp302 has joined #nixos
clever has quit [(Disconnected by services)]
clever_ is now known as clever
MightyJoe has joined #nixos
systemfault has joined #nixos
dtzWill_ has joined #nixos
Ralith__ has joined #nixos
hiratara_ has joined #nixos
Aexoden_ has joined #nixos
nahamu_ has joined #nixos
pkogan_ has joined #nixos
georges-duperon has quit [(Ping timeout: 252 seconds)]
dvim_ has joined #nixos
eqyiel_ has joined #nixos
abcrawf_ has joined #nixos
pmade_ has joined #nixos
acowleyz has joined #nixos
Sigma[Mtp] has joined #nixos
tylerjl has joined #nixos
Foul has joined #nixos
Sigma[Mtp] is now known as Sigma
pmade_ is now known as pmade
hiratara_ is now known as hiratara
pkogan_ is now known as pkogan
qewzee has joined #nixos
stig_ is now known as stig
modafinil_ is now known as modafinil
cstrahan has joined #nixos
codedmart has quit [(Ping timeout: 263 seconds)]
mguentner has quit [(Ping timeout: 240 seconds)]
stepcut has joined #nixos
rmarko has joined #nixos
edwtjo has joined #nixos
zimbatm has joined #nixos
rmarko is now known as srk
stepcut has quit [(Remote host closed the connection)]
<jchee>
I'm trying to get `firejail` working using nix as my package manager on a centos machine and I'm getting a bunch of euid/egid warnings and then the program fails: http://lpaste.net/4529242952949563392
<NixOS_GitHub>
[nixpkgs] fpletz pushed 2 new commits to release-16.09: https://git.io/vDPjO
<NixOS_GitHub>
nixpkgs/release-16.09 a93c2f8 Franz Pletz: gitlab service: fix database creation...