worldofpeace_ changed the topic of #nixos-dev to: #nixos-dev NixOS Development (#nixos for questions) | NixOS stable: 20.03 ✨ https://discourse.nixos.org/t/nixos-20-03-release/6785 | https://hydra.nixos.org/jobset/nixos/trunk-combined https://channels.nix.gsc.io/graph.html | https://r13y.com | 19.09 RMs: disasm, sphalerite; 20.03: worldofpeace, disasm | https://logs.nix.samueldr.com/nixos-dev
joepie91 has quit [Quit: killed]
rajivr has joined #nixos-dev
Emantor has quit [Quit: ZNC - http://znc.in]
Emantor has joined #nixos-dev
<Graypup_> would anyone get mad at me if I put a dependency on readline in libnixexpr? asking for a friend
<Graypup_> I don't like doing that personally, and I will try to refactor that part out
<Graypup_> but if I fail? idk
<Graypup_> what I'm doing is extracting the repl to libnixexpr so that it can have custom commands added
<Graypup_> this refactor is getting ugly :(
<Graypup_> specifically I want to kill the global variable `curRepl` because it seems bad, but am hitting trouble with passing that state properly since capturing lambdas don't coerce to function pointers (for good reason, to be fair)
<Graypup_> hmmmmm
betawaffle has quit [Ping timeout: 240 seconds]
betawaffle has joined #nixos-dev
betawaffle has quit [Max SendQ exceeded]
claudiii has quit [Ping timeout: 260 seconds]
betawaffle has joined #nixos-dev
angerman has quit [Ping timeout: 244 seconds]
mdlayher has quit [Ping timeout: 244 seconds]
claudiii has joined #nixos-dev
mdlayher has joined #nixos-dev
emilazy has quit [Ping timeout: 244 seconds]
georgyo has quit [Ping timeout: 272 seconds]
emilazy has joined #nixos-dev
betawaffle has quit [Ping timeout: 260 seconds]
angerman has joined #nixos-dev
mdlayher has quit [Read error: Connection reset by peer]
claudiii has quit [Ping timeout: 265 seconds]
emilazy has quit [Ping timeout: 260 seconds]
mdlayher has joined #nixos-dev
angerman has quit [Ping timeout: 256 seconds]
claudiii has joined #nixos-dev
mdlayher has quit [Ping timeout: 240 seconds]
mdlayher has joined #nixos-dev
georgyo_ has joined #nixos-dev
betawaffle has joined #nixos-dev
emilazy has joined #nixos-dev
angerman has joined #nixos-dev
<Graypup_> uhh, does anyone know how I can run a specific test with make?
<Graypup_> I'm reading tests.mk and have no idea what the targets it's creating are
<Graypup_> `make -pRrq : | grep --color plugins.sh`
<Graypup_> => tests/plugins.sh.test
<Graypup_> apparently
AlwaysLivid has joined #nixos-dev
AlwaysLivid has quit [Ping timeout: 256 seconds]
orivej has joined #nixos-dev
AlwaysLivid has joined #nixos-dev
alp has joined #nixos-dev
FRidh has joined #nixos-dev
orivej has quit [Ping timeout: 265 seconds]
cole-h has joined #nixos-dev
justanotheruser has quit [Ping timeout: 260 seconds]
alp has quit [Ping timeout: 272 seconds]
marek_ is now known as marek
cole-h has quit [Quit: Goodbye]
orivej has joined #nixos-dev
__monty__ has joined #nixos-dev
AlwaysLivid has quit [Remote host closed the connection]
AlwaysLivid has joined #nixos-dev
LnL has quit [Ping timeout: 256 seconds]
nschoe has joined #nixos-dev
alp has joined #nixos-dev
nschoe has quit [Ping timeout: 272 seconds]
orivej has quit [Ping timeout: 240 seconds]
AlwaysLivid has quit [Remote host closed the connection]
AlwaysLivid has joined #nixos-dev
alp has quit [Ping timeout: 272 seconds]
enick_617 has joined #nixos-dev
enick_617 has quit [Changing host]
enick_617 has joined #nixos-dev
enick_617 has joined #nixos-dev
enick_617 is now known as joepie91
<philipp[m]> I really don't like this behaviour of security.acme: https://github.com/NixOS/nixpkgs/blob/release-20.03/nixos/modules/security/acme.nix#L332
<philipp[m]> It chowns and chmods the key, when it is not changed and postRun isn't called after.
<philipp[m]> That breaks the following very handy setup: `postRun = "chown -R ${(toString config.ids.uids.nginx)} .";`
alp has joined #nixos-dev
<philipp[m]> What would you say: Just move the chmod and chown into the if block above it? I think it's a good idea, I wouldn't want my files to change their uid/gid/umask out of the blue anyway.
<dtz> What if the group/user change but the keys don't? (And your postRun isn't used)
<dtz> Related thought is that as-written it makes it trivial to conclude the chosen/chmod are always executed before script ends, and therefore the keys always have the specified permissions. Maybe the keys' owner should be tied to the nginx uid/etc at the nix level? (Makes some sense maybe, at least if it's clear nginx needs the keys :))
<dtz> I don't have a preference really just trying to help :)
<philipp[m]> I decided that it's probably best and follows the most internal logic for changes by chmod/chown to trigger KEY_CHANGED=yes
<dtz> Ahh sure
<{^_^}> #95498 (by pstn, 20 seconds ago, open): nixos/security.acme: Only chmod/chown certificate files when changed.
alp has quit [Ping timeout: 272 seconds]
alp has joined #nixos-dev
orivej has joined #nixos-dev
MichaelRaskin has joined #nixos-dev
AlwaysLivid has quit [Remote host closed the connection]
nschoe has joined #nixos-dev
AlwaysLivid has joined #nixos-dev
alp has quit [Ping timeout: 272 seconds]
AlwaysLivid has quit [Ping timeout: 272 seconds]
AlwaysLivid has joined #nixos-dev
alp has joined #nixos-dev
justanotheruser has joined #nixos-dev
alp has quit [Remote host closed the connection]
alp has joined #nixos-dev
alp has quit [Ping timeout: 240 seconds]
orivej has quit [Ping timeout: 265 seconds]
Raito_Bezarius has quit [Remote host closed the connection]
Raito_Bezarius has joined #nixos-dev
Raito_Bezarius has quit [Changing host]
Raito_Bezarius has joined #nixos-dev
alp has joined #nixos-dev
AlwaysLivid has quit [Ping timeout: 240 seconds]
AlwaysLivid has joined #nixos-dev
<FRidh> I'm creating binary wrappers now with Nim. They're about 85 kB each. Next step, a make-wrapper compatible interface for creating them.
vcunat has joined #nixos-dev
<clever> FRidh: nice, that would also greatly simplify setuid root
<vcunat> trunk-combined won't evaluate on Hydra :-/ (repeatedly) So far I've been unable to even reproduce it locally - I expect that json exception line is the fatal one. https://hydra.nixos.org/jobset/nixos/trunk-combined#tabs-errors
<FRidh> bug in the nix version used?
<FRidh> i recall there was something about replacing json parser in nix
<FRidh> vcunat: how did you evaluate it locally?
<vcunat> I'm not running Hydra's evaluator, so maybe I just haven't found the problematic attribute.
<vcunat> Packages seem fine.
<vcunat> The one test touched since last success is also fine.
<FRidh> clever: and less complaints from darwin users :P
<vcunat> I suspect #82743
<{^_^}> https://github.com/NixOS/nixpkgs/pull/82743 (by Infinisil, 21 weeks ago, merged): Freeform modules
<vcunat> but I don't really know. We just need traces.
<vcunat> (in Hydra's evaluator)
<infinisil> Hmm
<infinisil> vcunat: How can you eval this locally?
<infinisil> nix-instantiate nixos/release-combined.nix -A tested ?
<vcunat> I haven't managed to reproduce the error so far.
<infinisil> Wait no the command of mine finishes in a second
<infinisil> vcunat: I feel like the error comes from hydra using --strict --json and then importing the result into some hydra structure for processing
AlwaysLivid has quit [Ping timeout: 272 seconds]
<infinisil> But there's an unexpected null somewhere. So maybe we can evaluate it with --json too and grep for nulls
<infinisil> But I'm not sure what is even evaluated there
AlwaysLivid has joined #nixos-dev
<vcunat> The more I look the less it seems relevant to those "Freeform modules".
<samueldr> Nix expression nixos/release-combined.nix
<vcunat> Well, yes... but I know of no nice way to evaluate exactly the set of attributes that Hydra does.
<infinisil> samueldr: Yeah I got that, but just doing `nix-instantiate nixos/release-combined.nix` can't be it
<samueldr> maybe I should clean-up that script for evaluating it just like hydra
<samueldr> but IIRC it requires a bunch of RAM
<infinisil> Ah maintainers/scripts/hydra_eval_check?
<vcunat> I have 32G at hand, if that's enough.
<infinisil> I also have 16GB RAM and 32GB swap
<samueldr> hmm, hydra_eval_check seems... odd
<samueldr> you might want to comment out the strace
<samueldr> or not
<samueldr> if hydra is eating the actual messages they'll be present in the strace dump
<samueldr> but strace makes this take *much* longer
<infinisil> Ah but at least hydra_eval_jobs seems to be the right path
<samueldr> ugh, re-reading this and I'm sad I don't have the ~20 minutes to rewrite that
<samueldr> yeah, it has to use the hydra's own evaluator
<samueldr> which differs a bit
<infinisil> Ah probably catches errors and stuff
<infinisil> > hydra-check.meta.homepage
<samueldr> I think it also is kind of stricter?
<infinisil> Hm that's not relevant
<samueldr> nah, hydra-check consumed hydra contents :)
<samueldr> I think we want to fix maintainers/scripts/hydra_eval_check so it works again
<samueldr> I started the eval locally, too
<samueldr> bbl
<infinisil> I'm trying `hydra-eval-jobs nixos/release-combined.nix -I $PWD`
<infinisil> hydra-eval-jobs from pkgs.hydra-unstable
<vcunat> I think -combined is overkill
<vcunat> I was already running the same with just release.nix
<vcunat> (we know that with packages this doesn't happen, so it has to be among tests)
<infinisil> Hm I see
<julm> hm, it looks like builtins.path cannot be used when using a flake.nix, I get errors like: string '/nix/store/[…]' cannot refer to other paths
alp has quit [Ping timeout: 240 seconds]
<samueldr> error: [json.exception.type_error.302] type must be string, but is null
<samueldr> took 2 minutes to get there
<samueldr> but no more context
* samueldr puts strace back in
<infinisil> This error is awful
<samueldr> without context, definitely
<infinisil> The fact that it doesn't have any context whatsoever by default is awful already
<samueldr> agreed
<infinisil> Oh wait, hydra is written in perl??
<infinisil> :o
<samueldr> depends which part
<samueldr> don't :o unless you're willing to put in the effort of making the changes, and making the community accept the change :)
<clever> infinisil: a mix of perl and c++
<samueldr> Hydra comes from another time
<infinisil> samueldr: Oh that's nothing inherently negative, :o is just as in it surprised me
<samueldr> :) often perl brings negativity in people
<clever> infinisil: hydra-evaluator (the daemon) is c++, it will run hydra-eval-jobset at regular intervals
<clever> infinisil: hydra-eval-jobset is perl, it will fetch all inputs, and run hydra-eval-jobs
<clever> infinisil: hydra-eval-jobs is back in c++, it will eval the release.nix and spit out json
<clever> hydra-eval-jobset then parses that json, and creates entries in postgres
<clever> hydra-queue-runner is c++, and then begins acting on those entries, spinning up jobs as needed
<clever> and hydra-server is a perl based http frontend over the db
<clever> thats most of the large pieces of hydra
nschoe has quit [Ping timeout: 272 seconds]
<infinisil> I see
<clever> why do i have all of that memorized? lol
<vcunat> I expect with this we can bisect the problem within several hours.
<jtojnar> fontconfig-penultimate probably does not eval
<jtojnar> refers to non-existing package
<infinisil> Imagine if everything in nixpkgs always evald (minus throw/asserts when broken and co.)
Guest76136 has joined #nixos-dev
Guest76136 is now known as LnL
FRidh has quit [Quit: Konversation terminated!]
orivej has joined #nixos-dev
<samueldr> may be coming from the manual
<samueldr> quickly I can't see anything else in the strace
<samueldr> 413M; took only 6 minutes to make though
cole-h has joined #nixos-dev
<vcunat> You manage this with stable Nix?
<vcunat> Manual seems to evaluate fine for me.
<vcunat> (and the first pass of hydra-eval-jobs still haven't finished - perhaps I'm doing something wrong)
<samueldr> hmm
<samueldr> I don't actually know which hydra and which nix it used
<samueldr> (and I'm going out for the rest of the day)
alp has joined #nixos-dev
<vcunat> Ah, now it finished - and successfully :-(
obadz has joined #nixos-dev
vcunat has quit [Ping timeout: 240 seconds]
bhipple has joined #nixos-dev
ixxie has joined #nixos-dev
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
bhipple has quit [Remote host closed the connection]
cole-h has quit [Quit: Goodbye]
alp has quit [Ping timeout: 272 seconds]
rajivr has quit [Quit: Connection closed for inactivity]
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
<Raito_Bezarius> Hi there, I'm trying to integrate https://www.recompile.se/mandos in NixOS, which requires surgery on the initramfs, as I understand it, I have to poke stuff around this file: https://github.com/NixOS/nixpkgs/blob/master/nixos/modules/system/boot/stage-1.nix — Mandos seems to be compatible with Debian initramfs and Dracut, is there a way to leverage those to maximize reuse or should I throw everything
<Raito_Bezarius> away and write a NixOS native thing? Thanks!
<infinisil> Oh neat, I was interested in trying mandos
<Raito_Bezarius> I have the server & "client" in Nix
<Raito_Bezarius> Now, all that's left, is the initramfs stuff and a NixOS config module :-)
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
alp has joined #nixos-dev
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
vcunat has joined #nixos-dev
bennofs has joined #nixos-dev
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
alp has quit [Ping timeout: 272 seconds]
dongcarl has quit [Read error: Connection reset by peer]
orivej has quit [Ping timeout: 256 seconds]
dongcarl has joined #nixos-dev
vcunat has quit [Ping timeout: 240 seconds]
AlwaysLivid has quit [Remote host closed the connection]
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
AlwaysLivid has joined #nixos-dev
vcunat has joined #nixos-dev
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
vcunat has quit [Quit: Leaving.]
dongcarl has quit [Read error: Connection reset by peer]
dongcarl has joined #nixos-dev
orivej has joined #nixos-dev
orivej has quit [Ping timeout: 264 seconds]
alp has joined #nixos-dev
orivej has joined #nixos-dev
orivej has quit [Ping timeout: 265 seconds]
ixxie has quit [Quit: Lost terminal]
orivej has joined #nixos-dev
bennofs has quit [Ping timeout: 256 seconds]
__monty__ has quit [Quit: leaving]
alp has quit [Ping timeout: 272 seconds]
bennofs has joined #nixos-dev
alp has joined #nixos-dev
alp has quit [Ping timeout: 272 seconds]
bennofs has quit [Read error: Connection reset by peer]
bridge[evilred] has quit [Remote host closed the connection]
bridge[evilred] has joined #nixos-dev
red[evilred] has joined #nixos-dev