gchristensen changed the topic of #nixos-dev to: NixOS Development (#nixos for questions) | https://hydra.nixos.org/jobset/nixos/trunk-combined https://channels.nix.gsc.io/graph.html | 18.03 release managers: fpletz and vcunat
pxc has quit [Ping timeout: 245 seconds]
* obadz spent some time writing some docbook today. Does not feel good :(
<obadz> Anyone wants to proofread/opine on https://github.com/NixOS/nixpkgs/pull/36272/files ?
pxc has joined #nixos-dev
ottidmes has quit [Ping timeout: 248 seconds]
<angerman> obadz: there is some higher level authoring Tool for it, right?
mbrgm has quit [Ping timeout: 252 seconds]
mbrgm has joined #nixos-dev
<Mic92> ekleog: I don't know
orivej has quit [Ping timeout: 255 seconds]
jtojnar has quit [Remote host closed the connection]
jtojnar has joined #nixos-dev
davidlt_ has joined #nixos-dev
jtojnar has quit [Remote host closed the connection]
davidlt_ has quit [Read error: Connection reset by peer]
davidlt has joined #nixos-dev
pie___ has quit [Remote host closed the connection]
pie___ has joined #nixos-dev
jtojnar has joined #nixos-dev
<sphalerite> dtz: nope, sandbox was enabled
orivej has joined #nixos-dev
ottidmes has joined #nixos-dev
vcunat has joined #nixos-dev
FRidh2 has joined #nixos-dev
ma27 has joined #nixos-dev
vcunat has quit [Ping timeout: 248 seconds]
<obadz> angerman: oh? didn't know… which do you recommand?
vcunat has joined #nixos-dev
<angerman> obadz: I was actually curious if there was one... writing xml seems like ... ugh.
<obadz> angerman: yup
<sphalerite> you don't have to write XML, this has been said many times before, there are multiple people willing to mark it up for you
<sphalerite> all we need is for it to actually be written
<obadz> sphalerite: saw that in the manual but kind of assuming that these people's time is better spent reviewing PRs than marking up my doc so thought best to do myself…
<sphalerite> docs not written in docbook are still better to get than no docs :)
<gchristensen> obadz: what editor do you use?
<gchristensen> if you use emacs, you'll probably have a better time with the config here in this readme: https://github.com/NixOS/nix-pills/blob/master/README.md
<obadz> I use vim
<gchristensen> maybe that config will get you on your way to a less bad experience
<gchristensen> the important bit being, obviously, rng-schema-locating-files
<obadz> didn't know about nix pills other than the lethalman blog series… this is cool. going to read those…
<gchristensen> same thing :)
<gchristensen> but ported over sothey can be improved
<obadz> in docbook :')
<gchristensen> this makes me a heretic in a community that loves to hate docbook, but I like docboox :$
<gchristensen> figuring out that emacs config and finding this: http://tdg.docbook.org/tdg/5.2/para.html was what opened my eyees
<obadz> If at least the output looked pretty, it would be worth the effort…
<MichaelRaskin> This doesn't make you just a heretic. Worse. This makes you the go-to docbook debugging person.
<gchristensen> how would you make the output pretty?
<gchristensen> MichaelRaskin: I'm just going to preemptively die now
<MichaelRaskin> But… but… who will coordinate the update to The Protocol?
<obadz> gchristensen: maybe at least syntax highlighting. searchable docs like devdocs.io
<gchristensen> cool
<vcunat> There are tools to convert between markups.
<vcunat> I used pandoc last time IIRC.
<gchristensen> pandoc is very good
<sphalerite> vcunat: but with those the semantics are limited to those expressible in the source language. Which is unhelpful as far as converting to docbook is concerned
<FRidh2> it's been a while again we had a doc language discussion! :)
<vcunat> Yeah, we've had very lengthy discussions :-)
<sphalerite> indeed
<gchristensen> this one has been less useless and less aggressive, FRidh2 :P
<sphalerite> maybe because it's not in #nixos? xD
* gchristensen looks around, counts
<gchristensen> maybe because everyone here contributes in significant ways
<MichaelRaskin> If you want to hide a discussion well, just hold it at #nixos-chat.
<MichaelRaskin> You mean, contributes often enough to have learned not to have opinions on the things one doesn't use?
<sphalerite> oooooft
<obadz> gchristensen: didn't you use to have a bot that would send a message when a channel advanced?
<gchristensen> yep
<obadz> did you put it to sleep?
<gchristensen> nope
<gchristensen> 2018-03-04 01:50:08 -- Notice({^_^}): Channel nixos-17.09 advanced to https://github.com/NixOS/nixpkgs/commit/1dcd022f01b (from 8 hours ago, history: https://channels.nix.gsc.io/nixos-17.09)
<sphalerite> the unstable ones just haven't been advancing :p
<obadz> sphalerite: one column just went green
<obadz> how long does it take these days from that point to the channel advance?
<gchristensen> now the rest of the evaluation has to finish building (pass or fail doesn't matter, just done) and then within a bit of that
<obadz> ah yeah :(
<Mic92> there was one broken network test i think on unstable. But I reverted it already
<sphalerite> was that the deterministic interface names one?
<Mic92> yes. I have to merge two of my past pull requests into one for that to work.
<Mic92> yes. that was about deterministic interfaces.
<Mic92> qt-svg was also broken for random reason. But this has been not reproducible locally.
<Mic92> ah and the firefox build timed out
<Mic92> that is likely the reason
<shlevy> niksnut: For https://github.com/NixOS/nixpkgs/issues/36268 , any objection to adding builtins.storeFeatures? For now it can just be an empty list (since the existence of it would imply a lower bound on the version), and then when we add new features to the daemon that affect evaluation we can add to that list?
<shlevy> Ah, well, it can't actually be an empty list to start :D Because the whole point is it's a function of the daemon
<gchristensen> /!\ I'll probably be taking ofborg down for a couple hours today, to move the core bits to a new home. all the work will be queued and picked up at the end, just delayed for "a while" /!\
<shlevy> blue-green that shit :P
<shlevy> (kidding, of course)
<sphalerite> heh, it's funny because gchristensen's nick is blue and shlevy's is green on my weechat
<gchristensen> hah
<gchristensen> I could do this with zero down time, the biggest challenge is zero down time in the log collection
<gchristensen> but the reality is it probably won't take a couple hours. underpromise / overdeliver :)
<shlevy> I'll take that as a promise ;)
<gchristensen> for you? anything :)
<MichaelRaskin> Technically, a promise that it won't take a couple of hours only binds to do the work in 7199 seconds.
<gchristensen> MichaelRaskin has my back
<gchristensen> I think I'll replace the currently useless homepage with a status report thing too.
ma27 has quit [Ping timeout: 240 seconds]
<Mic92> Was romildo ever ask, if we wants to become a nixos member? He has like 450 merged pull requests, while I have 187 pull requests.
<shlevy> Hmm, actually builtins.supportedFeatures doesn't make sense. There's no inherent connection between evaluation time and runtime :(
<shlevy> s/runtime/build-time
<MichaelRaskin> 450 sounds like romildo _could_ be in one of my mass-mailings before I gave up
<MichaelRaskin> Mic92: do you have the email handy or should I go the long way?
<Mic92> MichaelRaskin: malaquias@gmail.com
<Mic92> what is the long way?
<MichaelRaskin> Running my vanity.sh script, or at least the relevant parts
<MichaelRaskin> Or doing that by hand — look up the name, look up the email from git log
<Mic92> This is what I have done
<MichaelRaskin> Hm, no, my spam runs never included this emails.
<Mic92> He does a lot of desktop stuff and his pull requests rarely needs a review.
<sphalerite> what were these spam runs?
<MichaelRaskin> I ran my vanity.sh script (can be found in maintainers/scripts)
<MichaelRaskin> Then grabbed the NixOS member list
<MichaelRaskin> Then did a diff, removed a few people about whom I just remembered by name, reviewed the recent PR track record to make sure.
<MichaelRaskin> So people who had a lot of commits in master, and the recent PRs mostly accepted, and not yet members were put into a list.
<sphalerite> aah right
<MichaelRaskin> And then I wrote a letter (the same one) to everyone in the list telling that a) the chances for getting commit access are high and b) yes, it is a good idea to ask already, pretty please.
<MichaelRaskin> I can teach the logic and the use of the vanity counter to anyone who is willing to pick up the idea!
<MichaelRaskin> I guess I can just run it myself on request from any such a nice person.
<MichaelRaskin> I just don't want to do the final manual filtering (which removes almost nobody) and email writing step.
<obadz> NixOS is a much better virtualbox guest than Ubuntu
<obadz> nixos-generate-config detects that it's in a vbox and loads all the right module for guest/host interactions
* obadz just saying…
mbrgm has quit [Quit: ZNC 1.6.5 - http://znc.in]
mbrgm has joined #nixos-dev
<ikwildrpepper> fpletz: pong, added comment (TLDR: will release nixops this week)
<fpletz> ikwildrpepper: awesome, thanks
mbrgm has quit [Quit: ZNC 1.6.5 - http://znc.in]
mbrgm has joined #nixos-dev
goibhniu1 has joined #nixos-dev
goibhniu has quit [Ping timeout: 256 seconds]
<vcunat> fpletz: I don't know with the release branching. The worst thing I know right now is https://github.com/NixOS/nixpkgs/issues/36268
ottidmes has quit [Quit: WeeChat 1.9.1]
ma27 has joined #nixos-dev
<obadz> what's the latest re staging? doesn't look like it evals regularly anymore
<obadz> I've been merging a bunch of PRs into staging. Hope they'll still make their ways to master eventually?
<vcunat> It does evaluate regularly on Hydra.
<vcunat> But merging is more difficult, as the inflow of changes is large, and occasional breakages happen.
<vcunat> Maybe we would better make the process iterative in future.
<vcunat> (I.e. explicit merges from staging to a Hydra-built branch, then stabilizing that if needed, merge to master and re-iterate.)
<Dezgeg> +1
<vcunat> I'm mainly looking at nixpkgs/staging
<vcunat> wrt. binaries it's typically darwin and aarch64 that need to be waited for
pie___ has quit [Ping timeout: 256 seconds]
pie___ has joined #nixos-dev
ma27 has quit [Quit: WeeChat 2.0]
ma27 has joined #nixos-dev
ma27 has quit [Ping timeout: 240 seconds]
<obadz> ah I got confused because https://hydra.nixos.org/job/nixpkgs/staging/tested#tabs-constituents also looked stale but looks like that's just not in recent evals
pxc has quit [Ping timeout: 256 seconds]
ma27 has joined #nixos-dev
pie___ has quit [Ping timeout: 276 seconds]
<obadz> was is the proper job set to look at on staging now?
pie___ has joined #nixos-dev
<gchristensen> heads up: ofborg will stop processing new tasks for a little bit while I move its infra.
<jtojnar> could I get a hydra build of https://github.com/NixOS/nixpkgs/pull/35513?
<jtojnar> I need to check if it fixes gnome shell but am unable to build it mysefl
{^_^} has quit [Remote host closed the connection]
{^_^} has joined #nixos-dev
{^_^} has joined #nixos-dev
pxc has joined #nixos-dev
orivej has quit [Ping timeout: 240 seconds]
<jtojnar> actually nevermind splitting the pull request will allow me to build it
jtojnar has quit [Quit: jtojnar]
jtojnar has joined #nixos-dev
FRidh2 has quit [Quit: Konversation terminated!]
<gchristensen> ofborg will process new commands starting now. I'm still working on setting up evaluators and builders on the new infra
<domenkozar> \o/
<domenkozar> gchristensen: beefier machine?
<gchristensen> right now I'm moving the centralized bits -- rabbitmq, comment filtering, log collection
<gchristensen> off of "just some random machine" and on to a dedicated system, so then I can let other people help admin it
lopsided98 has quit [Remote host closed the connection]
lopsided98 has joined #nixos-dev
lopsided98 has quit [Remote host closed the connection]
lopsided98 has joined #nixos-dev
<gchristensen> update: I _think_ all ofborg things should be normal now.
vcunat has quit [Quit: Leaving.]
pxc has quit [Ping timeout: 252 seconds]
MichaelRaskin has quit [Ping timeout: 260 seconds]
<gchristensen> domenkozar: still around?
<domenkozar> ye
<domenkozar> gchristensen: ^^
<domenkozar> sorry got carried away :)
<gchristensen> can you do some maintenance on the github side for me? adding a new webhook (not deleting the old one right now)
<domenkozar> sir yes sir
<gchristensen> ok great :) see the PM
<gchristensen> it took exactly 2 hours to fully move it over shlevy, sorry
<gchristensen> jtojnar: looks like https://github.com/NixOS/nixpkgs/pull/36312 needs to be rebased?
<jtojnar> gchristensen: unfortunately it is non-trivial as I modified it on master
<gchristensen> ouch
<jtojnar> though, I shough be able to drop most of the commits at least