samueldr changed the topic of #nixos-dev to: #nixos-dev NixOS Development (#nixos for questions) | NixOS 19.09 is released! | | | 19.09 RMs: disasm, sphalerite |
MichaelRaskin has quit [Ping timeout: 245 seconds]
lopsided98 has quit [Ping timeout: 240 seconds]
<red[m]> QuickQ: I'm opening a PR against a module and afaict modules don't have maintainers? Who should I tag (other than the person who did the previous commit?)
lopsided98 has joined #nixos-dev
<red[m]> sorry - wrong channel - should be on #nixos - brb
tdeo has quit [Ping timeout: 268 seconds]
tdeo has joined #nixos-dev
tdeo has joined #nixos-dev
tdeo has quit [Changing host]
ris has quit [Ping timeout: 276 seconds]
<aanderse> red: btw modules can (and should) have maintainers... but hardly any of them do, unfortunately
tdeo has quit [Quit: Quit]
tdeo has joined #nixos-dev
tdeo has joined #nixos-dev
tdeo has quit [Changing host]
<worldofpeace> infinisil: If the commits are literally fixups they don't need to be in the history
<worldofpeace> infinisil: they should squash the commits into the one that added the service. only if the further commits add extra context should it be needed.
<worldofpeace> and yep, github has the co-authors feature. I pretty much try to use it always.
drakonis has quit [Quit: WeeChat 2.6]
drakonis has joined #nixos-dev
evanjs has quit [Quit: ZNC 1.7.4 -]
evanjs has joined #nixos-dev
<infinisil> worldofpeace: Yeah, we should probably add something like this to the manual
MichaelRaskin has joined #nixos-dev
genesis has joined #nixos-dev
asymmetric has quit [Ping timeout: 240 seconds]
ris has joined #nixos-dev
__monty__ has joined #nixos-dev
asymmetric has joined #nixos-dev
asymmetric has quit [Ping timeout: 240 seconds]
asymmetric has joined #nixos-dev
FRidh has joined #nixos-dev
<tilpner> ,matrixbridge nh2[m]
<{^_^}> nh2[m]: The IRC<->Matrix bridge sends multi-line messages as links: `* Someone sent a long message: < >`. This means that highlights don't work, people have to open your link, and it gets less visibility than multiple single-line messages
<tilpner> And #nixos-borg
<simpson> Huh. Wonder why that went multi-line; it seems like it could have been sent as one single IRC message. Matrix, you silly~
<tilpner> Probably because it was entered as multiple lines
<nh2[m]> tilpner: uh, thanks. Do you know if I can turn that off? I can in irccloud but would like to turn it in matrix as well.
<tilpner> nh2[m]: I don't. I still use both
<nh2[m]> yes, I did enter multiple lines for better readability
<nh2[m]> oh well the bridge also seems to have > 10 seconds delay currently :( So hard to send plain text messages
<nh2> tilpner: thanks, so it looks like I need to run my own one to have it work well
<tilpner> nh2: I linked that, because the main instance might have the same config, so 3 lines might be fine
<nh2> tilpner: right I sent 4
<tilpner> nh2: I don't recommend Matrix for IRC. With weechat you can use IRC and Matrix in the same client
<tilpner> I don't know how you would host your own bridge without having your own HS
<red[m]> So - were I a betting man I would say that they do the collapsing of text into a link because of flood control on irc servers
<red[m]> the irc protocol has a maximum number of lines per X as a DDOS control
<red[m]> and will disconnect you if you go over
<red[m]> In the background I've been writing an API binding for erlang for matrix
<red[m]> and the irc protocol is ridiculously simple too
<red[m]> I just have an aversion to runningjavascript on a server
<red[m]> bbiab
drakonis_ has joined #nixos-dev
drakonis has quit [Ping timeout: 276 seconds]
drakonis_ has quit [Read error: Connection reset by peer]
drakonis_ has joined #nixos-dev
adisbladis has quit [Quit: WeeChat 2.4]
drakonis has joined #nixos-dev
adisbladis has joined #nixos-dev
<worldofpeace> infinisil: I think we could expand to be more detailed on the commit message thing.
<worldofpeace> I think I started to pretty much instruct everyone to do `maintainers: add $your_name` in a separate commit, so it for sure should be there. These things going by word of mouth just isn't good for the most general step to contributions.
<worldofpeace> lol, our manual has a ref to travisCI
<MichaelRaskin> Historical artifact!
<MichaelRaskin> We need to fix it before it gets UNESCO protection
<worldofpeace> Hah, my issue would be I couldn't just fix one thing. It'd get a rewrite.
Cale has quit [Ping timeout: 250 seconds]
drakonis1 has joined #nixos-dev
drakonis_ has quit [Ping timeout: 245 seconds]
Cale has joined #nixos-dev
FRidh has quit [Quit: Konversation terminated!]
<clever> Nov 09 17:03:17 nas hydra-init[2557]: Base class package "DBIx::Class::Schema" is empty.
<clever> Nov 09 17:03:17 nas hydra-init[2557]: (Perhaps you need to 'use' the module which defines that package first,
<clever> after upgrading my nixpkgs, hydra no longer starts
<das_j> clever: Do you have CatalystAuthenticationStoreDBIxClass in your build inputs?
<clever> das_j: just services.hydra.enable=true;
<clever> das_j: it should be using whatever nixpkgs defined for hydra
<das_j> clever: We have overridden our hydra and modified the build inputs. I just don't know if any of them is what's missing for you
<clever> das_j: i also ran into superUser problems
<clever> 276 echo "create extension if not exists pg_trgm" | runuser -u ${} -- ${}/bin/psql hydra
<clever> # The postgres user is used to create the pg_trgm extension for the hydra database
<clever> hydra-users postgres postgres
<das_j> That's normal iirc
<clever> das_j: it goes out of its way to use the right user in hydra-init, then doesnt, and grants a non-existent user access
<clever> so root lacks access to hydra, and hydra-init fails hard
<clever> das_j: hydra from master fails the same way
<clever> das_j: CatalystAuthenticationStoreDBIxClass is already in the perlDeps
<das_j> oof
<ma27[m]> it's a known problem:
<{^_^}> #72783 (by lopsided98, 5 days ago, open): Hydra fails at runtime due to buildPerlPackage changes
<clever> ma27[m]: would changing the nixpkgs used to build hydra fix it?
<ma27[m]> the problem is some changes in the build scripts for perl changed quite recently and since then no transitive deps are resolved (previously `buildEnv` resolved automagically perl deps)
<ma27[m]> I actually tried to add transitive dependencies last week, but that escalated pretty quickly
<ma27[m]> there are some workarounds suggested in the issue I linked, I'm currently waiting for some feedback from volth, until then I'd recommend to use hydra from release-19.09
<kgz> what's the etiquette these days for getting package updates for my own things merged?
<{^_^}> #72036 (by kragniz, 2 weeks ago, open): sailsd: 0.2.0 -> 0.3.0
<clever> Nov 09 17:26:50 nas hydra-init[18246]: exception when others then
<clever> Nov 09 17:26:50 nas hydra-init[18246]: raise warning 'Can not create extension pg_trgm: %', SQLERRM;
<clever> Nov 09 17:26:50 nas hydra-init[18246]: NOTICE: extension "pg_trgm" already exists, skipping
<clever> ma27[m]: now its failing to create pg_trgm, because pg_trgm exists, lol
<ma27[m]> which hydra package and module do you use? (the expressions from nixos/hydra, release-19.09 or master?)
<clever> ma27[m]: hydra master, with 2 changes to it
<das_j> ma27[m]: Also think of the NetPrometheus dependency. Idk if it was added recently, or if it's just on the flakes branch, but it seems to be new
<clever> ma27[m]: fetchGit hangs and shows no progress for over 5mins
<clever> <nixpkgs> used the broken perl from the host
<clever> fetchTarball used the old nixpkgs that worked
<ma27[m]> das_j: the `NetPrometheus` dependency was added in a revision which isn't used in nixpkgs (yet). I actually prepared that bump last weekend but didn't push it yet due to the issue we have on master atm
<das_j> ma27[m]: Nice, thanks for the work!
<das_j> Also, is the bridge properly translating [ ] when I ping you?
drakonis_ has joined #nixos-dev
<clever> Nov 09 17:26:50 nas hydra-init[18246]: raise warning 'The pg_trgm index on builds.drvpath has been skipped (slower complex queries on builds.drvpath)';
<clever> hydra=# create index IndexTrgmBuildsOnDrvpath on builds using gin (drvpath gin_trgm_ops);
<clever> ERROR: relation "indextrgmbuildsondrvpath" already exists
<clever> errors claim the index was skipped
<clever> but manually creating it claims it already existed
<ma27[m]> das_j: I got notified by your previous message at least %)
drakonis2 has joined #nixos-dev
drakonis2 has quit [Read error: Connection reset by peer]
drakonis2 has joined #nixos-dev
<clever> failed to write output file '/build/jam129317b.000'!
<clever> /nix/store/k6kjp4ljs9l7kn8fwizz2zf46ank83ih-stdenv-linux/setup: line 111: /dev/null: No such file or directory
<clever> and i think this was due to GC, but no clue how even /dev/null is missing, enless it cleaned up the chroot dir?
drakonis1 has quit [Ping timeout: 246 seconds]
drakonis_ has quit [Ping timeout: 245 seconds]
<jtojnar> wow, the python test driver has much better CLI
<clever> in an effort to avoid the perl test-driver in nixos, i once wrote a c++ testing framework, with lua scripting, and protobuf<->lua-table encoding, lol
drakonis_ has joined #nixos-dev
drakonis2 has quit [Ping timeout: 252 seconds]
<worldofpeace> Jan Tojnar: Totally, I was surprised also. autocomplete is pretty nice.
pie_ has quit [Ping timeout: 265 seconds]
__monty__ has quit [Quit: leaving]
pie_ has joined #nixos-dev