| 27 Oct 2021 |
 infinisil | Oh wait I think I got it | 19:37:45 |
| infinisil | https://github.com/NixOS/hydra/blob/master/src/hydra-eval-jobs/hydra-eval-jobs.cc#L400 | 19:38:57 |
| infinisil | So like, what if I just builtins.trace "error" before anything else lol | 19:39:32 |
| infinisil left the room. | 20:21:21 |
| infinisil joined the room. | 20:21:33 |
| infinisil | Apparently I can't have this room be in the non-spaces section, oh well.. | 20:21:57 |
 das_j | In reply to @infinisil:matrix.org
Apparently I can't have this room be in the non-spaces section, oh well.. You can toggle that globally | 20:22:17 |
| das_j | In reply to @infinisil:matrix.org
So like, what if I just builtins.trace "error" before anything else lol Or just add the trace prefix to the if? | 20:22:27 |
| infinisil | das_j: Then I don't think the filter has any use, I'll investigate the history of this when I have time | 20:23:46 |
|  fabianhjr joined the room. | 21:10:54 |
| 28 Oct 2021 |
|  NixOS Moderation Bot banned  David Arnold (blaggacao) (<no reason supplied>). | 20:04:59 |
| 29 Oct 2021 |
 ma27 | any opinions about https://github.com/NixOS/hydra/pull/922? Yes, it isn't the prettiest solution, but I'm still quite interested in finding a solution to use plugins with flake jobsets. Also, this is my only pending Hydra PR that can probably be reviewed without sacrificing at least a day %) | 18:20:14 |
| 30 Oct 2021 |
 Mic92 | I documented hydra webhooks for github: https://github.com/NixOS/hydra/pull/1052 | 07:48:15 |
|  jeonwh joined the room. | 11:07:19 |
| 31 Oct 2021 |
|  toonn changed their profile picture. | 20:19:14 |
| 3 Nov 2021 |
|  jbedo changed their profile picture. | 10:25:29 |
| 4 Nov 2021 |
| Mic92 | Why is the prometheus endpoint per job and not per jobset? | 18:36:55 |
| Mic92 | Nevermind. I can just use telegraf's sql input for that. | 18:45:20 |
 @grahamc:nixos.org | what would you like it to say per jobset? | 18:56:55 |
| @grahamc:nixos.org | one reason I didn't make it per jobset is indeed I wanted to monitor a specific job | 18:57:13 |
| @grahamc:nixos.org | so, probably the reason is nobody built it yet :P | 18:57:23 |
| Mic92 | grahamc: getting an alert if one of my build fails: https://hydra.thalheim.io/eval/160 | 18:58:45 |
| @grahamc:nixos.org | gotcha, so just monitor the entire jobset instead of an aggregate job | 18:59:13 |
| @grahamc:nixos.org | makes good sense to me | 18:59:16 |
| Mic92 | @grahamc: only issue is if you use this endpoint on nixpkgs with 100.000 jobs per jobset. | 19:00:04 |
| @grahamc:nixos.org | yeah, that is also why I haven't tried to implement it :) | 19:00:30 |
| @grahamc:nixos.org | there is a gulf between "Most users" and "hydra.nixos.org" on appropriate features | 19:00:55 |
| das_j | sounds like return if Sys::Hostname == "NAME_OF_THE_NIXOS_HOST" | 19:01:42 |
| @grahamc:nixos.org | puke | 19:01:52 |
| das_j | * sounds like return if Sys::Hostname::hostname == "NAME_OF_THE_NIXOS_HOST" | 19:01:53 |
