| 10 Jan 2022 |
 Winter (she/her) | So an issue that I ran into was that https://github.com/NixOS/nixpkgs/pull/153038#discussion_r780785733 (go to the last resolved discussion and expand it) didn’t cause OfBorg to detect myself as a maintainer. But I’m… not really sure why. I can successfully go into a REPL and see that thelounge.meta.maintainers exists.
Does anyone have any pointers on why this would be happening, or where to look in the codebase, or something? | 00:29:16 |
| Winter (she/her) | * So an issue that I ran into was that https://github.com/NixOS/nixpkgs/pull/153038#discussion_r780785733 (go to the last resolved discussion and expand it) didn’t cause OfBorg to detect myself as a maintainer (and add the by package maintainer label). But I’m… not really sure why. I can successfully go into a REPL and see that thelounge.meta.maintainers exists.
Does anyone have any pointers on why this would be happening, or where to look in the codebase, or something? | 00:30:09 |
| Winter (she/her) | Nothing immediately stands out to me :/ | 00:32:51 |
 sterni | Winter (she/her): ofborg relies on the attribute names in the commit message mostly, so the PR wouldn't trigger that code because there's no thelounge: … commit, I would assume? | 11:10:29 |
| 12 Jan 2022 |
 @grahamc:nixos.org | I think “thelounge:” is only for triggering builds, not @ing maintainers. This is to encourage good commit message format, but still @ing relevant people even if the author didn’t do it quite right | 14:33:27 |
| Winter (she/her) | In reply to @grahamc:nixos.org
I think “thelounge:” is only for triggering builds, not @ing maintainers. This is to encourage good commit message format, but still @ing relevant people even if the author didn’t do it quite right I’m confused. You said it’s not for @ing but it would @ anyways? | 14:38:37 |
| Winter (she/her) | (I’ll reword my commits to use the actual attribute name, though; thanks for the clarification.) | 14:39:46 |
| Winter (she/her) | * In reply to @grahamc:nixos.org
I think “thelounge:” is only for triggering builds, not @ing maintainers. This is to encourage good commit message format, but still @ing relevant people even if the author didn’t do it quite right I’m confused. You said it’s not for @ing but that it would @ anyways? | 14:44:14 |
| @grahamc:nixos.org | There is a rule that determines what things to build, and that rule is based on attribute as named in the commit message with the prefix. It has nothing to do with @ing people, iirc. | 14:45:44 |
| Winter (she/her) | Ah, understood. | 14:46:50 |
| sterni | yeah seems like that should use the changed attribute paths | 14:58:34 |
|  ncfavier joined the room. | 16:08:55 |
| 13 Jan 2022 |
|  NixOS Moderation Bot unbanned  David Arnold (blaggacao). | 18:43:45 |
| 14 Jan 2022 |
|  Chinchilla Washington changed their display name from Rev. CornWallace III (novus ordo seclorum) to coilWinder. | 04:39:47 |
| Chinchilla Washington changed their display name from coilWinder to CoilWinder (novus ordo seclorum). | 04:42:11 |
| 17 Jan 2022 |
 pennae | does ofborg not parse commands from review comments? https://github.com/NixOS/nixpkgs/pull/153346#pullrequestreview-854622718 apparently hasn't started a test run | 16:44:25 |
 7c6f434c | I would expect that review comments is a completely different kind of thing that Github peruidically breaks in a different way than comments (so probably ofBorg does not receive these events) | 16:46:14 |
| pennae | annoying, but not entirely surprising | 16:47:51 |
| Winter (she/her) | Do NixOS tests always have to be triggered manually? | 17:39:16 |
| Winter (she/her) | I’ve never seen them run automatically for changed modules/tests. | 17:39:35 |
 hexa | they run automatically if a package updated includes passthru.tests = [ inherit (nixosTest) foobar ]; | 18:00:47 |
| hexa | * they run automatically if a package updated includes passthru.tests = { inherit (nixosTest) foobar }; | 18:00:56 |
| 7c6f434c | Package specified in the commit message, I think | 18:04:09 |
| Winter (she/her) | Ah, got it. So changes to the module itself won’t trigger it automatically. | 18:07:28 |
| hexa | these parts need more love | 18:45:23 |
| 18 Jan 2022 |
| pennae | wanted to check ofborg monitoring to see why a pr is waiting for eval, noticed that https://monitoring.nix.ci/dashboard/db/ofborg seems to not exist | 15:16:42 |
| Winter (she/her) | In reply to @hexa:lossy.network
they run automatically if a package updated includes passthru.tests = { inherit (nixosTest) foobar }; this doesn’t seem to be true either :/
unless by updated you mean version bump and not just rebuild? | 15:59:31 |
| hexa | i meant that if the package definition changes in any way, and it has a nixosTest in passthru, then ofBorg should build it | 16:00:23 |
| Winter (she/her) | yeah i can’t get that to happen sadly :/ | 16:04:53 |
 cole-h | In reply to @pennae:matrix.eno.space
wanted to check ofborg monitoring to see why a pr is waiting for eval, noticed that https://monitoring.nix.ci/dashboard/db/ofborg seems to not exist Grafana broke last year some time when a power outage killed the machine it was running on. I haven't had time to try to piece the dashboard back together from scratch, so you'll want to look at Prometheus (https://nix.ci/prometheus) | 16:16:43 |
