| 16 Jan 2026 |
 K900 | The second one is nothing | 22:39:41 |
| K900 | The first one I may have misread | 22:39:47 |
| K900 | It's almost 2AM | 22:39:51 |
 emily | yeah heap overflow in a case that is maybe compiler UB regardless and I'm any case involves giving attackers crazy levels of control of memory allocation, plus uncommon calls leaking small amounts of stack to DNS server = I sleep | 22:40:41 |
| emily | I'd expect -next contains juicier fixes already | 22:41:40 |
| 17 Jan 2026 |
 Sergei Zimmerman (xokdvium) | There's a slight include messup with cppnix 2.33 and glibc 2.42. I should send that to staging-next now? https://github.com/NixOS/nix/pull/15011 | 18:45:42 |
| Sergei Zimmerman (xokdvium) | Or just master and the regular merge will do the thing? | 18:46:54 |
| K900 | master is fine | 18:48:17 |
| emily | staging-nixos, no? | 18:53:29 |
| emily | given the test rebuilds? | 18:53:37 |
| emily | or is it not default yet? | 18:53:43 |
| Sergei Zimmerman (xokdvium) | Not the default | 18:54:59 |
| Sergei Zimmerman (xokdvium) | Going to also grab aarch64-darwin patches to fix darwin | 19:00:11 |
| Sergei Zimmerman (xokdvium) | * Going to also grab aarch64-darwin patches to fix darwin sandbox shenanigans | 19:00:25 |
| 18 Jan 2026 |
 Randy Eckenrode | python3Packages.setproctitle is failing to build on 25.11. It happened to build on unstable, but that may have been a happy accident, so I’m going to fix the failure on staging first. When I do the backport, should it still target staging-25.11, or can it be retargeted to release-25.11 since it’s not technically causing rebuilds (but it will cause a bunch of builds)? | 19:53:52 |
 leona | Redacted or Malformed Event | 19:54:15 |
| leona | Redacted or Malformed Event | 19:54:22 |
 hexa | depends on the number of rebuilds | 20:12:45 |
| Randy Eckenrode | It was tagged 2501–5000, which is why I’m targeting staging for unstable. Since it’s broken on 25.11, would that still go through staging-25.11 due to the number of builds? | 20:14:10 |
 Vladimír Čunát | Can you make it rebuild on darwin only? (for now) | 20:23:43 |
| Vladimír Čunát | It's broken exactly on staging-next and release-25.11 currently. | 20:24:29 |
| Vladimír Čunát | If it's a darwin-only rebuild for setproctitle, I'd target these two. | 20:24:58 |
| Vladimír Čunát | * If it's a darwin-only rebuild for setproctitle, I'd target these two, probably. | 20:25:11 |
| Vladimír Čunát | I notified upstream, BTW:
https://github.com/dvarrazzo/py-setproctitle/issues/111#issuecomment-3765074486 | 20:27:31 |
| Randy Eckenrode | It rebuilds only on Darwin because the tests are only disabled on Darwin. I can retarget staging-next if it’s already broken there. | 20:28:18 |
| Randy Eckenrode | https://github.com/NixOS/nixpkgs/pull/481408 fixes it on staging-next. | 20:30:17 |
| Vladimír Čunát | Why not tag it as backport-release-25.11 directly? | 20:34:53 |
|  isabel changed their profile picture. | 20:43:58 |
 Arian | Heya I'm super confused
https://github.com/NixOS/nixpkgs/pull/475079 was opened on 29.12.2025 and merged on 10.01.2026
https://github.com/NixOS/nixpkgs/pull/474788 was merged on 28.12.2025
the 258.3 bump still didn't land on 25.11
Did I miss the window by a few seconds? | 22:05:22 |
| Arian | At this rate , given that the next staging-next-25.11 iteration will take at least a week; it means it'll take ~5 weeks from merge to landing in a channel. That feels long. What could I have done differently to speed this up? Should I have proactively asked to include it in the previous staging-next cycle manually? | 22:15:48 |
