| 21 Aug 2025 |
 dramforever | 2025 and early 2026 risc-v is just not happening | 16:05:00 |
 hexa (signing key rotation when) | (the pattern is weird, because meet wants it that way 😄) | 16:05:13 |
| hexa (signing key rotation when) | (will probably be changed down the road, not sure) | 16:05:22 |
 Mic92 | Will join in 5 min | 16:06:57 |
 Jeremy Fleischman (jfly) | nevermind, looks like i have access | 16:07:54 |
| Jeremy Fleischman (jfly) | err, view only access. so edit would be nice | 16:10:16 |
| Jeremy Fleischman (jfly) | oh, silvan's here. we'll get access =) | 16:10:50 |
| 22 Aug 2025 |
|  elikoga changed their profile picture. | 17:28:52 |
 Winter | you | 21:58:04 |
| Winter | * you'd think the drv would always be available on mimas? | 21:58:16 |
| Winter | unless we just like... gc drvs ~consistently | 21:58:26 |
| Winter | which would explain it | 21:58:31 |
| Winter | but either way surprised this isn't persisted in the hydra db | 21:58:40 |
| hexa (signing key rotation when) | we keep derivations, but not outputs | 22:06:08 |
| 23 Aug 2025 |
 dgrig | Copy pasting from the nix manpage:
keep-derivations ⟨#conf-keep-derivations⟩
If true (default), the garbage collector will keep the derivations from which non-garbage store paths were built. If false,
they will be deleted unless explicitly registered as a root (or reachable from other roots).
Keeping derivation around is useful for querying and traceability (e.g., it allows you to ask with what dependencies or op‐
tions a store path was built), so by default this option is on. Turn it off to save a bit of disk space (or a lot if keep-
outputs is also turned on).
Since we're gc'ing outputs and it specifies "will keep the derivations from which non-garbage store paths were built", I believe this will also gc derivations?
| 00:34:24 |
| hexa (signing key rotation when) | ah, yeah | 00:35:17 |
| 25 Aug 2025 |
|  @federicodschonborn:matrix.org changed their display name from Cat McFishface 🏳️🌈 (he/they) to Mr. Bones (he/they). | 13:37:34 |
| 26 Aug 2025 |
| Winter | is it expected that channels.nixos.org lags behind the hydra jobsets? e.g. https://hydra.nixos.org/jobset/nixos/trunk-combined#tabs-evaluations, i fetched nixos-unstable on 2025-08-25 10:55 -0400 and got https://hydra.nixos.org/eval/1817796 from the 19th | 21:58:21 |
| Winter | ~5-6 days of lag seems... odd? | 21:58:35 |
 Ramses 🇵🇸 | Those evals didn't cause channel bumps, IIUC. The tested job was broken for them | 22:35:48 |
| Ramses 🇵🇸 | So that's intended | 22:36:21 |
| Ramses 🇵🇸 | Winter: ^ | 22:36:37 |
| 27 Aug 2025 |
 vcunat | Details are in https://wiki.nixos.org/wiki/Channel_branches#Internal_channel_update_process | 05:04:03 |
 Arian | Aggregate jobs always trip me up. Especially because they're so hard to find in the hydra UI. Wish we pinned them to the top of the eval view or something | 08:31:30 |
| vcunat | You'd better use https://status.nixos.org/ | 08:33:53 |
| vcunat | Hydra itself does not know that some jobs are special. | 08:34:18 |
| vcunat | (in the sense that they block channels) | 08:34:33 |
| vcunat | That logic is external: https://github.com/NixOS/nixos-channel-scripts/blob/master/mirror-nixos-branch.pl | 08:35:14 |
| hexa (signing key rotation when) | will reboot haumea and mimas for an update and the nftables migration | 13:35:22 |
| hexa (signing key rotation when) | now that we've essentially run out of linux runnables 😉 | 13:35:42 |
