| 18 Nov 2021 |
jonringer | In reply to @tomberek:matrix.org We're already a bit behind. btw: would it make sense to increase the staging-next shares to get more priority? welcome to the most annoying part of doing a release, waiting for hydra to crunch builds | 16:30:06 |
jonringer | This is why I'm particularly annoyed around the schedules slipping, it snowballs the anxiety | 16:30:59 |
tomberek | I had a discussion with Equinix metal about interest in supporting the next release with additional hardware, devoted specifically to ZHF and staging-next builds. | 16:32:17 |
hexa | In reply to @timdeh:matrix.org is there any way to wire up your megaserver as one of the builders temporarily? I don't think anyone questions your integrity here 🙂 that's not going to happen out of pure principle | 19:12:24 |
nrdxp | what principle is that? | 19:12:45 |
hexa | the trust in hydra builders is somewhat ultimate. AFAIU they're all wiped and then netbooted into the same setup and everything happens on a trusted network. | 19:15:29 |
@grahamc:nixos.org | Yeah we have a blank check basically and can allocate more resources as needed without much trouble | 19:24:16 |
tomberek | This goes back to the idea of scaling up/down during ZHF/releases. | 19:27:03 |
@grahamc:nixos.org | yep, we can do that | 19:27:32 |
@grahamc:nixos.org | mind opening a ticket on the nixos-org-configurations repo? I can get that done in a few hours | 19:27:48 |
@grahamc:nixos.org | in particular what types of hw which would be useful | 19:28:00 |
@grahamc:nixos.org | (x86 vs. aarch64 I mean) | 19:28:11 |
@grahamc:nixos.org | * (x86-linux vs. aarch64-linux I mean) | 19:28:16 |
tomberek | at the moment it seems something else is slowing down/blocking x86_64-linux builds, not necessarily machine limitations. But i'll post something anyway, just so we get into the habit of it for future releases. | 19:35:39 |
@grahamc:nixos.org | 👍️ it'd be good to have "open a ticket to scale up" and "open a ticket to scale down" in the RM process documentation | 19:36:21 |
@grahamc:nixos.org | * 👍️ it'd be good to have "open a ticket to scale up" and "open a ticket to scale down" in the RM process documentation, too | 19:36:22 |
tomberek | https://github.com/NixOS/nixos-org-configurations/issues/186 | 19:43:35 |
tomberek | https://github.com/NixOS/release-wiki/pull/22 | 19:47:20 |
| 19 Nov 2021 |
| sterni joined the room. | 11:04:25 |
| Linux Hackerman is moving: @linus:schreibt.jetzt joined the room. | 14:19:49 |
nrdxp | If I'm not mistaken, it is branch off day 🙂
Would my fellow RMs wish to have a meeting sometime today to do the honors? | 16:03:05 |
jonringer | woah there | 16:39:05 |
jonringer | branch off is coniditional on staging-next getting merged :( | 16:39:15 |
jonringer | * branch off is condiitional on staging-next getting merged :( | 16:39:22 |
nrdxp | Yeah, I figured that might be the case, I just remember deciding on today originally, so I just wanted to ask | 16:45:11 |
jonringer | you work on hydra's time, not the other way around | 16:45:29 |
jonringer | I'm rubbing its belly and saying nice things into its ear | 16:45:42 |
jonringer | found out that hydra resources were very low (we get spot instances, just high demand in those regions+instances) | 19:33:15 |
jonringer | updated discourse thread https://discourse.nixos.org/t/lets-have-a-great-21-11-release-cycle/15262/9? | 19:33:23 |
jonringer | Also, @domenkozar:matrix.org would like to do a mass nixfmt across the code base before branch off.
And I would agree. A lot of merge conflicts are related to odd formatting, and it would reduce the amount of "don't look at that code" which has been hanging around in nixpkgs for too long
| 19:35:02 |