| 27 Jun 2026 |
 emily | seems about true looking at the job counts on the Hydra jobset | 13:07:56 |
 hexa | it has fewwer build slots | 13:07:59 |
| hexa | Redacted or Malformed Event | 13:08:03 |
| emily | yeah, but # of completed builds / queued builds seems comparable | 13:08:17 |
| emily | the builds themselves are likely faster because of single core perf | 13:08:37 |
| hexa | yeah, we could render steps completed per platform I would hope | 13:08:51 |
| hexa | https://grafana.nixos.org/d/MJw9PcAiz/hydra-jobs?orgId=1&from=now-7d&to=now&timezone=utc&var-machine=$__all&var-job=hydra_queue_runner&viewPanel=panel-30 | 13:32:07 |
| hexa | I'd say … no | 13:32:44 |
| hexa | but we have a budget to buy new macs as you may remember | 13:33:00 |
| hexa | I didn't want to make too many changes at once with the new queue-runner, the x86_64-darwin removal, so I delayed buying them | 13:33:25 |
| hexa | and hoped we could wait for m5 minis | 13:33:31 |
 Vladimír Čunát | Lower "steps / time" might be OK for darwin. We surely have less darwin jobs now than for either *-linux, at least on 26.11+ | 13:37:34 |
| emily | https://hydra.nixos.org/eval/1826634?filter=.aarch64-darwin&compare=1826594&full= has 21165 done / 44420 queued
https://hydra.nixos.org/eval/1826634?filter=.aarch64-linux&compare=1826594&full= has 5999 done / 71800 queued
https://hydra.nixos.org/eval/1826634?filter=.x86_64-linux&compare=1826594&full= has 32739 done / 48016 queued | 13:41:30 |
| emily | seems healthy enough? | 13:41:34 |
| emily | although what doesn't seem healthy is the number of new failures on both Darwin and Linux 😅 | 13:41:59 |
| emily | seems like Darwin now has 0.84× the number of aarch64-linux jobs, 0.81× the number of x86_64-linux ones | 13:43:43 |
| emily | but ofc previously it had ~1.5× either Linux platform | 13:44:19 |
| emily | dunno, maybe I am misreading the stats but it seems like we could broadly expect Darwin to finish at around the same time the Linux platforms do and if anything aarch64-linux is the one lagging quite a lot | 13:45:08 |
 Myria | https://github.com/NixOS/nixpkgs/pull/536047. Still building the sage test, but that might take a while. | 14:02:33 |
| Myria | Have built pandas though, and all the numpy tests succeed | 14:02:51 |
 K900 | Merged master -> next | 14:15:51 |
| K900 | 
Download uh oh lol | 14:17:33 |
 Alyssa Ross | ugh, did a big musl test a couple of days out, and then last minute systemd update got in that doesn't build :( | 14:20:01 |
| hexa | huh | 14:20:48 |
| hexa | I didn't think 261 happend close to the cycle | 14:20:57 |
| hexa | but rather they missed 261.1 | 14:21:01 |
| K900 | -next is on 261, at least | 14:21:38 |
| K900 | (.0) | 14:21:41 |
| emily | can just cherry-pick the relevant PR before landing -next | 14:21:50 |
| emily | I ordered it specifically to avoid breaking CI but didn't think about landing the -next | 14:21:59 |
