| 11 Sep 2021 |
 sterni | everything built on hydra is available via the binary cache, so if something has been built, it's substituted | 21:15:40 |
| sterni | staging-next is built on hydra, staging not | 21:16:30 |
| sterni | * staging-next is built on hydra, staging not atm | 21:16:36 |
 asbachb | I see. | 21:16:40 |
| asbachb | Sorry. | 21:18:02 |
| 12 Sep 2021 |
 Vladimír Čunát | staging as well but only in a very small jobset: https://hydra.nixos.org/jobset/nixos/staging-small | 08:22:49 |
 Sandro | What is blocking the current cycle? Darwin only? | 08:30:41 |
| Vladimír Čunát | *-linux has 20k unfinished jobs. | 08:32:02 |
| Vladimír Čunát | and maybe a couple hundred regressions? | 08:32:34 |
| Sandro | We need to do a hands on on how to read hydra... | 08:32:35 |
| Vladimír Čunát | https://hydra.nixos.org/eval/1703680?filter=linux&compare=trunk | 08:32:53 |
| Vladimír Čunát | On a quick glance there's lots of python stuff that's regressed. | 08:33:59 |
| Sandro | Great! That view is helpful. Not sure how to get there. Maybe we should always post such links to make it easier for people to help in the effort. | 08:35:04 |
| Sandro | I might be able to look at some of those things later | 08:35:21 |
| Vladimír Čunát | Well, you get the first part by clicking on the latest eval. Then I ususally edit to add ?compare=trunk but on the right there's a blue "compare to" button. (and the filtering has an easy box) | 08:38:42 |
| Vladimír Čunát | * Well, you get the first part by clicking on the latest eval (when viewing the jobset). Then I ususally edit to add ?compare=trunk but on the right there's a blue "compare to" button. (and the filtering has an easy box) | 08:39:02 |
| Vladimír Čunát | BTW, compare= also accepts number of any particular evaluation. | 08:39:24 |
| Sandro | Thanks | 09:10:16 |
 lukegb (he/him) | https://p.lukegb.com/raw/ViolentlySteadyHound.png heh | 13:36:23 |
| Sandro | that is not going anywhere | 13:41:11 |
| lukegb (he/him) | The linux builds are fine | 13:41:20 |
| lukegb (he/him) | (the queue runner seems to believe that there are 4 jobs scheduled on each of the m1 macs, although they don't show up on the machine status page) | 13:57:45 |
| Sandro | I think we should merge staging-next at this point. Most of the python regressions are known or can be easily fixed after the merge. | 18:50:12 |
 hexa | there are no substantial python regressions | 18:53:02 |
| Sandro | at this point there is more and more stuff on staging-next that is needed on master. One example is rust 1.54. Crates start to use newer rust features and compilation falls apart. | 18:53:05 |
| Vladimír Čunát | Linux builds will probably catch up within several hours from now. | 18:57:22 |
| Vladimír Čunát | So far the regression numbers+list look OK-ish to me. (on Linux... on Darwin there isn't much yet, really) | 18:59:36 |
| Sandro | Based on the darwin progress the last days we can't really wait for it | 19:00:31 |
| Sandro | you take the queued numbers from https://hydra.nixos.org/queue-summary ? or is there something fancier? | 19:01:52 |
| Vladimír Čunát | Including last-minute stdenv rebuilds is problematic. Especially the early bootstrapping needs lots of serial time, too. | 19:02:33 |
