!RROtHmAaQIkiJzJZZE:nixos.org

NixOS Infrastructure

360 Members
Next Infra call: 2024-07-11, 18:00 CEST (UTC+2) | Infra operational issues backlog: https://github.com/orgs/NixOS/projects/52 | See #infra-alerts:nixos.org for real time alerts from Prometheus.113 Servers

Load older messages


SenderMessageTime
10 Jul 2025
@zhaofeng:zhaofeng.liZhaofeng Liit could even be done on the S3 end too19:22:18
@arianvp:matrix.orgArianI'm not sure howe we could do this on the S3 end19:35:07
@arianvp:matrix.orgArianAh you mean: detect when something gets pushed to S3?19:35:31
@arianvp:matrix.orgArianoh yeh we could do that. We could have a Lambda or something that triggers on S3 uploads and purges the path from fastly19:36:13
@arianvp:matrix.orgArian

we do 12,078,871 PUTs per month

lambda costs 20cts per million requests. So this would cost $2.40

We could have a simple lambda trigger for each PUT and send a curl -XPURGE to the specific url in fastly

19:45:18
@emilazy:matrix.orgemilyis it billed on the fastly end?19:53:12
@arianvp:matrix.orgAriandont think so20:08:55
11 Jul 2025
@rnhmjoj:eurofusion.eurnhmjoj joined the room.09:32:39
@rnhmjoj:eurofusion.eurnhmjoj

just to be sure, to get the name of a test for the tested job (nixos/release-combined.nix) I take the attribute name and replace nixosTests with nixos.test, right?

es. nixosTests.networking.networkd.bridge -> nixos.tests.networking.networkd.bridge

09:38:57
@sternenseemann:systemli.orgsterni (he/him)Seems to happen a lot in staging-next at the moment as well, haven't been able to trigger this on any of my own machines yet even with big rebuilds…09:48:27
@k900:0upti.meK900 I wonder if this is related to @DavHau's patchShebangs changes 09:52:51
@dramforever:matrix.orgdramforever sterni: do you have a concrete example package? i want to take a look at what could possibly be in the build process 09:55:02
@hsngrmpf:matrix.orgDavHaustaging-next doesn't seem to have the latest fix for patch-shebangs09:55:06
@k900:0upti.meK900We're getting close to the end of the cycle now09:58:02
@hsngrmpf:matrix.orgDavHauOpened a backport PR against staging-next: https://github.com/NixOS/nixpkgs/pull/424278/files09:58:03
@dramforever:matrix.orgdramforever oh that would make sense, the haskell Setup is usually binary 09:58:42
@k900:0upti.meK900 We can't 09:58:59
@k900:0upti.meK900It'll scrap the entire cycle09:59:03
@sternenseemann:systemli.orgsterni (he/him) dramforever: https://hydra.nixos.org/build/301811455/nixlog/1 09:59:49
@k900:0upti.meK900I'm running a new eval on -next10:00:10
@sternenseemann:systemli.orgsterni (he/him) DavHau: does that try to patchShebang every binary in parallel with other phases? 10:00:13
@k900:0upti.meK900And will restart all failed after10:00:16
@k900:0upti.meK900 In parallel but not with other phases 10:00:22
@sternenseemann:systemli.orgsterni (he/him) hm but when? Setup is created in compileBuildDriverPhase which is only executed before configurePhase 10:01:24
@hsngrmpf:matrix.orgDavHauNo, this is just a bug fix for something I implemented for macos compat. It is not adding parallism10:01:28
@k900:0upti.meK900patchShebangs doesn't auto-run10:01:56
@k900:0upti.meK900Except in fixupPhase10:02:02
@sternenseemann:systemli.orgsterni (he/him)oh that was unrelated, got confused10:06:49
@hexa:lossy.networkhexasounds correct10:16:43
@hexa:lossy.networkhexawe then append the system at the end10:16:52

Show newer messages


Back to Room ListRoom Version: 6