| 4 Jun 2023 |
 Robert Hensing (roberth) | You could also speed up the test by not starting any vms. It's all about evaluation anyway so if we reach the build, we know that the lib wiring was ok | 10:18:39 |
 baloo | Ha I missed the all-tests I couldn’t find them yesterday night. I’ll send that later today. | 13:49:21 |
|  Federico Damián Schonborn changed their profile picture. | 17:40:04 |
| baloo | Robert Hensing (roberth): I pushed an new commit, but I'm not sure that's much better. It still runs a vm to build, not too sure if I can make it not evaluate:w | 22:04:26 |
| baloo | * Robert Hensing (roberth): I pushed an new commit, but I'm not sure that's much better. It still runs a vm to build, not too sure if I can make it not evaluate | 22:04:30 |
| baloo | nevermind, I can just pull the toplevel of the inner system >< | 22:06:26 |
| 10 Jun 2023 |
 raitobezarius | Robert Hensing (roberth): you might want to take a look to https://github.com/NixOS/nixpkgs/pull/228346 | 16:17:32 |
| raitobezarius | (this PR is like only a "model PR" that I hacked to understand the current challenges of netboot testing) | 16:18:19 |
| raitobezarius | (directBoot is now merged, diskless VMs might be much simpler thanks to the persistent PR from nikstur, etc, etc.) | 16:18:36 |
| 12 Jun 2023 |
| baloo | would it be possible to get a review on https://github.com/NixOS/nixpkgs/pull/235874 ? Or maybe even a merge? :) | 16:34:21 |
| 13 Jun 2023 |
 chreekat | I commented :P | 18:05:32 |
| baloo | I'm not too sure I understand how the code is different from the description to be honest. | 18:39:43 |
| baloo | maybe only review the first commit, the second adds noise because it brings a test to cover the issue. | 18:40:08 |
| Federico Damián Schonborn changed their profile picture. | 20:56:03 |
| 14 Jun 2023 |
| Robert Hensing (roberth) | In reply to @raitobezarius:matrix.org
(directBoot is now merged, diskless VMs might be much simpler thanks to the persistent PR from nikstur, etc, etc.) a rebase would reviewers I think | 10:27:31 |
| Robert Hensing (roberth) | * a rebase would help reviewers I think | 12:12:07 |
| 8 Jul 2023 |
 jonge | hey friends, i created a PR that syncs the docstrings of the Machine class's methods and the nixos docs of the test driver. if anybody would be interested to look at this this would be a great help: https://github.com/NixOS/nixpkgs/pull/242135 | 10:09:27 |
| raitobezarius | Will take a look when I can :-) | 10:53:52 |
| raitobezarius | Thank you! | 10:53:55 |
| 10 Jul 2023 |
|  nikstur joined the room. | 14:29:36 |
| nikstur | Is there any simple way I can wait for a machine to reboot? I want to test a service that automatically reboots the machine on success (as unitConfig.SuccessAction = "reboot"; so the reboot is not initiated by the test itself | 14:31:18 |
| nikstur | I'll even take a complex way because I can't figure it out :D | 14:43:53 |
| raitobezarius | wait_for_unit("$unit.service") ? :p | 14:44:13 |
| raitobezarius | after a machine.crash() ? | 14:44:21 |
| nikstur | Works only if I trigger the crash from within the test. But the service running in the VM triggers the reboot | 15:01:01 |
| raitobezarius | hmm | 15:01:12 |
| raitobezarius | so you want something to say "now you are rebooting" | 15:01:22 |
| nikstur | And waiting for reboot.target etc doesnt work either | 15:01:27 |
| raitobezarius | our protocol to assert the state of the machine is suboptimal for that | 15:01:43 |
| nikstur | wait_for_reboot() or something | 15:01:44 |
