| 27 Jul 2025 |
 emily | well I dunno. it's probably trivial to own the box with --option sandbox false already | 16:13:05 |
| emily | well | 16:13:24 |
| emily | "trivial" :) | 16:13:25 |
 raitobezarius | so dropping Darwin | 16:13:27 |
| emily | it's probably hard | 16:13:33 |
| emily | but it's somewhat scary still | 16:13:43 |
| raitobezarius | In reply to @gilice:matrix.org
sent a file. this is freaking awesome, thank you so much!! | 16:13:44 |
| emily | 2× VM per physical host, always rolled back between builds, is what would make me feel comfortable doing builds sans sandbox | 16:14:10 |
| emily | ofc this requires integration work with e.g. tart | 16:14:35 |
| raitobezarius | i mean one physical host cannot afford much more than one VM I think | 16:14:37 |
| emily | but shouldn't be too bad | 16:14:37 |
| raitobezarius | In reply to @emilazy:matrix.org
ofc this requires integration work with e.g. tart if you open an issue with pointers, i can take a look | 16:14:47 |
| raitobezarius | but i cannot embark on a self learning journey with tart :D | 16:14:55 |
| emily | then one job slot? | 16:14:59 |
| emily | if I context switch much more I'll drop the ball on stuff I need to do first :) but I will try to open an issue today | 16:15:23 |
| raitobezarius | hmmmmmmmmmmmmmm | 16:15:26 |
| raitobezarius | what I'd like to do is to ensure that a given build can use 2 job slots of the same machine | 16:15:37 |
| raitobezarius | but 2 builds cannot use 2 job slots of the same machine inside the same VM | 16:15:46 |
| emily | ok | 16:15:50 |
| emily | so just mutex around the CI? | 16:15:54 |
| raitobezarius | maybe 2 VMs and 1 job slot for each VM | 16:15:56 |
| raitobezarius | In reply to @emilazy:matrix.org
so just mutex around the CI? only halal proposals please | 16:16:03 |
| emily | I thought by job slot you meant like, CLs | 16:16:10 |
| emily | not max-jobs | 16:16:12 |
| raitobezarius | sorry | 16:16:16 |
| raitobezarius | job slot in my jargon refers to buildkite slots | 16:16:23 |
| raitobezarius | 1 slot is consumed by one derivation in our CI currently | 16:16:33 |
| emily | I know nothing of Buildkite :) | 16:16:35 |
| emily | ok | 16:16:37 |
| emily | per derivation or per leaf | 16:16:45 |
