| 2 Apr 2024 |
 raitobezarius | I'm afraid I need the meeting to move to another day | 12:15:10 |
| raitobezarius | I'm unable to attend on Tuesday from now on | 12:15:14 |
 edef | okay, what would work? | 12:15:40 |
| edef | i'm having a pretty rough day, so punting it forwards seems fine to me | 12:15:55 |
| edef | and i'm pretty indifferent to which particular day we move it to | 12:16:23 |
|  mvnetbiz joined the room. | 12:23:09 |
| raitobezarius | Here's my proposal: https://meet.dgnum.eu/nixos-archivists-weekly-meeting-421217 | 12:52:47 |
| raitobezarius | edef: ^ | 12:53:05 |
| raitobezarius | Let's fill this to find the optimal weekly slot | 12:53:20 |
| raitobezarius | Jonas Chevalier: flokli Eelco as usuals of the meeting ^ | 12:53:40 |
| 9 Apr 2024 |
|  symphorien joined the room. | 20:40:03 |
|  @patka_123:matrix.org joined the room. | 20:40:41 |
|  SomeoneSerge (back on matrix) joined the room. | 20:43:39 |
| symphorien | hey, I'm interested in having more debug info in nixpkgs, which has storage size implication (debuginfo is frequently as large as the executable, sometimes more). I have proposed a way https://github.com/NixOS/nix/pull/8080 to mark debuginfo NARs as debuginfo so that a different AWS retention policy can be set for those, but this approach has been deemed not generic enough. Do you have other approaches in mind that could allow having debuginfo in the cache for, say, 6 months? | 20:46:29 |
|  @federicodschonborn:matrix.org joined the room. | 21:01:26 |
|  thubrecht joined the room. | 21:03:58 |
|  connor (burnt/out) (UTC-8) joined the room. | 21:24:25 |
|  Toma joined the room. | 21:34:00 |
|  ajaxbits joined the room. | 22:35:36 |
| 10 Apr 2024 |
|  @aloisw:kde.org joined the room. | 04:34:47 |
| edef | In reply to @symphorien:xlumurb.eu
hey, I'm interested in having more debug info in nixpkgs, which has storage size implication (debuginfo is frequently as large as the executable, sometimes more). I have proposed a way https://github.com/NixOS/nix/pull/8080 to mark debuginfo NARs as debuginfo so that a different AWS retention policy can be set for those, but this approach has been deemed not generic enough. Do you have other approaches in mind that could allow having debuginfo in the cache for, say, 6 months? the biggest problem there is that it places much stronger requirements on reproducibility | 07:33:30 |
| symphorien | Reproducibility is already blamed today https://github.com/NixOS/nix/issues/7756 so it's not sure it would make the situation worse | 08:15:09 |
| edef | oof | 22:29:25 |
| 11 Apr 2024 |
 flokli | 🍿 | 15:16:42 |
| 23 Apr 2024 |
|  dnr joined the room. | 03:06:54 |
 @delroth:delroth.net | Hello cache wizards. I was curious how things are progressing regarding running a cache garbage collection (and doing it periodically in the future). Is this blocked on tooling right now? Or is it more a case of not having a good consensus on what to delete? (I think I've heard different plans from different people that I'd consider "involved")
(I have a few infra motivations beyond just cost saving that would benefit from the cache having more temporary properties for orphan/disconnected-from-roots paths.) | 03:52:20 |
| raitobezarius | AFAIK there have been no recent meetings since I asked to move the meeting to another day and sent the rescheduling link | 06:39:01 |
| 24 Apr 2024 |
|  Artturin joined the room. | 18:05:17 |
| 25 Apr 2024 |
| @delroth:delroth.net left the room. | 14:44:46 |
| SomeoneSerge (back on matrix) changed their display name from SomeoneSerge (void) to SomeoneSerge (UTC+1). | 23:02:30 |
