| 20 Mar 2025 |
 jade_ | (and it does not seem like it can be trivially changed to recursive hash mode there, since that breaks a bunch of observable behaviour. i have no idea how this is supposed to behave, but it seems like a really quite large bug) | 19:01:44 |
| jade_ | * (and it does not seem like it can be trivially changed to recursive hash mode there, since that breaks a bunch of observable behaviour. i have no idea how this is supposed to behave, but it seems like a really quite tricky bug) | 19:01:50 |
| jade_ | okay update: i can confirm that nixos/nix master emits that 9hagx path into the store and then additionally gets the q9prs one | 21:56:28 |
| jade_ | * okay update: i can confirm that nixos/nix master emits that 9hagx path into the store and then additionally makes the q9prs one | 21:56:33 |
| jade_ | that seems at least inefficient if the files are large | 21:56:41 |
| jade_ | repro:
builtins.fetchTree {
narHash = "sha256-WmkitUnTwTCCdq0a08EdUBC/UTo/i4+xwD3mk19NA1Q=";
type = "file";
url = "https://releases.lix.systems/lix/lix-2.90.0/manifest.nix";
}
| 21:57:01 |
| jade_ | found some bonus context: https://github.com/NixOS/nix/pull/6548#discussion_r877921756 | 23:09:50 |
 hexa | the builders are returning to nix 2.24.12 tonight | 23:24:13 |
| hexa | the version on nixos-24.11 | 23:24:32 |
| jade_ | wow. i can change it to recursive and it passes tests? absurd. | 23:26:21 |
| jade_ | that should not be possible. our tests clearly don't work if that's true | 23:26:38 |
| jade_ | explains the bug though :/ | 23:26:53 |
| jade_ | (i broke tests by writing code that was wrong, but once i fixed my code, it passed, which, aaaaa) | 23:29:27 |
| 21 Mar 2025 |
|  @orzklv:floss.uz joined the room. | 00:24:35 |
| jade_ | fix on the lix side https://gerrit.lix.systems/c/lix/+/2864 | 00:24:52 |
| jade_ | if nothing else please steal my tests | 00:25:02 |
| jade_ | * if nothing else please feel free to take my tests | 00:25:09 |
|  @0rzklv:matrix.org left the room. | 00:30:01 |
|  soopyc (she/it) changed their profile picture. | 02:19:19 |
|  Domen Kožar changed their profile picture. | 11:39:47 |
| hexa | most likely https://github.com/NixOS/nix/issues/8946 | 15:13:31 |
| hexa | we've seen like 500 coredumps since last night | 15:13:48 |
| 22 Mar 2025 |
|  isabel joined the room. | 00:40:09 |
 John Ericson | In reply to @hexa:lossy.network
we've seen like 500 coredumps since last night @hexa:lossy.network: wow ok, so that's a ton of coredumps, definitely the bug is back. What is the baseline rate? | 18:56:17 |
| hexa | really depends on how busy hydra is | 18:57:17 |
| hexa | but this was half a day's worth | 18:57:34 |
| John Ericson | @joerg:thalheim.io @picnoir:alternativebit.fr @jade_:matrix.org: thank you all for investigating this | 19:00:30 |
| John Ericson | (I see the PR now) | 19:00:38 |
|  @icarus_dh:matrix.org removed their profile picture. | 21:48:13 |
| @icarus_dh:matrix.org removed their display name icarus_dh. | 21:48:37 |
