| 1 Jun 2023 |
 hexa | and yet it fails on ofborg | 15:26:00 |
| hexa | the systemd unit is a bit involved | 15:26:16 |
| hexa | https://github.com/NixOS/nixpkgs/blob/master/nixos/modules/services/security/kanidm.nix | 15:27:41 |
| hexa | e.g. | 15:52:44 |
| hexa |
server # [ 5.835867] (kanidmd)[834]: kanidm.service: Failed at step NAMESPACE spawning /nix/store/ln4cfw2x9v5n41axccgb6b6mz27lz2dv-kanidm-1.1.0-alpha.12/bin/kanidmd: No such file or directory
| 15:52:45 |
| hexa | so errors related to cgroups | 15:53:02 |
| hexa | or on the aarch64 community builder | 15:56:26 |
| hexa |
server # [ 5.579879] (kanidmd)[835]: kanidm.service: Failed at step NAMESPACE spawning /nix/store/2325mmgcc3mdz1l8v6jp0s2mhkv1kch8-kanidm-1.1.0-alpha.12/bin/kanidmd: No such file or directory
| 15:56:27 |
| hexa | [hexa@aarch64:~]$ /nix/store/2325mmgcc3mdz1l8v6jp0s2mhkv1kch8-kanidm-1.1.0-alpha.12/bin/kanidmd
daemon
USAGE:
kanidmd <SUBCOMMAND>
OPTIONS:
-h, --help Print help information
SUBCOMMANDS:
configtest Test the IDM Server configuration, without starting network listeners
database Database maintenance, backups, restoration etc
db-scan Inspect the internal content of the database datastructures
domain Change domain settings
healthcheck Load the server config and check services are listening
help Print this message or the help of the given subcommand(s)
recover-account Recover an account's password
server Start the IDM Server
version Print the program version and exit
[hexa@aarch64:~]$
| 15:56:36 |
 cole-h | Not as far as I'm aware. The x86 builders' configs live here: https://github.com/ofborg/infrastructure/tree/master/morph-network/machines | 16:21:37 |
| 3 Jun 2023 |
|  zseri changed their display name from zseri to Alain Φογτια Zscheile. | 12:23:28 |
| zseri changed their profile picture. | 12:35:26 |
|  @emilytrau:matrix.org left the room. | 17:46:24 |
| 6 Jun 2023 |
|  Vladimír Čunát changed their display name from Vladimír Čunát to @vcunat. | 17:20:57 |
| 7 Jun 2023 |
 Artturin | https://github.com/NixOS/ofborg/pull/642 | 04:16:17 |
 K900 | https://github.com/NixOS/nixpkgs/pull/236329 any idea why eval didn't fail here? | 08:39:05 |
 Lily Foster | In reply to @k900:0upti.me
https://github.com/NixOS/nixpkgs/pull/236329 any idea why eval didn't fail here? If it's an alias issue, see my message from about a month ago: https://matrix.to/#/!sBfrWMVsLoSyFTCkNv:nixos.org/$HPOU0iB59i57szcLa8l4X8aAPfQPnUP3_uH81nAP-iw?via=lily.flowers&via=matrix.org&via=nixos.dev | 11:53:00 |
| K900 | It is | 11:53:10 |
| Lily Foster | I did do that testing to see how much it impacted eval time and it impacted it more than I would have liked, so I was going to look for another solution (and then stopped having spare time...) | 11:53:57 |
| Lily Foster | Yeah I just took a look at that diff and it is the same issue (fishing an alias out of pkgs doesn't get caught by that test since it really doesn't eval much past callPackage args) | 11:56:31 |
| Lily Foster | I'll try to make a PR in a few hours to ofborg to just do --drv-path for that test to get feedback/discussion on it (assuming I have a spare moment to do so at $dayjob this morning) | 11:58:07 |
 Mindavi | We should also just use a lot less aliases and stop introducing them haphazardly. Just make them throw immediately maybe, with the new attr name (if any) | 16:44:17 |
| hexa | People got mad at me the last time I converted aliased to throws | 16:47:59 |
| hexa | Primarily due to out of tree usage | 16:48:12 |
 maralorn | *cough* warnings RFC *cough* | 21:46:42 |
| 8 Jun 2023 |
|  henrirosten joined the room. | 07:45:53 |
| Lily Foster | I opened https://github.com/NixOS/ofborg/pull/643 for the alias checking problem | 13:49:23 |
| 9 Jun 2023 |
| K900 | Another alias issue: https://github.com/NixOS/nixpkgs/pull/236828 | 11:43:39 |
| K900 | And another one: https://github.com/NixOS/nixpkgs/pull/236833 | 12:14:26 |
| K900 | Today is definitely one of the days of all time | 12:14:32 |
