| 19 Jun 2023 |
 @rendakuenthusiast:imperishable.name | also an example of how to use nodeSpecialArgs, although I don't think I personally need that | 05:59:15 |
 Zhaofeng Li | In reply to @zhaofeng:zhaofeng.li
I usually just do meta.specialArgs = inputs so any config module can access all flake inputs Wait oops, should be specialArgs = { inherit inputs; }; | 06:00:08 |
| @rendakuenthusiast:imperishable.name | error: syntax error, unexpected INHERIT | 06:00:37 |
| @rendakuenthusiast:imperishable.name | oh wait need to wrap it in a set | 06:00:53 |
| Zhaofeng Li | otherwise it will look like { pkgs, config, home-manager, ... } etc, but might be fine depending on what you like | 06:00:58 |
| @rendakuenthusiast:imperishable.name | although now it objects to specialArgs = { inherit inputs }; | 06:01:16 |
| @rendakuenthusiast:imperishable.name | er, forgot the closing ; | 06:01:30 |
| @rendakuenthusiast:imperishable.name | inputs is undefined within the body of the outputs fn | 06:01:54 |
| @rendakuenthusiast:imperishable.name | does it need to be self.inputs? | 06:02:05 |
| @rendakuenthusiast:imperishable.name | or I guess I could just pass in the flake I care about in the outputs fn and then pass that as a special arg, I only need 2 or 3 flakes | 06:02:29 |
| Zhaofeng Li | Ah, you do something like outputs = { self, nixpkgs, ... } @ inputs: { ... } | 06:02:30 |
| Zhaofeng Li | The @ inputs will bind the entire set as inputs | 06:02:46 |
| @rendakuenthusiast:imperishable.name | @ syntax works on either side of the fn input set, right? | 06:02:52 |
| Zhaofeng Li | yes | 06:02:57 |
| @rendakuenthusiast:imperishable.name | I'm not sure why they needed to implmenet it both ways | 06:03:17 |
| @rendakuenthusiast:imperishable.name | anyway that seems to work | 06:03:19 |
| @rendakuenthusiast:imperishable.name | and that did update my flake.lock | 06:03:35 |
| @rendakuenthusiast:imperishable.name | with actually kind of a lot of stuff. I set up that flake I'm importing myself, like two days ago, I may not have done it correctly | 06:04:23 |
| @rendakuenthusiast:imperishable.name | but in any case I can deploy | 06:04:25 |
| @rendakuenthusiast:imperishable.name | thanks for your help! | 06:04:29 |
| 22 Jun 2023 |
|  @h4636oh:matrix.org joined the room. | 02:47:16 |
| @h4636oh:matrix.org left the room. | 03:37:55 |
|  kenji joined the room. | 05:55:29 |
 @palo:xaos.space | I'm sketching an Architecture for a service in an openstack cluster, which consists of multiple machines (and there will be new machines every now and than). I would like to have one machine that shares it's /nix/store with all the other instanzes in the cluster. That is not a problem. I have a mono repository and would like to build all machines (not deploy) and nix-copy-closure the resulting closures to this machine. Meaning the build server (which is not the cache server) is in a loop while true; git pull ; colmena build ; nix-copy-closure ???? private_cache_server How to find all closures that need to be copied? | 08:12:45 |
 Wanja Hentze | would you even be using colmena then? sounds more like you'd be reimplementing it | 08:14:48 |
| @palo:xaos.space | I want to use colmena. Because we would not deploy all machines instantly. | 08:15:40 |
| @palo:xaos.space | I just want to enrich the cache. | 08:16:07 |
| Wanja Hentze | oh, cache warming | 08:16:37 |
| @palo:xaos.space | All machines have a very similar base setup, and updates should be pulled from the cache, to avoid uploading 50 times the same closures from my office to the data center. | 08:18:00 |
| Wanja Hentze | then why not use `colmena apply push` | 08:18:07 |
