Replies: 7 comments
-
|
It's tough to draw a line between personal and professional when you manage them in the same way. I think the scope of the original vision does include professional work. I Just wasn't 100% positive of that in the beginning. I think this would be remarkably useful, although perhaps the work on Arion would be a better nix-first approach? Of course, there is nothing wrong with supporting both, since the bulk of the implementation and maintanence is left to the user. |
Beta Was this translation helpful? Give feedback.
-
Indeed, arion might be a better intermediate approach, though it doesn't come with the same distributed self healing capabilities of k8s. For the reference, some prime useful applications that I would find interesting to deploy that way (since they are "k8s-native"):
( we could eventually put together a secured company template environment on a
) |
Beta Was this translation helpful? Give feedback.
-
|
Even further reference:
|
Beta Was this translation helpful? Give feedback.
-
|
if there is to be a remote deployment solution, I would prefer this type of repository to remain nix-only configurations. I don't think its a good idea to deviate from that, after all this is supposed to be a gateway into nixos. If a deployment system does come to exist here, I've been meaning to combine my fork with nixus, a 100% nix-lang deployment system. Mostly to run |
Beta Was this translation helpful? Give feedback.
-
Given nix's excellence, I totally understand the desire to favour nix-centric technologies and keep out it's direct competitor (which happen to also get a lot more attention — unfairly so because of huge marketing budgets). Let's split this issue's proposal into it's components though: There is NixOS and nix-lang (a configuration management language). NixOS shines at almost anything, except for self healing capabilities. nix-lang shines at everything. K8s brings not only those self-healing capabilities, but also has a huge ecosystem for us to leverage, where we should read the word ecosystem as being a reified massive existing configuration database. Combining k8s with nix-lang (later nickel) and complement where nixOS falls short is — in my opinion — a match made in heaven. |
Beta Was this translation helpful? Give feedback.
-
@tgunnoe, I agree with this sentiment. Though, as long as the implementation remains unobtrusive and completely optional, I'm okay with kubenix support as well. I just don't want it to be mandatory. |
Beta Was this translation helpful? Give feedback.
-
|
I think though kubenix needs some overhauling in its current state. I'll be trying to support @adrian-gierakowski — who agreed to adopt maintainership of kubenix — in this endeavor in any way I (usefully) can. ... maybe |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Some things are just better (more stably) deployed in kubernetes.
Acknowledging this statement, it would be extremely nice to integrate or at least have a pathway to deploy kubernetes applications on devos hosts (presumably through kubenix).
Broader context: I feel while targeted at individual's uses, this repository can have a perfect second life to structure small business / company environments.
Even broader context: nickel is explicitly designed as a general purpose configuration language. And divnix/devos has already done an excellent job at integrating complex, but genious, stuff.
Feel very free to close, if this is off-direction. 😉
Further resources:
Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.
Beta Was this translation helpful? Give feedback.
All reactions