Yes, essentially the JSON schema converter needs to be aware of $defs. But that feels like half a rewrite (which I concluded…
Yeah but that's fine, nixops can keep living in its flake world.
@kiara is this small enough to be reviewable for you? I'd maybe split adding the new Python package into a separate commit and squash the rest. Making it nicer is blocked on the JSON schema…
I squashed it because the change is rather self-contained, and the individual commits didn't make all that much sense in isolation
Not sure how that's related to where the code lives. We wanted the deployment service to be able to display progress, so we need to expose such an interface.
Is this closed because migrations are out of scope? Because notifications would still be a thing even without a GUI.
We'd still need to expose some sort of streaming API, no?
Don't we need to somehow feed back whatever netbox provides us to the API exposed to operators?
I think we still need that one way or another. We have a [rudimentary glossary](5db04d0c50/README.md (glos)…
What I meant here is our infrastructure setup, not the deployments we run for (even demo) operators. Currently it's a table with which machines are used for what, and since we have that in an…
This needs to be rewritten to a user story, and if I try imagining what it would be like, I don't see the difference to Fediversity/Fediversity#109. Isn't all of…
Apart from the "viewing" (as in "in the browser") the problem of dealing with what amounts to API versioning for the union of all services still remains, so I don't think this issue can easily be…
They were hard-coded to have something for an initial demo, so writing down how to get rid of them is not wrong. Of course the next better solution would be to mock away an API for registering DNS…
hm, would it not be better to mark this as WIP while as per the current PR description that is its current status?
@kiara the WIP in the PR description talks about follow-up work. This one is…