Matthew Booth
Matthew Booth
> Also, I think we might surface why the desired output cannot achieved by simply naming KCD or the MachineDeployment according to the 'corporate naming policy'. For MachineDeployments we *can*...
> There are 3 layers of this problem: > > 1. How to generate the name for objects in the management cluster > 2. Provide guidelines to infrastructure providers about...
> @mboersma to check this from the MachinePool PoV (if applicable) FYI @sbueringer already had some thoughts on this: https://kubernetes.slack.com/archives/C8TSNPY4T/p1715178789376599?thread_ts=1715173795.198159&cid=C8TSNPY4T tl;dr My initial suggestion was to add this naming scheme...
> @lentzi90 @mdbooth what about adding a few detail about implementation + bringing this up at the office hours, so then we can move on with implementation? Unfortunately I don't...
@fabriziopandini With the workaround in place we have some breathing space and I'm not in a rush. How long should I leave it?
Ok. I'll update the description before tomorrow's office hours. It'll be based on @lentzi90 's namingStrategy proposal.
> I'm not opposed to start the implementation from the control plane (TBD if we consider this something KCP specific or if we want somehow to make this a feature...
/remove-lifecycle stale
I think I've said this elsewhere, but just in case: a brief recap. Microversions define different APIs with different interfaces and different semantics. They are usually (but not always) pretty...
The 'user' in the context of that documentation is the entity making the REST API call, so in that context the user is CAPO. The 'user' (i.e. CAPO) gets to...