Dr. Stefan Schimanski
Dr. Stefan Schimanski
### Feature Description The mentioned endpoints should return the same status as the global one. ### Proposed Solution Redirect those endpoints under a workspace URL to the top-level one. This...
A CI.md or TESTING.md at the root would be great, e.g. with a link to the openshift/release repo where everything is configured.
Follow-up of https://github.com/kcp-dev/kcp/issues/417. Problem: the names of CRDs in the `system:bound-crds` workspace have cryptic names. For debugging it would be very helpful to have the resource and group name **plus**...
We do not want all system components (shards, virtual workspace apiserver, syncers, proxy, ...) have admin access each. Hence, we have to work out a plan how to allow them...
### What happened? When trying to connect to an unreachable space, the interactive mode shows it greyed out: But when trying to connect non-interactively it shows: ```sh up ctx upbound/unreachable/default...
### What happened? Trying to add a profile using `up profile --profile robot -a "7b97ab5c..." config set token 'eyJh...`. Surprisingly this mutated my `default` profile. Epic # ### How can...
**What this PR does / why we need it**: The client-go scheme is global and not go-routine-safe (by design). Adding to it from the client constructor leads to data races...
### What happened? https://github.com/crossplane/crossplane/blob/master/internal/initializer/crds.go has no code to wait for the CRDs to be established and served. The result is a race between further init steps and kube-apiserver that can...
## Summary The old code only added webhook sources, but never cleaned up. This PR bump kube to get CEL optimization for webhook filters, and it properly caches the webhook...
## Summary NOT SURE THIS PR IS CORRECT: found https://github.com/kcp-dev/kcp/issues/2935 and https://github.com/kcp-dev/kubernetes/pull/132. The comparison in the replication controller between cached and local object didn't filter `generation` and `managedFields`. This was...