Natalie Klestrup Röijezon
Natalie Klestrup Röijezon
FWIW, you can actually do that kind of filtering without memory leaks, by putting the filter as a stage between `watcher` and `reflector`. But I can't really see a good...
Relying on the Kubeconfig itself to decide this would solve nothing, no? The reasoning behind doing this makes sense, but it's awkward that this pretty much breaks any cluster that...
We could let users enable it via an env variable (in addition to the API, like we handle impersonation at the moment). That still wouldn't help for cases where users...
Going into personal opinion, I think 1 is pretty much unavoidable, 2 is often undesirable but may be the least bad option, and 3+ are walking and talking DoS vulnerabilities...
For documenting the policy itself, probably throw it in a document in the repo and maybe link to it from CONTRIBUTING? For enforcing it, I'd probably go as far as...
Not sure I see how `ObjectReference` helps.. its fields are as `Option`al as `ObjectMeta`'s as far as I can tell?
Yeah.. I think finalizers are a kind of special case, since they have "fun" timing requirements (the finalizer MUST be acknowledged by K8s BEFORE we start performing work that they'd...
Oh sorry, didn't see that the canonical repo seems to be on GitLab.
Huh, apparently this is backwards-incompatible, because Rust doesn't automatically downcast `Box` into `Box`. :/
I'd argue that `#[serde(bound)]` is pretty much useless for schemars, since they largely require completely separate sets of bounds. Inheriting the bound from serde also makes this a breaking change,...