Lei Zhang (Harry)
Lei Zhang (Harry)
Btw, for end users they should not use ingress or service at all. So the target solution for this issue should focus on how to let users plug-in their own...
@ryanzhang-oss Please help to review the implementation plan? It generally looks good to me. @xishengcai Please contribute the code to https://github.com/oam-dev/catalog/tree/master/workloads , we will keep this repo clean and provide...
@ryanzhang-oss according to this implementation plan, I believe we could reserve some value like `childResource: dynamic` or simply leave it empty. In either case, oam runtime will then check the...
@xishengcai It's great to see the PoC code is shipped! Let's we split the workload and trait code from runtime and move them to https://github.com/oam-dev/catalog? The plan is we will...
@wonderflow Note that for non-core workload it still need revision labels to be automatically patched (not propagation).
@zzxwill I agree. OAM Runtime should always patch exactly same label to PodTemplate as long as it exist. That's also why I proposed a characteristic named `podspecable` in https://github.com/oam-dev/spec/issues/392 to...
@wonderflow properties in most cases indicates key-value pairs. I would prefer `spec` instead. Also, I prefer `name` than `type` because trait is already a type. That is to say, if...
Hi @mgianatagh , cross-namespace operation is complex and unpredictable in Kubernetes and I am afraid it will create heavy burden on the trait implementation side. Not sure if you envision...
@wonderflow this issue is about whether we want to inject (and how to inject) such information to user brought capabilities (Service, HPA, their own Operators etc), so they don't need...