Garry Fang

Thanks @Alphonse-iwnl, I truely agree with you. We should let karmada do all cleanup jobs. So if ttl-after-finished controller is enabled, we'd better not propagate `job.spec.ttlSecondsAfterFinished`. Since we already find...

We should never delete a object managed by karmada from member clusters. From this point of view, now I think there is no need to propagate `TTLSecondsAfterFinished` no matter if...

IMO, I prefer not propagating this filed.

> In which release? 1.23, https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.23.md#horizontalpodautoscaler-v2-graduates-to-ga

/assign

> Hi @Garrybest I guess the task(`Remove cluster scheduling result in spec.clusters when the cluster is terminating.`) is included in #1945. > > Should we update the task list? Well,...

I have an idea. Maybe we could implement this in cmd `karmadactl drain`. Just like what `kubectl drain` does, when we want to unjoin a cluster, we cordon this cluster...

Well, I prefer implicit priority, it's easier for users. And I think if there are a lot pps, maybe it's a burden for users to care about priority of every...

> I suppose the implicit priority and explicit priority don't conflict with each other. The explicit priority will have a default priority, but with a higher priority than the implicit...

In my scenario, I create a default pp for users to select all deployments in a namespace called `common-pp`. ``` resourceSelectors: - apiVersion: apps/v1 kind: Deployment namespace: ns1 ``` So...