Patrick Ohly

/test pull-kubernetes-e2e-gce Some NFS failures (known issue, if I am not mistaken). /skip Cluster creation worked again https://prow.k8s.io/view/gs/kubernetes-ci-logs/pr-logs/pull/135692/pull-kubernetes-local-e2e/1998669135306821632, but there are some know test failures that prevent a fully successful...

> I don't think this is a risk for code freeze but I do think it's less than ideal to have one of our test signals remain broken. So you...

Another idea: perhaps this can be combined with [binding conditions](https://github.com/kubernetes/enhancements/issues/5007) to implement gang scheduling? * We add claim.spec.bindingConditions and friends. * The app controller sets that when creating the claim....

There is also a sanity check in the kubelet which prevents running pods with claims where they are not listed in `ReservedFor`. This is to catch mistakes and already mitigated...

> So do we need to use the ReservedFor list for cleaning up the generated ResourceClaims? My first thought is that the owner references should be sufficient, but there might...

`ReservedFor` does one thing: "keeping track of the pods that are currently using the ResourceClaim." That the same pod sometimes happens to be also the owner of the ResourceClaim is...

> But is there any scenario where the references will be to something else than pods with this definition? Let me clarify: the field tracks "consumers" of the claim, and...

> Handle the kubelet check that prevents pods from running with a claim unless they are listed in ReservedFor. The ideas in https://github.com/kubernetes/enhancements/issues/5194#issuecomment-2715318412 works around this by skipping this step...

/wg device-management /title DRA: MatchExpression This is valid and consistent with the original design. I can see use cases for it which aren't possible with the existing MatchAttributes. Let's discuss...

The entire feature then needs to be renamed. How about `DRA: CEL Constraints` with `DRACELConstraints` as feature gate? Alternative: "DRA: Constraints with CEL", `DRAConstraintsWithCEL` - more verbose, but avoids direct...