DRA: Update resourceslice controller filtering logic

[klueska]

The logic has been updated to ensure that a controller started for non-node-local resources filters out all resourceslices created for node-local resources. Without this change, a single driver with both node-local and non-node-local resources would end up in a constant battle of creating and deleting node-local resource slices in the controller it setup for its non-node-local resources. This change fixes that.

/kind bug
/wg device-management

Does this PR introduce a user-facing change?

NONE

[klueska]

/cc @pohly

[klueska]

/hold
Since I have approval status

[pohly]

/wg device-management

[pohly]

"gofmt" failed.

[bart0sh]

/king bug
/triage accepted
/priority important-soon

[klueska]

@pohly do you know anything about the erros in the dra tests:

+ kubetest2 noop --test=node -- --repo-root=. --gcp-zone=us-west1-b --parallelism=1 '--label-filter="Feature: containsAny DynamicResourceAllocation && Feature: isSubsetOf DynamicResourceAllocation && !Flaky && !Slow"' '--test-args=--feature-gates="DynamicResourceAllocation=true" --service-feature-gates="DynamicResourceAllocation=true" --runtime-config=api/alpha=true,api/beta=true --container-runtime-endpoint=unix:///var/run/crio/crio.sock --container-runtime-process-name=/usr/local/bin/crio --container-runtime-pid-file= --kubelet-flags="--cgroup-driver=systemd --cgroups-per-qos=true --cgroup-root=/ --runtime-cgroups=/system.slice/crio.service --kubelet-cgroups=/system.slice/kubelet.service" --extra-log="{\"name\": \"crio.log\", \"journalctl\": [\"-u\", \"crio\"]}"' --image-config-file=/home/prow/go/src/k8s.io/test-infra/jobs/e2e_node/crio/latest/image-config-cgroupv1-serial.yaml
Error: unknown flag: --label-filter

[pohly]

It looks like --label-filter is being passed to kubetest2? That's wrong, it must be set as test flags that kubetest2 then passes on to Ginkgo.

[pohly]

See kubernetes/test-infra#33550

[pohly]

kubernetes/test-infra#32567 (comment)

[bart0sh]

@klueska I believe it's safe to ignore -kubetest2 failures for now as it's ongoing work to move jobs to use kubetest2. Here is a tracking issue: kubernetes/test-infra#32567

[klueska]

I will wait until that fix for --label-filter gets merged. I want to make sure that the DRA tests don't break with the change I introduced. I can't imagine they would, but this change isn't urgent anyway.

[pohly]

We are still testing with the non-kubetest2 jobs (pull-kubernetes-node-e2e-crio-cgrpv1-dra , pull-kubernetes-node-e2e-crio-cgrpv2-dra), which are green.

[pohly]

/lgtm
/approve

[k8s-ci-robot]

LGTM label has been added.

Git tree hash: 8ecc9498f1818dbe38735587268501303f432829

[pohly]

/hold cancel

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: klueska, pohly

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• staging/src/k8s.io/dynamic-resource-allocation/OWNERS [klueska,pohly]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[pacoxu]

/kind bug
/skip

[k8s-ci-robot]

@klueska: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
pull-kubernetes-node-e2e-crio-cgrpv1-dra-kubetest2	cfd6037	link	false	/test pull-kubernetes-node-e2e-crio-cgrpv1-dra-kubetest2
pull-kubernetes-node-e2e-crio-cgrpv2-dra-kubetest2	cfd6037	link	false	/test pull-kubernetes-node-e2e-crio-cgrpv2-dra-kubetest2

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.

[pacoxu]

/test pull-kubernetes-unit
flake