[WIP] CNTRLPLANE-371: Update to Kubernetes v1.33#2261
[WIP] CNTRLPLANE-371: Update to Kubernetes v1.33#2261bertinatto wants to merge 2404 commits intoopenshift:masterfrom
Conversation
Signed-off-by: carlory <baofa.fan@daocloud.io>
add device-plugin-test e2e log
Remove general available feature-gate CPUManager
…ularContainerImgChangeE2E Revert "Add e2e test for Regular Container image change"
The defaulting of TimeAdded randomly broke some of the tests:
TestList:
resttest.go:1393: expected:
[]runtime.Object{(*resource.DeviceTaintRule)(0xc000b83080), (*resource.DeviceTaintRule)(0xc000b831e0)},
got:
[]runtime.Object{(*resource.DeviceTaintRule)(0xc0003db608), (*resource.DeviceTaintRule)(0xc0003db750)}
...
TestCreate:
resttest.go:346: unexpected obj: &resource.DeviceTaintRule{TypeMeta:v1.TypeMeta{Kind:"", APIVersion:""}, ObjectMeta:v1.ObjectMeta{Name:"foo2", GenerateName:"", Namespace:"", SelfLink:"", UID:"18d3084d-7d11-4575-8730-4650b81cf1a7", ResourceVersion:"8", Generation:1, CreationTimestamp:time.Date(2025, time.March, 21, 8, 27, 23, 0, time.Local), DeletionTimestamp:<nil>, DeletionGracePeriodSeconds:(*int64)(nil), Labels:map[string]string(nil), Annotations:map[string]string(nil), OwnerReferences:[]v1.OwnerReference(nil), Finalizers:[]string(nil), ManagedFields:[]v1.ManagedFieldsEntry(nil)}, Spec:resource.DeviceTaintRuleSpec{DeviceSelector:(*resource.DeviceTaintSelector)(nil), Taint:resource.DeviceTaint{Key:"example.com/taint", Value:"", Effect:"NoExecute", TimeAdded:time.Date(2025, time.March, 21, 8, 27, 23, 0, time.Local)}}}, expected &resource.DeviceTaintRule{TypeMeta:v1.TypeMeta{Kind:"", APIVersion:""}, ObjectMeta:v1.ObjectMeta{Name:"foo2", GenerateName:"", Namespace:"", SelfLink:"", UID:"18d3084d-7d11-4575-8730-4650b81cf1a7", ResourceVersion:"8", Generation:1, CreationTimestamp:time.Date(2025, time.March, 21, 8, 27, 23, 0, time.Local), DeletionTimestamp:<nil>, DeletionGracePeriodSeconds:(*int64)(nil), Labels:map[string]string(nil), Annotations:map[string]string(nil), OwnerReferences:[]v1.OwnerReference(nil), Finalizers:[]string(nil), ManagedFields:[]v1.ManagedFieldsEntry(nil)}, Spec:resource.DeviceTaintRuleSpec{DeviceSelector:(*resource.DeviceTaintSelector)(nil), Taint:resource.DeviceTaint{Key:"example.com/taint", Value:"", Effect:"NoExecute", TimeAdded:time.Date(2025, time.March, 21, 8, 27, 24, 0, time.Local)}}}
Failure rate before: 3m40s: 1332 runs so far, 7 failures (0.53%)
It's not obvious from the test failure, but the difference is the
TimeAdded. Setting it beforehand to a value that can be encoded (i.e. truncated
to seconds) fixes the flake.
Failure rate after: 5m0s: 1825 runs so far, 0 failures
De-flake infeasible resize test
goos: linux goarch: amd64 pkg: k8s.io/client-go/tools/cache cpu: Intel(R) Xeon(R) CPU @ 2.60GHz BenchmarkLister_Match_1k_100 BenchmarkLister_Match_1k_100-48 41910 28255 ns/op 16384 B/op 1 allocs/op BenchmarkLister_Match_10k_100 BenchmarkLister_Match_10k_100-48 3487 337728 ns/op 163848 B/op 1 allocs/op BenchmarkLister_Match_100k_100 BenchmarkLister_Match_100k_100-48 222 7040793 ns/op 1605659 B/op 1 allocs/op BenchmarkLister_Match_1M_100 BenchmarkLister_Match_1M_100-48 12 97962328 ns/op 16007172 B/op 1 allocs/op PASS ok k8s.io/client-go/tools/cache 10.480s
benchmarks inefficiency on listers linear search lookup
…t-flake DRA API: fix DeviceTaintRule REST storage test
…efs-tests Revert "Separate SeparateDiskTests from eviction"
Run in-place resize tests in default CI jobs
Includes v2beta2 HPA round-trip conversion, defaulting, and validation.
also debug inotify limits before/after setting
…erd-node-e2e CI jobs Signed-off-by: Davanum Srinivas <davanum@gmail.com>
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
The existing patch retried any etcd error returned from storage with the code "Unavailable". Writes can only be safely retried if the client can be absolutely sure that the initial attempt ended before persisting any changes. The "Unavailable" code includes errors like "timed out" that can't be safely retried for writes.
Signed-off-by: Peter Hunt <pehunt@redhat.com> UPSTREAM: <carry>: authorization: add minimumkubeletversion package MinimumKubeletVersion is a way for an admin to declare that nodes any older than the minimum version cannot authorize with the apiserver. This effectively prevents them from joining. Doing so means the apiservers can trust newer features are usable on clusters with version skews Signed-off-by: Peter Hunt <pehunt@redhat.com> UPSTREAM: <carry>: authorizer: move mininum kubelet version authorizer to pkg/kubeapiserver and add authorization mode this does require a line of code be moved from the enablement package to stop a cyclical import Signed-off-by: Peter Hunt <pehunt@redhat.com> UPSTREAM: <carry>: crdvalidation: move latency profile file to be agnostic of field Signed-off-by: Peter Hunt <pehunt@redhat.com> UPSTREAM: <carry>: features: add MinimumKubeletVersion feature Signed-off-by: Peter Hunt <pehunt@redhat.com>
Upstream enables volume group snapshots by editing yaml files in a shell script [1]. We can't use this script in openshift-tests. Create a brand new, OCP specific test driver based on csi-driver-hostpath, only with the --feature-gate=VolumeGroupSnapshot on external-snapshotter command line. We will need to carry this patch until the feature graduates to GA. I've chosen to create brand new files in this carry patch, so it can't conflict with the existing ones. 1: https://github.com/kubernetes/kubernetes/blob/91d6fd3455c4a071408df20c7f48df221f2b6d30/test/e2e/testing-manifests/storage-csi/external-snapshotter/volume-group-snapshots/run_group_snapshot_e2e.sh
The test that checks that a volume can be accessed from multiple nodes should create ReadWriteMany volume and not ReadWriteOnce.
Co-authored-by: Hemant Kumar <gnufied@users.noreply.github.com> Signed-off-by: carlory <baofa.fan@daocloud.io>
Co-authored-by: Allen Ray <alray@redhat.com>
…1 as deprecated." This reverts commit 24bf6d5.
…ates various priority Pods preempt expectedly with the async preemption: replace finalizers with preStop hook and TerminationGracePeriodSeconds
…espaceDeletion test serial
00de7eb to
052606a
Compare
|
/payload 4.20 nightly blocking Previous results were bad because there was a missing patch. |
|
@bertinatto: trigger 11 job(s) of type blocking for the nightly release of OCP 4.20
See details on https://pr-payload-tests.ci.openshift.org/runs/ci/eb6bcc70-3a3a-11f0-924b-9c9399bb924c-0 trigger 4 job(s) of type blocking for the ci release of OCP 4.20
See details on https://pr-payload-tests.ci.openshift.org/runs/ci/eb6bcc70-3a3a-11f0-924b-9c9399bb924c-1 trigger 65 job(s) of type informing for the nightly release of OCP 4.20
See details on https://pr-payload-tests.ci.openshift.org/runs/ci/eb6bcc70-3a3a-11f0-924b-9c9399bb924c-2 |
|
/payload 4.20 nightly blocking |
|
@bertinatto: trigger 11 job(s) of type blocking for the nightly release of OCP 4.20
See details on https://pr-payload-tests.ci.openshift.org/runs/ci/129e1a80-3a88-11f0-9515-89c7c5c11a9c-0 trigger 65 job(s) of type informing for the nightly release of OCP 4.20
See details on https://pr-payload-tests.ci.openshift.org/runs/ci/129e1a80-3a88-11f0-9515-89c7c5c11a9c-1 |
|
This PR served its purpose and can be closed. Payload testing will be done at #2314. /close |
|
PR needs rebase. DetailsInstructions 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. |
|
@bertinatto: Closed this PR. DetailsIn response to this:
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. |
This is a temporary PR create to get the initial steps of the kube bump ready.
This will be closed before starting the payload-testing phase.