issue_owner_repo listlengths 2 2 | issue_body stringlengths 0 261k ⌀ | issue_title stringlengths 1 925 | issue_comments_url stringlengths 56 81 | issue_comments_count int64 0 2.5k | issue_created_at stringlengths 20 20 | issue_updated_at stringlengths 20 20 | issue_html_url stringlengths 37 62 | issue_github_id int64 387k 2.91B | issue_number int64 1 131k |
|---|---|---|---|---|---|---|---|---|---|
[
"kubernetes",
"kubernetes"
] | ### What happened?
Node has 64 vCPUs. CPU manager policy is 'static'
8 vCPUs are marked as resreved (via --reserved-cpus)
The remaining 56 vCPUs are all isolated via kernel command line.
systemd.cpu_affinity is set for all of the --reserved-cpus also (I think this is redundant with isolated CPU command line tho... | containers get scheduled on --reserved-cpus | https://api.github.com/repos/kubernetes/kubernetes/issues/115994/comments | 49 | 2023-02-23T06:09:44Z | 2025-01-07T16:41:47Z | https://github.com/kubernetes/kubernetes/issues/115994 | 1,596,278,268 | 115,994 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Getting log messages like:
```
kube-apiserver-master-k8s-mczx.c.uw-dev.internal kube-apiserver Trace[2027316592]: ---"About to Encode" 521ms (05:35:39.110)
kube-apiserver-master-k8s-mczx.c.uw-dev.internal kube-apiserver Trace[2027316592]: [535.208261ms] [535.208261ms] END
kube-apiserver-master... | Trace messages from API Server with log verbosity of 0 | https://api.github.com/repos/kubernetes/kubernetes/issues/115993/comments | 10 | 2023-02-23T05:45:09Z | 2024-04-23T18:34:02Z | https://github.com/kubernetes/kubernetes/issues/115993 | 1,596,261,001 | 115,993 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
# Some Context
But only after the Pod call is completed, we know the node where the Pod is running, and then we can know the availability zone information.
At this time, it is no longer possible to define environment variables in the spec template.
A tricky way is to add the en... | Add NodeFieldSelector to EnvVarSource to add environment generated on the kubelet | https://api.github.com/repos/kubernetes/kubernetes/issues/115985/comments | 7 | 2023-02-23T02:47:41Z | 2024-01-19T00:59:57Z | https://github.com/kubernetes/kubernetes/issues/115985 | 1,596,127,228 | 115,985 |
[
"kubernetes",
"kubernetes"
] | ### Failure cluster [1bf2be60ce0078971afe](https://go.k8s.io/triage#1bf2be60ce0078971afe)
##### Error text:
```
Failed
=== RUN TestLegacyFallbackNoCache
I0222 09:04:49.496617 108495 handler_discovery.go:392] Starting ResourceDiscoveryManager
I0222 09:04:49.498038 108495 handler.go:149] Adding GroupVersion s... | Failure cluster [1bf2be60...] TestLegacyFallbackNoCache flakes for order | https://api.github.com/repos/kubernetes/kubernetes/issues/115983/comments | 3 | 2023-02-23T02:30:11Z | 2023-02-23T15:59:08Z | https://github.com/kubernetes/kubernetes/issues/115983 | 1,596,117,008 | 115,983 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
```
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx
labels:
app: nginx
spec:
replicas: 1
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
affinity:
nodeAffinity:
required... | Required value: must be only one value when `operator` is 'In' or 'NotIn' for node field selector | https://api.github.com/repos/kubernetes/kubernetes/issues/115980/comments | 4 | 2023-02-23T00:39:42Z | 2023-02-24T02:47:32Z | https://github.com/kubernetes/kubernetes/issues/115980 | 1,596,044,224 | 115,980 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
When running the e2e tests on AWS, the following two tests are failing:
```
[sig-network] LoadBalancers should be able to preserve UDP traffic when server pod cycles for a LoadBalancer service on different nodes [Suite:openshift/conformance/parallel] [Suite:k8s]
```
```
[sig-network] LoadBala... | e2e loadbalancer udp conntrack test not working properly on AWS variant | https://api.github.com/repos/kubernetes/kubernetes/issues/115975/comments | 7 | 2023-02-22T21:34:13Z | 2023-02-27T18:26:53Z | https://github.com/kubernetes/kubernetes/issues/115975 | 1,595,867,185 | 115,975 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
We are running Kubernetes with an OpenID connector.
our API server is configured with 2 flags in this regard:
```console
oidc-client-id: ourKubernetes
oidc-issuer-url: https://our-openid-connect.tld
```
In our issuer, we have the `groups` key within the claims similar to: `{"s... | OIDC groups are not handles by default | https://api.github.com/repos/kubernetes/kubernetes/issues/115954/comments | 3 | 2023-02-22T10:56:51Z | 2023-02-22T14:53:49Z | https://github.com/kubernetes/kubernetes/issues/115954 | 1,594,891,871 | 115,954 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
When deploying an application, the version iterates several times over time and its cinfigmap file.
In the process of running, when time jump back,the deployment version fallback occasionally.
However, the specific trigger point has not been known, and it can not recur stably.
### What did yo... | Deployment version roll back when os time happen time jump. | https://api.github.com/repos/kubernetes/kubernetes/issues/115949/comments | 6 | 2023-02-22T08:51:37Z | 2024-01-18T23:59:57Z | https://github.com/kubernetes/kubernetes/issues/115949 | 1,594,702,141 | 115,949 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
Using kubespray to upgrade 1.20.15 to 1.21.14 had no problems, but failed to upgrade from 1.21.14 to 1.22.16. The key commands are as follows:
kubeadm upgrade apply -y v1.22.16 --certificate-renewal=True --config=/etc/kubernetes/kubeadm-config.yaml --ignore-preflight-errors=all --allow-exp... | “kube-apiserver” cannot be pulled up by kubelet | https://api.github.com/repos/kubernetes/kubernetes/issues/115948/comments | 6 | 2023-02-22T08:26:31Z | 2024-01-18T23:59:57Z | https://github.com/kubernetes/kubernetes/issues/115948 | 1,594,669,110 | 115,948 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
I use a configMap to mount at container , when i update the configMap , the container volume takes 20s or much longer to catch up the update .
And very strange ,if i update configMap again, the previous change will immediately show at the container volume.
### What did you expect to happen?
th... | ConfigMap Hot Update Delay | https://api.github.com/repos/kubernetes/kubernetes/issues/115946/comments | 3 | 2023-02-22T06:56:55Z | 2023-02-26T03:13:57Z | https://github.com/kubernetes/kubernetes/issues/115946 | 1,594,561,633 | 115,946 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are flaking?
master-informing
- gce-cos-master-serial
### Which tests are flaking?
Kubernetes e2e suite.[It] [sig-storage] PersistentVolumes-local Stress with local volumes [Serial] should be able to process many pods and reuse local volumes
### Since when has it been flaking?
2023-02-15 15:12:38... | [Flaking Test] gce-cos-master-serial | https://api.github.com/repos/kubernetes/kubernetes/issues/115945/comments | 18 | 2023-02-22T05:32:30Z | 2023-06-07T15:48:26Z | https://github.com/kubernetes/kubernetes/issues/115945 | 1,594,487,010 | 115,945 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
headless services without selectors can be used to access external resources. Example:
```
apiVersion: v1
kind: Service
metadata:
name: yb-masters
spec:
clusterIP: None
ports:
- name: tcp-rpc-port
port: 7100
protocol: TCP
targetPort: 7100
---
api... | control pod name on headless without selector | https://api.github.com/repos/kubernetes/kubernetes/issues/115941/comments | 3 | 2023-02-22T01:14:25Z | 2023-02-22T03:05:13Z | https://github.com/kubernetes/kubernetes/issues/115941 | 1,594,300,141 | 115,941 |
[
"kubernetes",
"kubernetes"
] | Today we construct the request object for every http request for the http probes. This involves concatenating some strings, setting headers, etc.
Probes are simple http get requests that are executed synchronously (for a single probe) and has no request body.
The optimization may be to reuse the request object fo... | Reuse the http request object for http probes | https://api.github.com/repos/kubernetes/kubernetes/issues/115939/comments | 27 | 2023-02-21T23:07:09Z | 2025-03-06T19:38:07Z | https://github.com/kubernetes/kubernetes/issues/115939 | 1,594,211,960 | 115,939 |
[
"kubernetes",
"kubernetes"
] | - This is part of pre-work for sidecar KEP: https://github.com/kubernetes/enhancements/issues/753
- See also uber issue: https://github.com/kubernetes/kubernetes/issues/115934
In sidecar container KEP we plan to use the Startup probes state as a way to decide whether the next Init container need to start being init... | Refactor prober_manager to update the runtime state vs. API state | https://api.github.com/repos/kubernetes/kubernetes/issues/115937/comments | 15 | 2023-02-21T22:05:44Z | 2025-02-07T07:05:02Z | https://github.com/kubernetes/kubernetes/issues/115937 | 1,594,157,127 | 115,937 |
[
"kubernetes",
"kubernetes"
] | KEP: https://github.com/kubernetes/enhancements/issues/753
This issue is to discuss and approve the implementation plan to minimize the amount of rebase needed. Issue was discussed at [working group meeting](https://docs.google.com/document/d/1E1guvFJ5KBQIGcjCrQqFywU9_cBQHRtHvjuqcVbCXvU/edit#bookmark=id.x7ms8ybmhtfv... | Sidecar containers implementation steps: uber issue | https://api.github.com/repos/kubernetes/kubernetes/issues/115934/comments | 9 | 2023-02-21T21:06:59Z | 2023-11-01T01:15:33Z | https://github.com/kubernetes/kubernetes/issues/115934 | 1,594,100,455 | 115,934 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are flaking?
ci-kubernetes-e2e-capz-master-containerd-windows
### Which tests are flaking?
Kubernetes e2e suite: [It] [sig-windows] [Feature:WindowsHostProcessContainers] [MinimumKubeletVersion:1.22] HostProcess containers container stats validation
### Since when has it been flaking?
feb... | [Flaky] HostProcess containers container stats validation | https://api.github.com/repos/kubernetes/kubernetes/issues/115932/comments | 2 | 2023-02-21T18:15:12Z | 2023-03-01T23:11:29Z | https://github.com/kubernetes/kubernetes/issues/115932 | 1,593,909,810 | 115,932 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
According to the [docs](https://kubernetes.io/docs/reference/command-line-tools-reference/kube-proxy/#:~:text=will%20be%20masqueraded%20and-,traffic%20sent%20from%20pods%20to%20an%20external%20LoadBalancer%20IP%20will%20be%20directed%20to%20the%20respective%20cluster%20IP%20instead.,-For%20dual)... | kube-proxy IPVS mode: traffic sent from pods to an external LoadBalancer IP is NOT directed to the respective cluster IP instead | https://api.github.com/repos/kubernetes/kubernetes/issues/115931/comments | 12 | 2023-02-21T17:50:16Z | 2023-04-07T07:06:34Z | https://github.com/kubernetes/kubernetes/issues/115931 | 1,593,882,465 | 115,931 |
[
"kubernetes",
"kubernetes"
] | ```
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
┃ [CVE-2021-3636] CWE-287: Improper Authentication ... | kubernetes@v1.26.1 has vulnerability - CVE-2021-3636 | https://api.github.com/repos/kubernetes/kubernetes/issues/115921/comments | 3 | 2023-02-21T08:22:06Z | 2023-02-21T14:04:45Z | https://github.com/kubernetes/kubernetes/issues/115921 | 1,593,032,243 | 115,921 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
If we have defined several validations that are violated, only one validation result seems to be displayed.
### What did you expect to happen?
All results of validation that have been violated will be displayed.
### How can we reproduce it (as minimally and precisely as possible)?
Defi... | CEL ValidatingAdmissionPolicy only one validation result is displayed | https://api.github.com/repos/kubernetes/kubernetes/issues/115920/comments | 18 | 2023-02-21T06:17:59Z | 2025-02-06T21:24:06Z | https://github.com/kubernetes/kubernetes/issues/115920 | 1,592,880,382 | 115,920 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
When using external dependencies inside healthchecks or using fixed thread pool applications like python+gunicorn, ruby+passenger, PHP etc,... upstream service failures tend to cause cascading impact on the service as well.
one solution is removing the upstream services insid... | Panic Threshold for Readiness Probe Failures | https://api.github.com/repos/kubernetes/kubernetes/issues/115917/comments | 20 | 2023-02-21T04:05:21Z | 2023-03-30T16:56:22Z | https://github.com/kubernetes/kubernetes/issues/115917 | 1,592,743,752 | 115,917 |
[
"kubernetes",
"kubernetes"
] |
There is a known limitation in this feature that when pods scale down there topology spread may not be even. How does one overcome this limitation to keep there pods spread across multiple availability zones. This can be a huge issue for apps that need to be highly available.
As per documentation:
“There’s no... | Topology spread constraint not respected when pods scale down | https://api.github.com/repos/kubernetes/kubernetes/issues/115911/comments | 20 | 2023-02-20T15:16:17Z | 2024-03-29T10:47:08Z | https://github.com/kubernetes/kubernetes/issues/115911 | 1,592,048,693 | 115,911 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The dynamic resource created by kubelet DRA plugin is not cleaned because NodeUnprepareResource is not called.
### What did you expect to happen?
kubelet DRA plugin will free the allocated dynamic resource
### How can we reproduce it (as minimally and precisely as possible)?
1. create a POD w... | kubelet DRA plugin not cleaning resources if kubelet restarted | https://api.github.com/repos/kubernetes/kubernetes/issues/115902/comments | 2 | 2023-02-20T10:51:51Z | 2023-03-12T19:20:41Z | https://github.com/kubernetes/kubernetes/issues/115902 | 1,591,603,784 | 115,902 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The PR [1] introduced ``PluginExistsWithCorrectUUID`` as an alternative method to ``PluginExistsWithCorrectTimestamp``, deprecating the latter with the intention of removing in a few releases.
This issue is created to track its deprecation and removal.
[1] https://github.com/kubernetes/kuber... | TODO: Remove deprecated PluginExistsWithCorrectTimestamp | https://api.github.com/repos/kubernetes/kubernetes/issues/115897/comments | 10 | 2023-02-20T08:52:32Z | 2024-03-22T13:28:01Z | https://github.com/kubernetes/kubernetes/issues/115897 | 1,591,409,871 | 115,897 |
[
"kubernetes",
"kubernetes"
] | An error as follow:
```
# kubectl expose deployment -n kube-system virtual-kubelet --port=10250 --type=LoadBalancer
The Service "virtual-kubelet" is invalid: spec.ports[0]: Invalid value: 10250: may not expose port 10250 externally since it is used by kubelet
#
```
People who see it for the first time may feel c... | Is it still necessary to maintain the limitation on exposing port 10250 externally? | https://api.github.com/repos/kubernetes/kubernetes/issues/115896/comments | 42 | 2023-02-20T08:33:43Z | 2024-12-14T06:27:46Z | https://github.com/kubernetes/kubernetes/issues/115896 | 1,591,388,054 | 115,896 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
periodic-conformance-main-k8s-main
### Which tests are failing?
capa-e2e-conformance: [SynchronizedBeforeSuite]
### Since when has it been failing?
since 2/18 09:25 PST
### Testgrid link
https://k8s-testgrid.appspot.com/sig-release-master-informing#periodic-conformance-main-k8s-main
... | [Failing Test] periodic-conformance-main-k8s-main | https://api.github.com/repos/kubernetes/kubernetes/issues/115894/comments | 7 | 2023-02-20T01:28:01Z | 2023-02-25T03:11:56Z | https://github.com/kubernetes/kubernetes/issues/115894 | 1,590,959,014 | 115,894 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
The go-jose dependency version should be updated from v2.2.2 to v2.6.0.
### Why is this needed?
This is to make the kubernetes code compatible with newer go-jose v2 versions that have a small breaking change: `jwt.NewNumericDate()` returns a pointer in newer versions which ... | Update go-jose version | https://api.github.com/repos/kubernetes/kubernetes/issues/115892/comments | 2 | 2023-02-19T23:33:41Z | 2023-03-02T04:23:08Z | https://github.com/kubernetes/kubernetes/issues/115892 | 1,590,906,299 | 115,892 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
`golangci-lint` reports a number of unused function, field and variable errors. Shall we fix them?
```
pkg/util/iptables/iptables_unsupported.go:31:6: func `grabIptablesFileLock` is unused (unused)
func grabIptablesFileLock(f *os.File) error {
^
pkg/volume/util/hostutil/hostutil.go:81:... | Fix golangci-lint errors | https://api.github.com/repos/kubernetes/kubernetes/issues/115874/comments | 8 | 2023-02-19T02:34:10Z | 2023-02-23T07:59:22Z | https://github.com/kubernetes/kubernetes/issues/115874 | 1,590,535,756 | 115,874 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
Currently, when the Label Selector in the Pod Affinity is empty, no objects will be match.
That means if it's in a required PodAffinity, the Pod will never be schedulable. And if it's in a required PodAntiAffinity, that PodAntiAffinity will be no-op.
So, I'm thinking PodAffi... | add the warning in the Pod creation with empty label selector in the pod affinity and topology spread | https://api.github.com/repos/kubernetes/kubernetes/issues/115873/comments | 8 | 2023-02-18T16:06:46Z | 2023-04-13T23:00:41Z | https://github.com/kubernetes/kubernetes/issues/115873 | 1,590,388,523 | 115,873 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
root@LAPTOP-6IH0BQSV:/home/jiacheng/kubernetes# hack/make-rules/../../hack/verify-conformance-requirements.sh
go version go1.20 linux/amd64
hack/conformance/check_conformance_test_requirements.go:29:2: cannot find package "github.com/spf13/cobra" in any of:
/home/jiacheng/kuberne... | cannot find package "github.com/spf13/cobra" when make verify | https://api.github.com/repos/kubernetes/kubernetes/issues/115866/comments | 1 | 2023-02-18T04:21:31Z | 2023-02-22T15:17:58Z | https://github.com/kubernetes/kubernetes/issues/115866 | 1,590,200,051 | 115,866 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Running e2e tests with vSphere cloud panics in `vspheretest.GetReadySchedulableNodeInfos`: https://github.com/kubernetes/kubernetes/blob/1c97dbc8c9096495706b1e141b0fe256ff3185e9/test/e2e/storage/vsphere/vsphere_utils.go#L755-L756 when it sues global variable `f`: https://github.com/kubernetes/kubern... | vSphere e2e tests panic in cleanup | https://api.github.com/repos/kubernetes/kubernetes/issues/115862/comments | 1 | 2023-02-17T19:18:33Z | 2023-02-20T19:09:50Z | https://github.com/kubernetes/kubernetes/issues/115862 | 1,589,850,364 | 115,862 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are flaking?
pull-kubernetes-unit sometimes fails with data race.
https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/pull/115585/pull-kubernetes-unit/1622986339706212352
```sh
~/go/src/k8s.io/kubernetes/staging/src/k8s.io/kube-aggregator/pkg/apiserver > go test -c -race ./
~/go/src/k8s.io/kube... | [Flaking test] pull-kubernetes-unit - TestDirty fails sometimes | https://api.github.com/repos/kubernetes/kubernetes/issues/115858/comments | 2 | 2023-02-17T14:04:24Z | 2023-03-01T18:09:21Z | https://github.com/kubernetes/kubernetes/issues/115858 | 1,589,409,557 | 115,858 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
[MCS](https://cloud.google.com/kubernetes-engine/docs/concepts/multi-cluster-services) mechanism to communicate multiple GKE clusters stopped work in all our clusters since 09.02.2023. We noticed that MCS service account (`service-PROJ_NUMBER@gcp-sa-mcsd.iam.gserviceaccount.com`) deleted the privat... | GKE - Multi cluster services stopped work after DNS zone was removed | https://api.github.com/repos/kubernetes/kubernetes/issues/115857/comments | 4 | 2023-02-17T13:18:04Z | 2023-02-17T16:11:21Z | https://github.com/kubernetes/kubernetes/issues/115857 | 1,589,346,099 | 115,857 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
start of a simple pod with just an emptydir volume needs between 2 and 5 minutes. the same issue is reproducible with basically every volume type (secret, configmap, PVC,..) but during troubleshooting, we used just an eptydir volume
following log output we found during troubleshooting that the ... | emtpydir volume slow mount | https://api.github.com/repos/kubernetes/kubernetes/issues/115853/comments | 11 | 2023-02-17T08:51:27Z | 2023-05-02T07:39:10Z | https://github.com/kubernetes/kubernetes/issues/115853 | 1,588,974,136 | 115,853 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are flaking?
release-master-informing
ci-crio-cgroupv1-node-e2e-conformance
### Which tests are flaking?
Overall
kubetest.Prepare
kubetest.Node tests
### Since when has it been flaking?
2/15
### Testgrid link
https://k8s-testgrid.appspot.com/sig-release-master-informing#ci-crio-cgroupv1... | [Flaking Test] ci-crio-cgroupv1-node-e2e-conformance | https://api.github.com/repos/kubernetes/kubernetes/issues/115849/comments | 6 | 2023-02-17T02:44:08Z | 2023-02-20T02:33:00Z | https://github.com/kubernetes/kubernetes/issues/115849 | 1,588,667,462 | 115,849 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The same ResourceRequirements struct is used in two only loosely related entities (Container and PersistentVolumeClaim). Extending ResourceRequirements for dynamic resource requirement had the unintended effect that the new `Claims` field also became available in a PersistentVolumeClaim.
##... | API: ResourceRequirements usage in Container and PVC | https://api.github.com/repos/kubernetes/kubernetes/issues/115845/comments | 2 | 2023-02-16T19:21:27Z | 2023-08-21T21:08:07Z | https://github.com/kubernetes/kubernetes/issues/115845 | 1,588,238,730 | 115,845 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
When a Pod belonging to a Job is deleted, the job controller creates a replacement pod immediately, even if parallelism is set to 1 and backoffLimit is set to 0.
### What did you expect to happen?
This depends on the use case, but some applications fail when there are more than one pod.
F... | Jobs create replacement Pods as soon as a Pod is marked for deletion | https://api.github.com/repos/kubernetes/kubernetes/issues/115844/comments | 56 | 2023-02-16T18:07:30Z | 2023-07-21T13:46:11Z | https://github.com/kubernetes/kubernetes/issues/115844 | 1,588,143,328 | 115,844 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
KEP https://github.com/kubernetes/enhancements/pull/3837 outlined the fact that Kube-proxy today doesn't perform any validation of its `healthz` state when answering to load balancer HCs for eTP:Local services.
This is a potential problem, because if Kube-proxy is unhealthy then it might also not... | Kube-proxy does not validate `healthz` state when answering to HC for eTP:Local | https://api.github.com/repos/kubernetes/kubernetes/issues/115825/comments | 10 | 2023-02-16T10:25:36Z | 2024-09-18T00:00:46Z | https://github.com/kubernetes/kubernetes/issues/115825 | 1,587,420,005 | 115,825 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
My newly installed kubeadm failed when executing `kubeadm init --config kubeadm-config.yaml`, It reports that kubelet isn't healthy. So, I did the following steps:
`root@localhost:~# systemctl status kubelet`
```
● kubelet.service - kubelet: The Kubernetes Node Agent
Loaded: loaded (/etc/... | kubelet.service: Main process exited, code=exited, status=1/FAILURE | https://api.github.com/repos/kubernetes/kubernetes/issues/115824/comments | 12 | 2023-02-16T09:57:23Z | 2023-02-21T02:04:51Z | https://github.com/kubernetes/kubernetes/issues/115824 | 1,587,375,929 | 115,824 |
[
"kubernetes",
"kubernetes"
] | Probes of all kinds currently support FailureThreshold (and SuccessThreshold), these properties allow a user to specify that Kubernetes should not take action in response to a failed probe unless it fails a successive number of times.
This is useful for end-users as it allows them to mitigate the effects of any pro... | Give an indication in container events for probe failure as to whether the failure was ignored due to FailureThreshold | https://api.github.com/repos/kubernetes/kubernetes/issues/115823/comments | 14 | 2023-02-16T09:02:13Z | 2024-10-04T10:55:58Z | https://github.com/kubernetes/kubernetes/issues/115823 | 1,587,295,074 | 115,823 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Having successfully deleted the pods using kubectl command in the k8s cluster(1.24.3), the containers of containerd(1.6.9) are still in the Running status, leading to the netns and the ip resources of them cannot be released
![微信图片_20230216152532](https://user-images.githubusercontent.com/28282... | Having successfully deleted the pods using kubectl command in the k8s cluster, the containers of containerd are still in the Running status, leading to the netns and the ip resources of them cannot be released | https://api.github.com/repos/kubernetes/kubernetes/issues/115822/comments | 10 | 2023-02-16T07:32:22Z | 2024-01-19T17:00:55Z | https://github.com/kubernetes/kubernetes/issues/115822 | 1,587,177,838 | 115,822 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
Running update-codegen
go version go1.20 linux/amd64
go: downloading github.com/spf13/pflag v1.0.5
go: downloading k8s.io/klog/v2 v2.80.1
go: downloading k8s.io/gengo v0.0.0-20220902162205-c0856e24416d
go: downloading k8s.io/kube-openapi v0.0.0-20230123231816-1cb3ae25d79a
cmd/defaulte... | Running update-codegen FAILED | https://api.github.com/repos/kubernetes/kubernetes/issues/115820/comments | 4 | 2023-02-16T05:52:01Z | 2023-02-25T08:12:04Z | https://github.com/kubernetes/kubernetes/issues/115820 | 1,587,064,453 | 115,820 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
When pods are evicted, i.e. with a hard eviction, the grace period is intended to be `0s` (i.e. do not respect the pod's `terminationGracePeriodSeconds`).
However, currently, evicted pods receive their full termination grace period seconds.
### What did you expect to happen?
The pod is e... | kubelet: Evicted and force deleted pods get their full termination grace period when they should not | https://api.github.com/repos/kubernetes/kubernetes/issues/115819/comments | 13 | 2023-02-16T03:21:59Z | 2024-05-15T23:14:14Z | https://github.com/kubernetes/kubernetes/issues/115819 | 1,586,937,584 | 115,819 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Termination grace period includes the termination time of prestop hooks and containers in the pod. The total time that the pod terminates should always be < termination grace period seconds.
A misconfigured pod with a preStopHook that hangs indefinitely and pod ignoring SIGTERM can result in th... | kubelet: PreStop Hooks can hang termination of a pod beyond pod's termination grace period | https://api.github.com/repos/kubernetes/kubernetes/issues/115817/comments | 5 | 2023-02-16T00:43:00Z | 2023-06-14T17:34:20Z | https://github.com/kubernetes/kubernetes/issues/115817 | 1,586,810,321 | 115,817 |
[
"kubernetes",
"kubernetes"
] | Ref: #115595
/assign | [KMS v2] Setup build and push for reference implementation provider | https://api.github.com/repos/kubernetes/kubernetes/issues/115812/comments | 1 | 2023-02-15T21:54:12Z | 2023-03-01T01:57:19Z | https://github.com/kubernetes/kubernetes/issues/115812 | 1,586,619,790 | 115,812 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
This is being proposed as change from https://github.com/kubernetes/enhancements/issues/2119
From` discussion among sig-storage, we agreed to make following fields inside `CSIDriver` mutable:
* FSGroupPolicy: it’s safe to be mutable
If two pods use the same volume, one of ... | Make specific fields inside CSIDriver.Spec mutable | https://api.github.com/repos/kubernetes/kubernetes/issues/115810/comments | 5 | 2023-02-15T19:52:56Z | 2023-12-13T20:24:59Z | https://github.com/kubernetes/kubernetes/issues/115810 | 1,586,461,659 | 115,810 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
As part of the apiserver's pre-shutdown sequence, we stop the lease-controller from renewing the apiserver's etcd lease anymore and delete the lease:
https://github.com/kubernetes/kubernetes/blob/125e38c0872521bec49351415cc9fb624b046659/pkg/controlplane/controller.go#L211-L215
As part of the... | Last apiserver to shutdown doesn't remove its IP from k8s svc endpoints | https://api.github.com/repos/kubernetes/kubernetes/issues/115804/comments | 24 | 2023-02-15T17:52:24Z | 2023-06-20T21:22:01Z | https://github.com/kubernetes/kubernetes/issues/115804 | 1,586,294,212 | 115,804 |
[
"kubernetes",
"kubernetes"
] | This is a feature request to add support for informers to filter objects for local cache. The motivation is to reduce the local cache size, which is memory overhead, for shared informers.
For example, a binary might initialize a shared informer for multiple controllers but some controllers only want to watch objects... | [FR] Add support in informers to filter objects for local cache | https://api.github.com/repos/kubernetes/kubernetes/issues/115803/comments | 10 | 2023-02-15T17:06:00Z | 2024-05-16T16:43:15Z | https://github.com/kubernetes/kubernetes/issues/115803 | 1,586,262,900 | 115,803 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Initially issue was observed using downstream version of ovn-kubernetes deployment. ovn-kubernetes central components include ovn database pod (ovnkube-db) along with a headless service and endpoints created by this pod. Database pod contains two databases - northbound and southbound, hence endpoint... | Manually created Endpoints for a headless Service are deleted when kube-controller-manager Pod restarted | https://api.github.com/repos/kubernetes/kubernetes/issues/115801/comments | 8 | 2023-02-15T16:50:21Z | 2023-07-16T13:24:38Z | https://github.com/kubernetes/kubernetes/issues/115801 | 1,586,201,458 | 115,801 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Hi kubernetes group, I so sorry to bothe yours.
My code include `k8s.io/component-base/logs` package, at compile ouccr this error
## my code is
```go
logs.InitLogs()
defer logs.FlushLogs()
```
## error phenomenon
```
../../pkg/mod/k8s.io/component-base@v0.19.11/logs/json/jso... | about component-base/logs package error at compile | https://api.github.com/repos/kubernetes/kubernetes/issues/115795/comments | 3 | 2023-02-15T12:23:22Z | 2023-02-16T02:50:45Z | https://github.com/kubernetes/kubernetes/issues/115795 | 1,585,767,329 | 115,795 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
when I deploy k8s, it occurs the follow errors.
journalctl -xeu kubelet
Feb 15 18:20:29 master kubelet[613371]: Flag --logtostderr has been deprecated, will be removed in a future release, see https://github.com/kubernetes/enhancements/tree/master/keps/sig-instrumentation/2845-deprecate-klog-spe... | kubelet report the error, and cannot create kata pod | https://api.github.com/repos/kubernetes/kubernetes/issues/115793/comments | 3 | 2023-02-15T10:37:58Z | 2023-02-16T01:56:27Z | https://github.com/kubernetes/kubernetes/issues/115793 | 1,585,618,524 | 115,793 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
master-informing
ci-kubernetes-e2e-gce-scale-performance.Pod
### Which tests are failing?
ci-kubernetes-e2e-gce-scale-performance.Pod
<img width="997" alt="image" src="https://user-images.githubusercontent.com/42276283/218947000-07688fd9-7ce0-4ef6-89e5-973252229077.png">
### Since wh... | [Failing test] ci-kubernetes-e2e-gce-scale-performance.Pod | https://api.github.com/repos/kubernetes/kubernetes/issues/115792/comments | 6 | 2023-02-15T06:38:56Z | 2023-02-15T22:15:00Z | https://github.com/kubernetes/kubernetes/issues/115792 | 1,585,305,999 | 115,792 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
Add a `responseResourceVersion` field to audit log events (`k8s.io/apiserver/pkg/apis/audit/v1.Event`). If the audit event refers to a successful resource update resulting in a change in resource version, it should contain the resource version of the response object.
This is equ... | Add `responseResourceVersion` to audit log events | https://api.github.com/repos/kubernetes/kubernetes/issues/115791/comments | 16 | 2023-02-15T06:12:34Z | 2024-12-20T00:02:03Z | https://github.com/kubernetes/kubernetes/issues/115791 | 1,585,283,552 | 115,791 |
[
"kubernetes",
"kubernetes"
] | I am tracking a changed behavior in networking with kubernetes 1.26. If I create a service of type externalTrafficPolicy: Local, and there are no local endpoints in a node, kube-proxy does not create hns loadbalancers for the clusterIP of that service.
With kubernetes 1.25, it creates a loadbalancer for clusterIP (... | windows kube-proxy: "externalTrafficPolicy: Local" results in no clusterIP entry in windows node | https://api.github.com/repos/kubernetes/kubernetes/issues/115783/comments | 7 | 2023-02-15T00:39:39Z | 2023-08-03T15:40:03Z | https://github.com/kubernetes/kubernetes/issues/115783 | 1,585,030,363 | 115,783 |
[
"kubernetes",
"kubernetes"
] | Write a new e2e test that will be marked as `[Serial]` and will be covering this bug fix: https://github.com/kubernetes/kubernetes/issues/89898
The test will serve as a replacement for the skipped unit test: https://github.com/kubernetes/kubernetes/pull/115329
The logic of the test will be as following:
- Crea... | Write the stress test for gRPC, http, and tcp probes | https://api.github.com/repos/kubernetes/kubernetes/issues/115782/comments | 21 | 2023-02-15T00:07:14Z | 2025-02-01T23:22:46Z | https://github.com/kubernetes/kubernetes/issues/115782 | 1,585,009,624 | 115,782 |
[
"kubernetes",
"kubernetes"
] | - Tests in `test/e2e/common/node/container_probe.go` are using `etcd` image for gRPC probe tests:
- should not be restarted with a GRPC liveness probe
- should be restarted with a GRPC liveness probe
We should use `agnhost` image for these tests as it is much smaller size.
/sig node
/kind cleanup
/priorit... | Switch image in gRPC probe tests to agnhost | https://api.github.com/repos/kubernetes/kubernetes/issues/115781/comments | 9 | 2023-02-15T00:06:54Z | 2023-02-18T18:19:38Z | https://github.com/kubernetes/kubernetes/issues/115781 | 1,585,009,207 | 115,781 |
[
"kubernetes",
"kubernetes"
] | - Instructions: https://github.com/kubernetes/community/blob/master/contributors/devel/sig-architecture/conformance-tests.md#promoting-tests-to-conformance
- Tests in `test/e2e/common/node/container_probe.go`:
- should not be restarted with a GRPC liveness probe
- should be restarted with a GRPC liveness probe
... | Promote gRPC probe tests to conformance | https://api.github.com/repos/kubernetes/kubernetes/issues/115780/comments | 8 | 2023-02-15T00:06:25Z | 2023-03-09T09:06:05Z | https://github.com/kubernetes/kubernetes/issues/115780 | 1,585,008,718 | 115,780 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
`NodeLifeCycleController` applies predefined `NoExecute` taints (e.g., `Unreachable`, `NotReady`) to mark nodes as unhealthy state, when it cannot obtain heartbeat acks from the nodes. After the nodes get tainted, the `taint-manager` does its due diligence to start evicting runni... | Decouple Taint-based Pod Eviction from Node LifeCycle Controller | https://api.github.com/repos/kubernetes/kubernetes/issues/115779/comments | 12 | 2023-02-14T22:46:07Z | 2023-05-23T15:58:50Z | https://github.com/kubernetes/kubernetes/issues/115779 | 1,584,940,253 | 115,779 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are flaking?
- https://testgrid.k8s.io/sig-windows-1.25-release#windows-containerd-azure-disk-1.25
### Which tests are flaking?
- kubetest.Up
### Since when has it been flaking?
Since 01/31/2023
### Testgrid link
https://testgrid.k8s.io/sig-windows-1.25-release#windows-containerd-azure-disk-1.25
... | windows-containerd-azure-disk-1.25 seems to be flaky | https://api.github.com/repos/kubernetes/kubernetes/issues/115778/comments | 8 | 2023-02-14T20:48:28Z | 2024-03-22T15:28:01Z | https://github.com/kubernetes/kubernetes/issues/115778 | 1,584,806,129 | 115,778 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are flaking?
sig-release-master-blocking
[build-master](https://k8s-testgrid.appspot.com/sig-release-master-blocking#build-master)
[gce-device-plugin-gpu-master](https://k8s-testgrid.appspot.com/sig-release-master-blocking#gce-device-plugin-gpu-master)
[gce-cos-master-default](https://k8s-testgrid.ap... | [Flaking Tests] multiple jobs - Pod deleted unexpectedly | https://api.github.com/repos/kubernetes/kubernetes/issues/115767/comments | 14 | 2023-02-14T16:28:09Z | 2023-02-15T22:13:47Z | https://github.com/kubernetes/kubernetes/issues/115767 | 1,584,469,434 | 115,767 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The metric `prober_probe_total` used to report acutal pod in the label `pod`, but now I see the podname prefix only.
### What did you expect to happen?
The full pod name in the label, so I can trace the metric to the failing pod.
### How can we reproduce it (as minimally and precisely as ... | Metric prober_probe_total reports invalid pod field | https://api.github.com/repos/kubernetes/kubernetes/issues/115766/comments | 11 | 2023-02-14T16:23:45Z | 2023-07-27T21:03:36Z | https://github.com/kubernetes/kubernetes/issues/115766 | 1,584,461,742 | 115,766 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
master-blocking
[Conformance - GCE - master - kubetest2](https://testgrid.k8s.io/sig-release-master-blocking#Conformance%20-%20GCE%20-%20master%20-%20kubetest2)
### Which tests are failing?
ci-kubernetes-gce-conformance-latest-kubetest2.Overall
ci-kubernetes-gce-conformance-latest-kub... | [Failing tests] Conformance - GCE - master - kubetest2 : Pod deleted unexpectedly | https://api.github.com/repos/kubernetes/kubernetes/issues/115765/comments | 9 | 2023-02-14T16:12:53Z | 2023-02-14T19:32:36Z | https://github.com/kubernetes/kubernetes/issues/115765 | 1,584,443,799 | 115,765 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
As part of adding support for dynamic resource allocation to `test/integration/scheduler_perf`, I ran `go test -race`. With DRA, scheduling failures occur, and that led to a race report:
```
$ go test -v -timeout=0 -bench=.*/SchedulingWithInlineResourceClaims/ -race ./test/integration/schedule... | scheduler: data race between PodInfo.Update and PriorityQueue.AddNominatedPod | https://api.github.com/repos/kubernetes/kubernetes/issues/115762/comments | 4 | 2023-02-14T15:49:51Z | 2023-03-10T06:44:20Z | https://github.com/kubernetes/kubernetes/issues/115762 | 1,584,403,119 | 115,762 |
[
"kubernetes",
"kubernetes"
] | How to repro:
1. Start kubelet
2. check that the node has been registered
```
kubectl get nodes test-worker
NAME STATUS ROLES AGE VERSION
test-worker Ready <none> 59m v1.27.0
```
3. delete the node
```
kubectl delete nodes test-worker
node "test-worker" deleted
```
kubelet neve... | Kubelet does not try to register again if its Node object is deleted | https://api.github.com/repos/kubernetes/kubernetes/issues/115761/comments | 10 | 2023-02-14T12:53:30Z | 2024-02-09T13:09:48Z | https://github.com/kubernetes/kubernetes/issues/115761 | 1,584,117,866 | 115,761 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
Assuming that updates the shared libs could affect the binaries provided. Would be possible to provide the binaries with static compilation to avoid this scenarios like that?
### Why is this needed?
This one is related to the concerns raised in: https://github.com/kuberne... | Provide binaries with static compilation | https://api.github.com/repos/kubernetes/kubernetes/issues/115757/comments | 7 | 2023-02-14T11:59:49Z | 2023-07-14T18:02:23Z | https://github.com/kubernetes/kubernetes/issues/115757 | 1,584,038,914 | 115,757 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Observed unexpected memory usage in follower/non-leader `kube-scheduler` instances
### What did you expect to happen?
The `kube-scheduler` in standby mode should be only having minimum memory overhead until being elected.
### How can we reproduce it (as minimally and precisely as possible)?
* cr... | Unexpected memory usage in follower kube-scheduler processes | https://api.github.com/repos/kubernetes/kubernetes/issues/115753/comments | 2 | 2023-02-14T09:49:41Z | 2023-05-15T22:41:37Z | https://github.com/kubernetes/kubernetes/issues/115753 | 1,583,838,563 | 115,753 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
Add automatic detection of DNS lookup latency regression in performance tests at scale.
### Why is this needed?
Recently, a major DNS regression has been manually detected in a performance test at scale (5k nodes). https://github.com/kubernetes/kubernetes/issues/115720
The chang... | Auto detection of DNS performance regression in performance tests at scale | https://api.github.com/repos/kubernetes/kubernetes/issues/115752/comments | 10 | 2023-02-14T09:34:16Z | 2024-03-22T23:32:58Z | https://github.com/kubernetes/kubernetes/issues/115752 | 1,583,814,712 | 115,752 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
It's much similar to https://github.com/kubernetes/kubernetes/issues/114399.
In https://github.com/kubernetes/kubernetes/issues/114827, we support Skip in PreScore/Score as well. (Thanks @kidddddddddddddddddddddd)
So, the plugin can return Skip in PreScore when it doesn't do an... | [Umbrella] Skip Score plugins when coupled PreScore plugin returns Skip status | https://api.github.com/repos/kubernetes/kubernetes/issues/115745/comments | 22 | 2023-02-14T03:13:23Z | 2024-01-07T04:28:57Z | https://github.com/kubernetes/kubernetes/issues/115745 | 1,583,416,494 | 115,745 |
[
"kubernetes",
"kubernetes"
] | As mentioned https://github.com/kubernetes/kubernetes/pull/110601/files#r1004432225 the return value of `getAppArmorFS` is not being used any longer and the entire method can be removed.
The usage of `appArmorFS` was removed in https://github.com/kubernetes/kubernetes/pull/97966
Now there is no need to have the f... | Remove the getAppArmorFS function | https://api.github.com/repos/kubernetes/kubernetes/issues/115741/comments | 8 | 2023-02-13T20:52:27Z | 2023-02-23T08:31:36Z | https://github.com/kubernetes/kubernetes/issues/115741 | 1,583,043,376 | 115,741 |
[
"kubernetes",
"kubernetes"
] | Do we have folks that use `linux/arm` on Raspberry Pi's that directly use Kubernetes release artifacts?
Asking as there are projects like k3s that have lightweight alternatives, so maybe we don't need to own this anymore?
/sig release
/sig testing
PS: we are not asking about 64 bit arm which goes by `linux/ar... | Do we need to support `linux/arm` build/release artifacts in kubernetes (32-bit arm for Raspberry PIs) | https://api.github.com/repos/kubernetes/kubernetes/issues/115738/comments | 46 | 2023-02-13T19:36:49Z | 2023-05-16T21:57:44Z | https://github.com/kubernetes/kubernetes/issues/115738 | 1,582,942,134 | 115,738 |
[
"kubernetes",
"kubernetes"
] | from the given instructions at kubenetes home site, while running below getting following issue, it is not working on any of the systems.
sudo yum install -y kubelet kubeadm kubectl --disableexcludes=kubernetes
```
Loaded plugins: rhnplugin
This system is receiving updates from Spacewalk server.
Repository uekr5... | Trouble installing Kubernetes components using `yum` | https://api.github.com/repos/kubernetes/kubernetes/issues/115743/comments | 10 | 2023-02-13T18:52:44Z | 2023-07-14T18:02:24Z | https://github.com/kubernetes/kubernetes/issues/115743 | 1,583,247,784 | 115,743 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
References:
https://github.com/kubernetes/kubernetes/pull/111155/files#r1096010921
https://github.com/kubernetes/kubernetes/pull/111155/files#r929904171
https://github.com/kubernetes/kubernetes/pull/115588#discussion_r1098970348
Opening this issue for discussion. Then i... | Decide on whether to pass context from defaulting code of KS config (to use with ctx logging) | https://api.github.com/repos/kubernetes/kubernetes/issues/115724/comments | 7 | 2023-02-13T14:26:52Z | 2025-03-07T18:20:00Z | https://github.com/kubernetes/kubernetes/issues/115724 | 1,582,453,265 | 115,724 |
[
"kubernetes",
"kubernetes"
] | On the documentation page for [`kube-controller-manager`](https://kubernetes.io/docs/reference/command-line-tools-reference/kube-controller-manager/)
The following doesn't seem correct
```
--client-ca-file string
If set, any request presenting a client certificate signed by one of the authorities in the client-... | kube-controller-manager supports `--client-ca-file` parameter | https://api.github.com/repos/kubernetes/kubernetes/issues/115725/comments | 7 | 2023-02-13T11:40:49Z | 2023-02-19T03:00:05Z | https://github.com/kubernetes/kubernetes/issues/115725 | 1,582,486,348 | 115,725 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
https://perf-dash.k8s.io/#/?jobname=gce-5000Nodes&metriccategoryname=DNS&metricname=Load_DNSLookupLatency&Metric=DnsLookupLatency
### What did you expect to happen?
performance should not change to be worse.
### How can we reproduce it (as minimally and precisely as possible)?
See th... | dns lookup latency performance downgrade | https://api.github.com/repos/kubernetes/kubernetes/issues/115720/comments | 15 | 2023-02-13T09:42:37Z | 2023-02-20T07:25:13Z | https://github.com/kubernetes/kubernetes/issues/115720 | 1,581,990,261 | 115,720 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Hi all,
Since some time ago we are facing an issue while displaying Prometheus metrics. We are seeing that for the same namespace and the same node pool CPU/Memory metrics are being displayed and some others not. We have checked the issue with Azure but they are pointing more to an upstream iss... | Prometheus metrics is not showing metrics for some PODs | https://api.github.com/repos/kubernetes/kubernetes/issues/115718/comments | 16 | 2023-02-13T09:30:12Z | 2024-04-18T16:38:52Z | https://github.com/kubernetes/kubernetes/issues/115718 | 1,581,973,795 | 115,718 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
I: Motivation:
We have some per-node setup works in our platform when provisioning new nodes, but they should operate as an one-shoot manner, so we tried:
- `daemonSet`, but pods will not run-to-complete, this is not fully what we want. But yes, it can work like idling in a s... | [Proposal] Run once daemonSet (daemonJob) support in Kubernetes | https://api.github.com/repos/kubernetes/kubernetes/issues/115716/comments | 48 | 2023-02-13T07:53:27Z | 2025-02-17T10:39:51Z | https://github.com/kubernetes/kubernetes/issues/115716 | 1,581,836,287 | 115,716 |
[
"kubernetes",
"kubernetes"
] | Default range of node port is 30000 ~ 32767 and I just wonder if there's a way to seperate it by service types.
The main concept would be
A service : 30000 ~ 30100
B service : 30101 ~ 30200
C service : 30201 ~ 30300
| (Question) Can Node Port Range be seperated by service?? | https://api.github.com/repos/kubernetes/kubernetes/issues/115703/comments | 4 | 2023-02-11T15:43:12Z | 2023-02-11T16:38:56Z | https://github.com/kubernetes/kubernetes/issues/115703 | 1,580,895,207 | 115,703 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
`kubectl logs POD_NAME -f ` won't output logs after kubelet rotate log filer of the container.
### What did you expect to happen?
kubectl output logs correctly after the log file is rotated.
### How can we reproduce it (as minimally and precisely as possible)?
1. Add the config `containerLogM... | kubectl logs will not output logs after the log file is rotated. | https://api.github.com/repos/kubernetes/kubernetes/issues/115701/comments | 6 | 2023-02-11T13:22:20Z | 2023-10-14T20:42:06Z | https://github.com/kubernetes/kubernetes/issues/115701 | 1,580,840,013 | 115,701 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
KCM uses generic informers to watch all k8s objects, which is unnecessary memory overhead as all generic controllers currently only use object metadata. This issue makes KCM have proportional memory usage to ETCD database size.
### What did you expect to happen?
KCM has low memory usage which is p... | High memory usage of kube-controller-manager for caching full objects | https://api.github.com/repos/kubernetes/kubernetes/issues/115699/comments | 7 | 2023-02-11T12:50:13Z | 2025-02-21T18:20:18Z | https://github.com/kubernetes/kubernetes/issues/115699 | 1,580,826,229 | 115,699 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Trying to run this
"kubectl config set-credentials system:kube-proxy --client-certificate=kube-proxy.pem --client-key=kube-proxy-key.pem --embed-certs=true --kubeconfig=kube-proxy.kubeconfig
"
### What did you expect to happen?
I need to have kube-proxy.kubeconfig file
### How can we reproduce... | Panic Runtime Error on Kali | https://api.github.com/repos/kubernetes/kubernetes/issues/115698/comments | 5 | 2023-02-11T12:25:20Z | 2023-02-16T09:07:53Z | https://github.com/kubernetes/kubernetes/issues/115698 | 1,580,816,999 | 115,698 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
When I tested the local storage capabilities of the CSI Plugin， The VG does not exist. Create a PVC in pending state，Creating a VG，It takes about 2 minutes for a PVC to change from the pending state to the bound state. According to the node controller log of the local storage, it... | The PVC is created first, and then the VG is created. Why does the PVC take a long time to change from pending to bound? | https://api.github.com/repos/kubernetes/kubernetes/issues/115696/comments | 8 | 2023-02-11T08:54:57Z | 2023-07-13T10:45:20Z | https://github.com/kubernetes/kubernetes/issues/115696 | 1,580,743,696 | 115,696 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
We appear to have two methods for creating kube-openapi's spec.Schema type being employed today.
For validation, there is a direct conversion `CustomResourceValidation -> spec.Schema` written in an ad-hoc function for validation. This schema is used to perform validation of CRs against the sche... | CRD validation uses separate CRD->OpenAPI conversion from SSA | https://api.github.com/repos/kubernetes/kubernetes/issues/115691/comments | 9 | 2023-02-10T20:54:25Z | 2025-02-19T21:05:15Z | https://github.com/kubernetes/kubernetes/issues/115691 | 1,580,313,148 | 115,691 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
I've noticed an edge case when trying out the PodFailurePolicy example from Story 3 of the KEP. When a pod is marked as a DisruptionTarget but then completes successful, it retries anyways. I wasn't able to add some sort of Ignore rule to exclude the 0 case
### What did you expect to happen?
I'm n... | Pod Failure Policy Edge Case: Job Retries When Pod Finishes Successfully | https://api.github.com/repos/kubernetes/kubernetes/issues/115688/comments | 47 | 2023-02-10T16:47:14Z | 2024-12-18T18:58:24Z | https://github.com/kubernetes/kubernetes/issues/115688 | 1,580,019,116 | 115,688 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
When constructing a `rest.Request`, _e.g._ to stream I/O from a pod:
```
req := k8s.CoreV1().RESTClient().Post().Resource("pods").Name(podName).Namespace(podNamespace).SubResource("exec")
req.VersionedParams(&v1.PodExecOptions{
Command: command,
Stdin: false,
Stdout: true,
St... | client-go: if an error occurs during request construction, URL() can return a useless result with no indication of an error | https://api.github.com/repos/kubernetes/kubernetes/issues/115683/comments | 6 | 2023-02-10T13:21:27Z | 2023-04-20T16:59:11Z | https://github.com/kubernetes/kubernetes/issues/115683 | 1,579,676,526 | 115,683 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
yaml:
```
lifecycle:
postStart:
exec:
command:
- /bin/sh
- -c
- |
cp -r conf/* ./
mount -t cifs xxxxx
mount -t cifs xxxxx
mount -t cifs xxxxx
```
FailedPostStartHook Exec lifecycle ho... | FailedPostStartHook exited with 137: , message: "" | https://api.github.com/repos/kubernetes/kubernetes/issues/115680/comments | 9 | 2023-02-10T09:41:02Z | 2023-07-14T04:50:24Z | https://github.com/kubernetes/kubernetes/issues/115680 | 1,579,355,326 | 115,680 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
Add an option to kubectl debug --copy-to to either on the fly add labels to the clone or copy the existing labels from the origin.
### Why is this needed?
`kubectl debug --copy-to` currently does not copy labels. I reckon there is a good reason for it. (Probably if a pod from a r... | kubectl debug --copy-to option copy or add labels | https://api.github.com/repos/kubernetes/kubernetes/issues/115679/comments | 4 | 2023-02-10T09:40:07Z | 2023-02-10T12:09:52Z | https://github.com/kubernetes/kubernetes/issues/115679 | 1,579,353,998 | 115,679 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
master-blocking
- build-master
### Which tests are failing?
Overall build is failing
### Since when has it been failing?
2023-02-09 23:09... | [Failing Test] build-master | https://api.github.com/repos/kubernetes/kubernetes/issues/115675/comments | 12 | 2023-02-10T05:17:43Z | 2023-03-11T04:27:06Z | https://github.com/kubernetes/kubernetes/issues/115675 | 1,579,000,373 | 115,675 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
https://cloud.google.com/kubernetes-engine/docs/deprecations/docker-containerd?authuser=2&hl=ko&_ga=2.242608960.-33629090.1617698791&_gac=1.22591817.1675752022.Cj0KCQiA54KfBhCKARIsAJzSrdqL2c_qiqLCRHtyBPnzMPSSY1W3m-C8JvvEtjJrCnsAYEIPEaqdMYwaAs7CEALw_wcB
### What did you expect to happen?
I saw so... | Kubernetes no support docker Runtime? | https://api.github.com/repos/kubernetes/kubernetes/issues/115673/comments | 2 | 2023-02-10T01:41:39Z | 2023-02-10T03:08:43Z | https://github.com/kubernetes/kubernetes/issues/115673 | 1,578,857,416 | 115,673 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
`kubectl debug --profile restricted` errors out with `container has runAsNonRoot and image will run as root`
We need to set a `RunAsUser` value for this debug profile, e.g. `RunAsUser: pointer.Int64(65534)`
| restricted debug profile needs a "RunAsUser" specified | https://api.github.com/repos/kubernetes/kubernetes/issues/115671/comments | 8 | 2023-02-10T00:45:36Z | 2023-02-16T14:00:16Z | https://github.com/kubernetes/kubernetes/issues/115671 | 1,578,820,042 | 115,671 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
https://testgrid.k8s.io/sig-windows-master-release#capz-windows-containerd-2022-master-serial-slow-hpa
https://testgrid.k8s.io/sig-windows-master-release#capz-windows-containerd-master-serial-slow-hpa
### Which tests are failing?
Kubernetes e2e suite.[It] [sig-autoscaling] [Feature:HPA] ... | Subset of HPA related tests are Flaking | https://api.github.com/repos/kubernetes/kubernetes/issues/115664/comments | 12 | 2023-02-09T19:59:57Z | 2024-03-26T13:13:41Z | https://github.com/kubernetes/kubernetes/issues/115664 | 1,578,504,155 | 115,664 |
[
"kubernetes",
"kubernetes"
] | SubjectAccessReview [expects "*"](https://github.com/kubernetes/api/blob/0478a3e95231398d8b380dc2a1905972be8ae1d5/authorization/v1/types.go#L106-L108) (not "") if the version is unspecified. We should audit all authz checks to ensure that "*" is being passed everywhere.
Example of where version is missing: https://g... | Ensure version "*" is passed instead of "" for all authz checks | https://api.github.com/repos/kubernetes/kubernetes/issues/115660/comments | 7 | 2023-02-09T18:23:19Z | 2023-04-25T16:06:19Z | https://github.com/kubernetes/kubernetes/issues/115660 | 1,578,389,998 | 115,660 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
For quite a while we have seen that some controllers/informers suddenly get out of sync with the state of the world, and keep `x` (different number every time).
After more investigation, we think we narrowed it down to the informer having a queue of events while a relist occur, and for some obj... | client-go/cache: informer misses delete events on relist for elements only present in the queue | https://api.github.com/repos/kubernetes/kubernetes/issues/115658/comments | 37 | 2023-02-09T15:21:40Z | 2023-03-20T16:16:22Z | https://github.com/kubernetes/kubernetes/issues/115658 | 1,578,094,915 | 115,658 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The [gRPC probe](https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/#define-a-grpc-liveness-probe) currently in beta will mark the pod as live (resp. ready) only if the [grpc Health Service](https://github.com/grpc/grpc/blob/master/doc/health-checkin... | gRPC probe semantics prevent their usage for both readiness and liveness | https://api.github.com/repos/kubernetes/kubernetes/issues/115651/comments | 12 | 2023-02-09T11:06:08Z | 2024-08-09T18:19:40Z | https://github.com/kubernetes/kubernetes/issues/115651 | 1,577,685,071 | 115,651 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Deployed Kubernetes on a single node setup (master+worker). The setup was UP, checked Kubelet + certs managed by Kubeadm has a validity of 1yr. Now, changed the date of the system to 30% of the cert's expiry date.
Ran Kubeadm certs renew all, certs managed by Kubeadm got renewed from the day t... | kubelet client cert is not getting auto-rotated (K8s 1.21) | https://api.github.com/repos/kubernetes/kubernetes/issues/115648/comments | 9 | 2023-02-09T07:43:19Z | 2023-08-19T15:29:31Z | https://github.com/kubernetes/kubernetes/issues/115648 | 1,577,399,468 | 115,648 |
[
"kubernetes",
"kubernetes"
] | ### Failure cluster [6c45426c2c004a6e3571](https://go.k8s.io/triage#6c45426c2c004a6e3571)
after stopping the nfs-server and deleting the (active) client pod, the NFS mount and the pod's UID directory should be removed.
> Kubernetes e2e suite [It] [sig-node] kubelet host cleanup with volume mounts [HostCleanup][Flak... | e2e: kubelet host cleanup with volume mounts [HostCleanup][Flaky] [NFS] | https://api.github.com/repos/kubernetes/kubernetes/issues/115647/comments | 5 | 2023-02-09T06:31:37Z | 2023-07-24T10:17:03Z | https://github.com/kubernetes/kubernetes/issues/115647 | 1,577,323,482 | 115,647 |
[
"kubernetes",
"kubernetes"
] | Part of https://github.com/kubernetes/kubernetes/issues/115467
Ref: https://github.com/kubernetes/enhancements/issues/1610
### What would you like to be added?
Currently, HPA controller doesn't have any metrics in it.
Ideas:
- `scaling_attempt_duration_seconds`: How long the HPA controller takes to comple... | expose some metrics from HPA controller | https://api.github.com/repos/kubernetes/kubernetes/issues/115639/comments | 19 | 2023-02-09T01:48:45Z | 2025-03-04T21:22:07Z | https://github.com/kubernetes/kubernetes/issues/115639 | 1,577,081,281 | 115,639 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
https://prow.k8s.io/?job=pull-kubernetes-unit
reported by @jiahuif here:
https://kubernetes.slack.com/archives/C7J9RP96G/p1675902134074689
### Which tests are failing?
the entire job
### Since when has it been failing?
recently
### Testgrid link
_No response_
... | pull-kubernetes-unit failing | https://api.github.com/repos/kubernetes/kubernetes/issues/115637/comments | 10 | 2023-02-09T00:35:25Z | 2023-02-09T04:59:50Z | https://github.com/kubernetes/kubernetes/issues/115637 | 1,577,012,698 | 115,637 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The`attachable-volumes-aws-ebs` Allocatable field attached to EC2 nitro nodes (c5/m5s) on start up by default is set to 25. However, this limit does not account for dynamic ENI allocation/deallocation that CNIs like the [aws-cni](https://github.com/aws/amazon-vpc-cni-k8s) performs. This causes the... | attachable-volumes-aws-ebs Allocatable Field Does Not Account For ENIs | https://api.github.com/repos/kubernetes/kubernetes/issues/115632/comments | 5 | 2023-02-08T21:51:13Z | 2023-05-05T15:40:17Z | https://github.com/kubernetes/kubernetes/issues/115632 | 1,576,846,816 | 115,632 |
[
"kubernetes",
"kubernetes"
] | With the [change to defaulting](https://kubernetes.io/blog/2022/11/28/registry-k8s-io-faster-cheaper-ga/) to registry.k8s.io as our image registry and the [planned freeze in April](https://kubernetes.io/blog/2023/02/06/k8s-gcr-io-freeze-announcement/), projects within the Kubernetes orgs should update their image refer... | [Action Required] Update references from k8s.gcr.io to registry.k8s.io | https://api.github.com/repos/kubernetes/kubernetes/issues/115631/comments | 6 | 2023-02-08T21:40:56Z | 2023-02-17T01:05:29Z | https://github.com/kubernetes/kubernetes/issues/115631 | 1,576,836,603 | 115,631 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Upgrading kubernetes-cni package from 1.1.1 to 1.2.0 removes and recreates the /opt/cni/bin directory.
This removes all CNI plugin binaries from the directory.
When using flannel CNI this causes pods to get stuck in ContainerCreating until flannel pod is deleted and recreated.
See https:/... | Upgrading kubernetes-cni package from 1.1.1 to 1.2.0 removes and recreates the /opt/cni/bin directory | https://api.github.com/repos/kubernetes/kubernetes/issues/115629/comments | 20 | 2023-02-08T19:37:20Z | 2024-01-27T10:42:46Z | https://github.com/kubernetes/kubernetes/issues/115629 | 1,576,697,183 | 115,629 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
when using server side apply to create an object, only a patch event is logged in the audit log and no create event is event logged.
cc @liggitt
### What did you expect to happen?
a create event should be logged in the audit log.
### How can we reproduce it (as minimally and precisely as poss... | Server Side Apply - No Create event in audit log | https://api.github.com/repos/kubernetes/kubernetes/issues/115627/comments | 6 | 2023-02-08T18:29:42Z | 2025-02-12T22:09:01Z | https://github.com/kubernetes/kubernetes/issues/115627 | 1,576,605,633 | 115,627 |
Subsets and Splits
Unique Owner-Repo Count
Counts the number of unique owner-repos in the dataset, providing a basic understanding of diverse repositories.