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?
1. sometimes, when we add or delete anti-affinity pods with pvc, the pod is pending with reason by "node(s) didn't match pod anti-affinity rules", I forgot to check the scheduler cache that time.
2. but when we restart kube-sheduler, the pod can be scheduled successfully again
the logs can... | kube-scheduler cache has not updated sometimes | https://api.github.com/repos/kubernetes/kubernetes/issues/120512/comments | 6 | 2023-09-08T02:11:28Z | 2023-12-05T06:54:39Z | https://github.com/kubernetes/kubernetes/issues/120512 | 1,886,824,608 | 120,512 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Since upgrading to Kubernetes `1.27.5` (from 1.27.4) we have experienced outages on our `kube-apiserver` pods where their container has been restarted due to failing liveness probes.
This has caused instability, usually via a number of important components within our clusters failing leader elect... | kube-apiserver instability for k8s 1.27.5 | https://api.github.com/repos/kubernetes/kubernetes/issues/120510/comments | 15 | 2023-09-08T00:31:37Z | 2023-09-18T00:30:49Z | https://github.com/kubernetes/kubernetes/issues/120510 | 1,886,765,364 | 120,510 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
During periods when there's webhook failures, it is possible for the apiserver to crash while generating an audit log entry for the given request. Specifically, when a large number of pods are deleted that are intercepted by a mutating webhook that is timing out, the following fatal error can be e... | apiserver crashes on concurrent map iteration and write | https://api.github.com/repos/kubernetes/kubernetes/issues/120507/comments | 31 | 2023-09-07T20:37:22Z | 2025-03-10T16:31:43Z | https://github.com/kubernetes/kubernetes/issues/120507 | 1,886,533,002 | 120,507 |
[
"kubernetes",
"kubernetes"
] | DRA (DynamicResourceAllocation) plugin tries to issue a synchronous API call in Reverse():
https://github.com/kubernetes/kubernetes/blob/c682d2b8c59f615d6896a1b1db22a5e7a5aaa23e/pkg/scheduler/framework/plugins/dynamicresources/dynamicresources.go#L883
However, raising direct API requests in main scheduling cycle ... | [discussion] DRA plugin raises UPDATE API request in scheduling cycle | https://api.github.com/repos/kubernetes/kubernetes/issues/120502/comments | 15 | 2023-09-07T16:38:20Z | 2024-03-06T17:32:32Z | https://github.com/kubernetes/kubernetes/issues/120502 | 1,886,234,216 | 120,502 |
[
"kubernetes",
"kubernetes"
] | previous discussion: https://github.com/kubernetes/kubernetes/pull/120461#discussion_r1317515863
https://github.com/kubernetes/kubernetes/blob/56cc5e77a10ba156694309d9b6159d4cd42598e1/pkg/kubelet/apis/podresources/server_v1.go#L51-L86
kubelet's PodResources API does not report the resources allocated to restartab... | PodResources API of kubelet does not report the resources allocated to restartable init containers | https://api.github.com/repos/kubernetes/kubernetes/issues/120501/comments | 5 | 2023-09-07T14:58:09Z | 2024-03-05T21:44:55Z | https://github.com/kubernetes/kubernetes/issues/120501 | 1,886,068,233 | 120,501 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
master-informing:
- [gce-master-scale-correctness](https://testgrid.k8s.io/sig-release-master-informing#gce-master-scale-correctness)
- [gce-master-scale-performance](https://testgrid.k8s.io/sig-release-master-informing#gce-master-scale-performance)
### Which tests are failing?
`kubetest... | [Failing Test] Error starting e2e cluster (kubetest.Up) | https://api.github.com/repos/kubernetes/kubernetes/issues/120500/comments | 7 | 2023-09-07T14:52:52Z | 2023-11-20T10:14:11Z | https://github.com/kubernetes/kubernetes/issues/120500 | 1,886,059,090 | 120,500 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
spec.variables should be marked omitempty.
### Why is this needed?
I'm trying to create a validating admission policy that uses `spec.variables` in v1.27 as follows:
```
// set validating admission policy spec
vap.Spec = v1alpha1.ValidatingAdmissionPolicySpec {
ParamK... | ValidatingAdmissionPolicy: spec.variables should be marked omitempty | https://api.github.com/repos/kubernetes/kubernetes/issues/120491/comments | 4 | 2023-09-07T13:24:20Z | 2023-10-19T13:57:23Z | https://github.com/kubernetes/kubernetes/issues/120491 | 1,885,893,202 | 120,491 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
A Deployment's number of replicas does not match what the `kubectl.kubernetes.io/last-applied-configuration` is showing.
This is the deployment status, Notice the number of replicas is 1 in `kubectl.kubernetes.io/last-applied-configuration`, but the actual number of replicas deployed is 3 (`rep... | Deployment number of replicas doesn't match last applied configuration | https://api.github.com/repos/kubernetes/kubernetes/issues/120487/comments | 6 | 2023-09-07T10:10:07Z | 2023-09-11T02:05:09Z | https://github.com/kubernetes/kubernetes/issues/120487 | 1,885,585,979 | 120,487 |
[
"kubernetes",
"kubernetes"
] | TLDR, There are many moving parts in a K8s cluster that may use / depend on
- the `"node.kubernetes.io/network-unavailable" condition` and/or
- `v1.TaintNodeNetworkUnavailable taint` Condition/Taint couple
In ambiguous/ad-hoc sorts of ways. This makes it confusing for users and vendors to know when they should/s... | (umbrella issue) Define, Deprecate, Roadmap the NodeNetworkUnvailable / NetworkUnavailable:false taints, tolerations, conditions - including CNI/CloudController reccomendations | https://api.github.com/repos/kubernetes/kubernetes/issues/120486/comments | 27 | 2023-09-07T09:46:41Z | 2024-04-19T20:41:57Z | https://github.com/kubernetes/kubernetes/issues/120486 | 1,885,532,245 | 120,486 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
master-informing:
- [capg-conformance-main-ci-artifacts](https://testgrid.k8s.io/sig-release-master-informing#capg-conformance-main-ci-artifacts)
### Which tests are failing?
`capg-e2e.[It] Conformance Tests Should run conformance tests`
 No Control Plane machines came into existence | https://api.github.com/repos/kubernetes/kubernetes/issues/120481/comments | 35 | 2023-09-07T07:26:35Z | 2024-08-09T10:10:53Z | https://github.com/kubernetes/kubernetes/issues/120481 | 1,885,278,130 | 120,481 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
https://testgrid.k8s.io/conformance-all#Conformance%20-%20GCE%20-%20master
### Which tests are failing?
[It] [sig-api-machinery] Namespaces [Serial] should apply changes to a namespace status [Conformance]
### Since when has it been failing?
Since https://github.com/kubernetes/kuberne... | Failed test Namespaces [Serial] should apply changes to a namespace status [Conformance] | https://api.github.com/repos/kubernetes/kubernetes/issues/120478/comments | 5 | 2023-09-07T05:37:12Z | 2024-02-20T19:50:14Z | https://github.com/kubernetes/kubernetes/issues/120478 | 1,885,148,958 | 120,478 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The HPA is encountering the "FailedComputeMetricsReplicas" error, preventing it from computing metrics and making scaling decisions and I have received the following error message:
I0902 11:57:32.576643 1 event.go:294] "Event occurred" object="***/storage" fieldPath="" kind="HorizontalPodAu... | FailedComputeMetricsReplicas: failed to get memory and CPU utilization | https://api.github.com/repos/kubernetes/kubernetes/issues/120477/comments | 11 | 2023-09-07T04:09:32Z | 2024-04-03T14:45:14Z | https://github.com/kubernetes/kubernetes/issues/120477 | 1,885,080,358 | 120,477 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
After performing the `kubectl apply app.yaml --server-side --force-conflicts` on the server side, the "managedFields" information appears to be missing.
The normal case should be
```yaml
# k get svc haproxy-nlb-jfg-0 --show-managed-fields=true -o yaml | yq .metadata.managedFields
- apiVersio... | [Bug] The "managedFields" field in abnormal state after applying resources | https://api.github.com/repos/kubernetes/kubernetes/issues/120476/comments | 14 | 2023-09-07T03:57:36Z | 2024-12-19T21:23:17Z | https://github.com/kubernetes/kubernetes/issues/120476 | 1,885,071,901 | 120,476 |
[
"kubernetes",
"kubernetes"
] | ### Failure cluster [63bfe963e52db942c847](https://go.k8s.io/triage#63bfe963e52db942c847)
##### Error text:
```
[FAILED] PV Create API error: persistentvolumes "local-pv8vhdt" is forbidden: expression 'object.metadata.namespace == 'production'' resulted in error: no such key: namespace
In [BeforeEach] at: test/e2... | Failure cluster [63bfe963...] [Volume type: blockfswithoutformat] Two pods mounting a local volume one after the other should be able to write from pod1 and read from pod2 exp | https://api.github.com/repos/kubernetes/kubernetes/issues/120475/comments | 8 | 2023-09-07T03:33:24Z | 2024-02-20T19:50:06Z | https://github.com/kubernetes/kubernetes/issues/120475 | 1,885,055,541 | 120,475 |
[
"kubernetes",
"kubernetes"
] | # Progress <code>[7/7]</code>
- [X] APISnoop org-flow : [StorageV1StorageClass-LifecycleTest.org](https://github.com/apisnoop/ticket-writing/blob/master/StorageV1StorageClass-LifecycleTest.org)
- [X] test approval issue : #120470
- [X] test pr : #120471
- [x] two weeks soak start date : 8 Sept 2023 [test... | Write e2e test for StorageClass Endpoints + 7 Endpoints | https://api.github.com/repos/kubernetes/kubernetes/issues/120470/comments | 4 | 2023-09-06T18:13:45Z | 2023-09-20T14:02:04Z | https://github.com/kubernetes/kubernetes/issues/120470 | 1,884,518,856 | 120,470 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
pull-kubernetes-e2e-gce-cos-alpha-features
### Which tests are failing?
"Up"
https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/pull/120417/pull-kubernetes-e2e-gce-cos-alpha-features/1699385890981285888/
### Since when has it been failing?
since this weekend, 1st of Sept
... | [Failing Test job] pull-kubernetes-e2e-gce-cos-alpha-features | https://api.github.com/repos/kubernetes/kubernetes/issues/120463/comments | 1 | 2023-09-06T14:30:51Z | 2023-09-06T16:38:54Z | https://github.com/kubernetes/kubernetes/issues/120463 | 1,884,144,978 | 120,463 |
[
"kubernetes",
"kubernetes"
] | ### Failure cluster [9339868fc1cf8e570508](https://go.k8s.io/triage#9339868fc1cf8e570508)
##### Error text:
```
[FAILED] couldn't find that PostStart-regular-1 ever exited, got 2023-08-29 00:39:12 +0000 UTC 49.060000 regular-1 Starting
2023-08-29 00:39:12 +0000 UTC 49.070000 regular-1 Started
2023-08-29 00:39:12... | Failure cluster [9339868f...] Containers Lifecycle should not launch second container before PostStart of the first container completed | https://api.github.com/repos/kubernetes/kubernetes/issues/120458/comments | 4 | 2023-09-06T13:13:40Z | 2023-09-19T00:35:54Z | https://github.com/kubernetes/kubernetes/issues/120458 | 1,883,993,122 | 120,458 |
[
"kubernetes",
"kubernetes"
] | ### Failure cluster [9f7f7053b14ff9d31a67](https://go.k8s.io/triage#9f7f7053b14ff9d31a67)
##### Error text:
```
[FAILED] Timed out after 300.001s.
Expected success, but got an error:
<*errors.errorString | 0x4000c86ea0>:
node condition "TestCondition" not found
{
s: "node condition \"Test... | [arm64] Failure cluster [9f7f7053...] SystemLogMonitor should generate node condition and events for corresponding errors | https://api.github.com/repos/kubernetes/kubernetes/issues/120457/comments | 5 | 2023-09-06T13:02:20Z | 2023-09-08T15:11:25Z | https://github.com/kubernetes/kubernetes/issues/120457 | 1,883,972,723 | 120,457 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Steps done:
1. Create a pvc
2. Create a deployment consuming this pvc
3. Deployment pods we up and running, fio workload was running in this pod
4. Power off all the worker nodes for 20 min
5. Power on all the worker nodes
Old deployment pod was in terminating state. New deployment pod was c... | Data loss due to node stage call missing during pod start up | https://api.github.com/repos/kubernetes/kubernetes/issues/120455/comments | 12 | 2023-09-06T12:23:32Z | 2024-07-20T18:58:51Z | https://github.com/kubernetes/kubernetes/issues/120455 | 1,883,902,136 | 120,455 |
[
"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?
- `Kubernetes e2e suite.[SynchronizedBeforeSuite]`
![Screenshot_20230906_174516-1](htt... | [Failing Test] fluentd-gcp pod not scheduled due to insufficient CPU | https://api.github.com/repos/kubernetes/kubernetes/issues/120453/comments | 11 | 2023-09-06T12:14:50Z | 2023-09-08T13:34:51Z | https://github.com/kubernetes/kubernetes/issues/120453 | 1,883,887,954 | 120,453 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
As far as we read the help of "kubectl delete", it looks like that --grace-period=0 immediately kills Pod's processes.
```
--force=false:
If true, immediately remove resources from API and bypass graceful deletion. Note that immediate deletion of
some resources may result i... | "kubectl delete pod" doesn't kill pod's process immediately when specifying --grace-period=0 | https://api.github.com/repos/kubernetes/kubernetes/issues/120449/comments | 15 | 2023-09-06T08:47:21Z | 2024-05-29T17:53:34Z | https://github.com/kubernetes/kubernetes/issues/120449 | 1,883,549,688 | 120,449 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Generated serviceaccount, secret with long-lived token, and pod with this serviceaccount
enable this feature gate and found after legacy-service-account-token-clean-up-period this secret still be cleaned.
In controller manager log, could see
```
I0906 01:59:13.727397 1 legacy_serviceaccoun... | alpha feature gate: LegacyServiceAccountTokenCleanUp token referenced by pod still be cleaned | https://api.github.com/repos/kubernetes/kubernetes/issues/120447/comments | 12 | 2023-09-06T07:50:41Z | 2023-09-11T16:18:28Z | https://github.com/kubernetes/kubernetes/issues/120447 | 1,883,453,672 | 120,447 |
[
"kubernetes",
"kubernetes"
] | ### Failure cluster [b23e74354cc687fda3ee](https://go.k8s.io/triage#b23e74354cc687fda3ee)
##### Error text:
```
[FAILED] number of deallocations
Expected
<int64>: 2
to equal
<int64>: 1
In [It] at: test/e2e/dra/dra.go:650 @ 08/22/23 03:04:45.666
```
#### Recent failures:
[9/4/2023, 5:05:18 AM ci-k... | DRA: reallocation flake - failure cluster [b23e7435...] | https://api.github.com/repos/kubernetes/kubernetes/issues/120429/comments | 1 | 2023-09-05T13:15:30Z | 2023-09-11T09:56:14Z | https://github.com/kubernetes/kubernetes/issues/120429 | 1,881,962,025 | 120,429 |
[
"kubernetes",
"kubernetes"
] | ### Failure cluster [ea6679417165f10786e6](https://go.k8s.io/triage#ea6679417165f10786e6)
##### Error text:
```
error during ./hack/e2e-internal/e2e-up.sh: exit status 1
```
#### Recent failures:
[9/5/2023, 6:07:02 AM ci-kubernetes-e2e-gci-gce-alpha-enabled-default](https://prow.k8s.io/view/gs/kubernetes-jenkin... | Failure cluster [ea667941...] CI jobs that use alpha features | https://api.github.com/repos/kubernetes/kubernetes/issues/120426/comments | 4 | 2023-09-05T12:22:54Z | 2023-09-06T00:57:29Z | https://github.com/kubernetes/kubernetes/issues/120426 | 1,881,870,184 | 120,426 |
[
"kubernetes",
"kubernetes"
] | We need to explore how DRA interacts with https://github.com/kubernetes/enhancements/tree/master/keps/sig-storage/2268-non-graceful-shutdown.
One likely outcome is that the DRA helper code watches for nodes with the special `node.kubernetes.io/out-of-service=nodeshutdown:NoExecute` label and helps DRA drivers clean up... | DRA: interaction with unexpected node shutdown KEP | https://api.github.com/repos/kubernetes/kubernetes/issues/120421/comments | 8 | 2023-09-05T10:37:54Z | 2023-10-20T09:58:49Z | https://github.com/kubernetes/kubernetes/issues/120421 | 1,881,695,883 | 120,421 |
[
"kubernetes",
"kubernetes"
] | From the DRA KEP:
For kube-controller-manager, metrics similar to the generic ephemeral volume controller will be added:
- [x] Metric name: `resource_controller_create_total`
- [x] Metric name: `resource_controller_create_failures_total`
- [X] Metric name: `workqueue` with `name="resource_claim"`
We need tests... | DRA: controller metrics | https://api.github.com/repos/kubernetes/kubernetes/issues/120420/comments | 7 | 2023-09-05T10:34:43Z | 2023-09-07T09:33:59Z | https://github.com/kubernetes/kubernetes/issues/120420 | 1,881,690,853 | 120,420 |
[
"kubernetes",
"kubernetes"
] | found this by running tests on this PR:
https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/pull/120417/pull-kubernetes-typecheck/1698995947637313536
https://github.com/kubernetes/kubernetes/pull/120417
it's problematic for a couple of reasons:
- prevents kubeadm from moving out (eventually)
- there are no AP... | cmd/kube-apiserver imports cmd/kubeadm | https://api.github.com/repos/kubernetes/kubernetes/issues/120419/comments | 6 | 2023-09-05T10:21:21Z | 2024-02-20T19:50:01Z | https://github.com/kubernetes/kubernetes/issues/120419 | 1,881,669,462 | 120,419 |
[
"kubernetes",
"kubernetes"
] | - [kube-apiserver] ensure `DisableCloudProviders` is `true` (by default or explicitly)
- [kubelet] ensure `DisableKubeletCloudCredentialProviders` is `true` (by default or explicitly)
- [kubelet] Need to switch on `--image-credential-provider-config` and `--image-credential-provider-bin-dir` ( see https://github.com/... | Collect tips for external cloud provider migration | https://api.github.com/repos/kubernetes/kubernetes/issues/120411/comments | 11 | 2023-09-04T15:54:52Z | 2024-03-31T23:52:23Z | https://github.com/kubernetes/kubernetes/issues/120411 | 1,880,550,717 | 120,411 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
While debugging a problem with pods getting stuck in the unschedulable queue, I looked at the code which handles the "unschedulable_pods". It's probably not working as intended and lacks unit tests.
See https://github.com/kubernetes/kubernetes/pull/120334#discussion_r1314302936.
### What did you... | scheduler: fix and test "unschedulable_pods" metric | https://api.github.com/repos/kubernetes/kubernetes/issues/120395/comments | 10 | 2023-09-04T07:36:02Z | 2024-05-27T06:33:51Z | https://github.com/kubernetes/kubernetes/issues/120395 | 1,879,691,980 | 120,395 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Can you customize the storage directory of /var/lib/kubelet during installation? Sometimes /var/lib/kubelet is stored in the / directory by default, but the actual environment system disk and data disk are separated, such as when kubelet starts When you want to store data in /data/kubelet
### What ... | customize the storage directory of /var/lib/kubelet | https://api.github.com/repos/kubernetes/kubernetes/issues/120394/comments | 4 | 2023-09-04T05:07:48Z | 2023-09-04T05:58:43Z | https://github.com/kubernetes/kubernetes/issues/120394 | 1,879,516,817 | 120,394 |
[
"kubernetes",
"kubernetes"
] |
### Which jobs are failing?
> **pull-kubernetes-e2e-gci-gce-ipvs**
pull-kubernetes-e2e-gce-correctness
https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/pull/120389/pull-kubernetes-e2e-gce-correctness/1698401238044905472
### Which tests are failing?
> [sig-apps] ReplicaSet should serve a basi... | job constantly failing on ReplicaSet and ReplicationController tests | https://api.github.com/repos/kubernetes/kubernetes/issues/120391/comments | 8 | 2023-09-03T18:36:51Z | 2023-09-14T16:55:46Z | https://github.com/kubernetes/kubernetes/issues/120391 | 1,879,193,113 | 120,391 |
[
"kubernetes",
"kubernetes"
] | Logs look like:
```
Sep 02 00:46:52.046603 bootstrap-e2e-minion-group-15cv kubelet[10550]: E0902 00:46:52.046567 10550 run.go:74] "command failed" err="failed to run Kubelet: cannot create certificate signing request: certificatesigningrequests.certificates.k8s.io is forbidden: User \"kubelet\" cannot create resour... | External GCE cloud provider has trouble with kubelet(s) in worker nodes joining the control plane node | https://api.github.com/repos/kubernetes/kubernetes/issues/120387/comments | 3 | 2023-09-03T11:08:18Z | 2023-09-03T19:19:48Z | https://github.com/kubernetes/kubernetes/issues/120387 | 1,879,052,996 | 120,387 |
[
"kubernetes",
"kubernetes"
] | Code refactoring has been done for the dra plugin, next step could be to mutualise this code between the others plugins: csi et device | generalise GRPC plugin code and use it for CSI, Device and DRA plugins | https://api.github.com/repos/kubernetes/kubernetes/issues/120385/comments | 7 | 2023-09-03T10:06:46Z | 2024-03-27T15:24:24Z | https://github.com/kubernetes/kubernetes/issues/120385 | 1,879,035,242 | 120,385 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
I would like to add an AddGoFlags function to k8s.io/component-base/featuregate, [similar to the one found in k8s.io/component-base/logs/api/v1](https://github.com/kubernetes/kubernetes/blob/v1.28.1/staging/src/k8s.io/component-base/logs/api/v1/options.go#L308-L311).
### Why i... | Add AddGoFlags Function to k8s.io/component-base/featuregate Similar to k8s.io/component-base/logs/api/v1 | https://api.github.com/repos/kubernetes/kubernetes/issues/120384/comments | 6 | 2023-09-03T06:42:20Z | 2024-09-10T20:31:49Z | https://github.com/kubernetes/kubernetes/issues/120384 | 1,878,962,811 | 120,384 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
When I use kubeadm to create a new cluster, kube-proxy always binds ipv6 manually, which brings some problems
BUT Don't disable the ipv6 address of the network , it has other uses for me.
### What did you expect to happen?
found a way to manually disable ipv6
### How can we reproduce it ... | How to manually disable ipv6 | https://api.github.com/repos/kubernetes/kubernetes/issues/120383/comments | 10 | 2023-09-03T03:37:59Z | 2024-01-31T01:06:07Z | https://github.com/kubernetes/kubernetes/issues/120383 | 1,878,922,272 | 120,383 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
i follow this document to set k8s environment (https://github.com/kubernetes/community/blob/master/contributors/devel/development.md),
i want to use this new feature : Kubernetes 1.27: In-place Resource Resize for Kubernetes Pods (alpha) (https://kubernetes.io/blog/2023/05/12/in-place-pod-resize... | all the pods status are pending in my environment | https://api.github.com/repos/kubernetes/kubernetes/issues/120378/comments | 5 | 2023-09-02T15:03:30Z | 2023-09-02T16:03:04Z | https://github.com/kubernetes/kubernetes/issues/120378 | 1,878,723,721 | 120,378 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
kubeadm cleanup kubelet mounted paths on umount failure
logs:
```
[preflight] Running pre-flight checks
[reset] No etcd config found. Assuming external etcd
[reset] Please, manually reset etcd to prevent further issues
[reset] Stopping the kubelet service
[reset] Unmounting mounted director... | kubeadm cleanup kubelet mounted paths on umount failure | https://api.github.com/repos/kubernetes/kubernetes/issues/120376/comments | 5 | 2023-09-02T10:02:41Z | 2023-09-02T15:43:16Z | https://github.com/kubernetes/kubernetes/issues/120376 | 1,878,561,249 | 120,376 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
master-blocking:
- [x] [gce-device-plugin-gpu-master](https://testgrid.k8s.io/sig-release-master-blocking#gce-device-plugin-gpu-master)
- [x] [gce-cos-master-default](https://testgrid.k8s.io/sig-release-master-blocking#gce-cos-master-default)
- [x] [gci-gce-ingress](https://testgrid.k8... | [Failing Test] Failure in setting up e2e tests (kubetest.Up) | https://api.github.com/repos/kubernetes/kubernetes/issues/120367/comments | 7 | 2023-09-01T18:16:29Z | 2024-02-20T19:49:58Z | https://github.com/kubernetes/kubernetes/issues/120367 | 1,877,850,647 | 120,367 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
https://github.com/kubernetes/kubernetes/tree/master/cluster/addons/addon-manager
This image must be multiarchitecture:
```
% docker manifest inspect --verbose registry.k8s.io/addon-manager/kube-addon-manager:v9.1.7 | grep architecture
"architecture": "amd64",
```
Nope. Its linux/amd64
... | addon-manager image is in trouble | https://api.github.com/repos/kubernetes/kubernetes/issues/120364/comments | 4 | 2023-09-01T17:19:38Z | 2023-09-02T09:32:39Z | https://github.com/kubernetes/kubernetes/issues/120364 | 1,877,784,538 | 120,364 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are flaking?
pull-kubernetes-integration: k8s.io/kubernetes/test/integration: scheduler
### Which tests are flaking?
TestRequeueByBindFailure
```
queue_test.go:494: Expect pod-1 to be rejected by the bind plugin
```
Running with stress reproduces the flake
```
go test -c ./test... | pull-kubernetes-integration: k8s.io/kubernetes/test/integration/scheduler: TestRequeueByBindFailure | https://api.github.com/repos/kubernetes/kubernetes/issues/120363/comments | 5 | 2023-09-01T16:59:19Z | 2023-09-06T21:18:58Z | https://github.com/kubernetes/kubernetes/issues/120363 | 1,877,761,114 | 120,363 |
[
"kubernetes",
"kubernetes"
] | There are some cases when kube-proxy has to exit and restart, in order to do that we panic, but this has the problem that the log output shows a literal "panic" message and engineers, support, customers and automation gets confused, thinking that is a problem when is the expected behavior.
As suggested by @pohly in ... | Improve logging on kube-proxy exit | https://api.github.com/repos/kubernetes/kubernetes/issues/120355/comments | 3 | 2023-09-01T13:40:38Z | 2023-09-10T19:08:12Z | https://github.com/kubernetes/kubernetes/issues/120355 | 1,877,450,001 | 120,355 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
In general, clients can request `WatchBookmark` events by specifying the `allowWatchBookmarks` query parameter. However, the server is not obliged to deliver any `WatchBookmark` events, as described in https://github.com/kubernetes/enhancements/tree/master/keps/sig-api-machinery/956-watch-bookmark#p... | Inability to receive WatchBookmarks from the etcd storage implementation | https://api.github.com/repos/kubernetes/kubernetes/issues/120348/comments | 6 | 2023-09-01T09:50:20Z | 2024-09-12T20:29:35Z | https://github.com/kubernetes/kubernetes/issues/120348 | 1,877,111,534 | 120,348 |
[
"kubernetes",
"kubernetes"
] | This happens on my k8s cluster. When too many tasks are submitted (these tasks usually take up a lot of CPU and memory), many pods in the Completed state cannot be recycled. What is the reason?Is it due to resource limitations? Is there any way to fix it?
![img_v2_7484e0b4-324b-4620-acbb-f2b3257864cg](https://github.... | Many Completed pods appear | https://api.github.com/repos/kubernetes/kubernetes/issues/120343/comments | 10 | 2023-09-01T09:07:43Z | 2024-05-10T14:32:37Z | https://github.com/kubernetes/kubernetes/issues/120343 | 1,877,046,755 | 120,343 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
```
root@controller:~/cyclinder# cat << EOF | kubectl apply -f -
> apiVersion: apps/v1
kind: DaemonSet
metadata:
name: nginx
spec:
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
name: nginx
spec:
containers:
... | error message need improve: failed to create daemonSet due to invalid resources definition | https://api.github.com/repos/kubernetes/kubernetes/issues/120342/comments | 8 | 2023-09-01T09:02:37Z | 2023-09-13T16:58:26Z | https://github.com/kubernetes/kubernetes/issues/120342 | 1,877,038,037 | 120,342 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
I have a system with max-pods count set to 100 on a node.There are pods with different status like Running/NodeAffinity/Completed/Terminating etc on the system.
Observed that pods with NodeAffinity and Completed status are not considered in the pods count.
1.Is there a way to get the list of p... | When max-pods count is set on a node , how to get the list of pods (or) pod status that are considered in that limit? | https://api.github.com/repos/kubernetes/kubernetes/issues/120333/comments | 8 | 2023-09-01T06:03:12Z | 2024-01-28T12:37:37Z | https://github.com/kubernetes/kubernetes/issues/120333 | 1,876,799,371 | 120,333 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
I wrote a deliberately failing Job. It should cause a `StartError`.
The problem is that the container start time is reported as '1970-01-01T00:00:00Z' whereas the `finishedAt` time is correct.
```
{
'terminated': {
'exitCode': 128,
'reason': 'StartError',
'message': 'faile... | Job Failed StartTime is 1970? | https://api.github.com/repos/kubernetes/kubernetes/issues/120331/comments | 23 | 2023-09-01T02:07:53Z | 2024-04-03T14:45:12Z | https://github.com/kubernetes/kubernetes/issues/120331 | 1,876,594,466 | 120,331 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Certain online upgrade workflows might need to restart kubernetes as a part of their upgrade procedure. This can lead to the following sequence:
* As a part of draining a kubernetes node, a pod might get deleted.
* The unmount of a pod could take more than 6 minutes.
* If the node is deemed unhea... | Flag to Disable the 6-minute Force Detach Window | https://api.github.com/repos/kubernetes/kubernetes/issues/120328/comments | 11 | 2023-08-31T23:53:21Z | 2024-02-22T21:02:40Z | https://github.com/kubernetes/kubernetes/issues/120328 | 1,876,450,719 | 120,328 |
[
"kubernetes",
"kubernetes"
] | https://github.com/kubernetes/kubernetes/pull/118384 updated evanphx/json-patch from 4.12.0 to 5.6.0, but `v5.6.0+incompatible` is *older* than `v4.12.0+incompatible` (both were tagged on master, and the `+incompatible` bit means this is referencing the content in the root of the repo and not the content inside the `v5... | evanphx/json-patch used by 1.28 drops a fix included in 1.27 | https://api.github.com/repos/kubernetes/kubernetes/issues/120326/comments | 2 | 2023-08-31T22:43:59Z | 2024-02-20T19:49:53Z | https://github.com/kubernetes/kubernetes/issues/120326 | 1,876,406,255 | 120,326 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The overall arch:
1. admin Kubeconfig
```
- cluster:
certificate-authority-data: xxx
proxy-url: http://172.18.0.3:8088
server: https://10.96.0.1
```
2.svc in clusters
... | kubectl proxy doesnt support `proxy-url` | https://api.github.com/repos/kubernetes/kubernetes/issues/120318/comments | 16 | 2023-08-31T16:18:12Z | 2024-01-31T05:03:31Z | https://github.com/kubernetes/kubernetes/issues/120318 | 1,875,853,652 | 120,318 |
[
"kubernetes",
"kubernetes"
] | ```[tasklist]
### Tasks
- [ ] Remove `APISelfSubjectReview` feature gate in `v1.31`, update docs
- [ ] Remove alpha APIs (can happen any time since we have the GA API), update docs
- [ ] Mark beta APIs deprecated (can happen any time since we have the GA API), update docs
- [ ] Remove beta APIs in `v1.31`, update ... | SelfSubjectReview post GA work | https://api.github.com/repos/kubernetes/kubernetes/issues/120315/comments | 6 | 2023-08-31T14:45:00Z | 2024-10-16T23:53:01Z | https://github.com/kubernetes/kubernetes/issues/120315 | 1,875,684,205 | 120,315 |
[
"kubernetes",
"kubernetes"
] | I think all of the changes will be focused around allowing `v2` as an acceptable value in addition to `v2beta1` (with associated unit/integration tests):
https://github.com/kubernetes/kubernetes/blob/76b06ee239a893dece6eae861090f06df994a878/staging/src/k8s.io/apiserver/pkg/storage/value/encrypt/envelope/kmsv2/envelo... | [KMSv2] update code to mark API as GA in v1.29 | https://api.github.com/repos/kubernetes/kubernetes/issues/120309/comments | 2 | 2023-08-31T12:59:25Z | 2023-10-27T00:23:52Z | https://github.com/kubernetes/kubernetes/issues/120309 | 1,875,482,956 | 120,309 |
[
"kubernetes",
"kubernetes"
] | /sig auth
/triage accepted
/assign aramase enj ritazh
/milestone v1.29 | [KMSv2] update KEP with targeting GA in v1.29 | https://api.github.com/repos/kubernetes/kubernetes/issues/120308/comments | 0 | 2023-08-31T12:54:43Z | 2023-09-22T14:24:52Z | https://github.com/kubernetes/kubernetes/issues/120308 | 1,875,474,856 | 120,308 |
[
"kubernetes",
"kubernetes"
] | recently [the `github.com/gorilla/...` project post an update](https://gorilla.github.io/blog/2023-07-17-project-status-update/), and un-archived all the projects.
These projects have all been listed on the [`unwanted_modules.json`]([a5fff9a](https://github.com/kubernetes/kubernetes/commit/a5fff9a5d09422cb8632777d79... | unwanted_modules: github.com/gorilla/... appear active again | https://api.github.com/repos/kubernetes/kubernetes/issues/120307/comments | 7 | 2023-08-31T12:38:50Z | 2024-01-28T14:55:29Z | https://github.com/kubernetes/kubernetes/issues/120307 | 1,875,448,490 | 120,307 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
fuse process's cpu/memory also should be added into the pod and container cgroup limit.
### Why is this needed?
Glusterfs volume, for example, the fuse process's parent process is 1. So it not add into the pod or container cgroup. But when large IO into the volume, the fuse proce... | storage fuse process managed by pod/container cgroup | https://api.github.com/repos/kubernetes/kubernetes/issues/120299/comments | 9 | 2023-08-31T09:49:55Z | 2024-05-11T19:58:38Z | https://github.com/kubernetes/kubernetes/issues/120299 | 1,875,177,326 | 120,299 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
installed k8s 1.28.1,
and enable controller manager's feature gate LegacyServiceAccountTokenCleanUp as below:
- --feature-gates=LegacyServiceAccountTokenCleanUp=true
- --legacy-service-account-token-clean-up-period=1h
Then controller-manager shall be restarted, but in the log it shows... | alpha feature gate: LegacyServiceAccountTokenCleanUp failed to get configmap kube-apiserver-legacy-service-account-token-tracking | https://api.github.com/repos/kubernetes/kubernetes/issues/120298/comments | 9 | 2023-08-31T08:55:05Z | 2023-09-15T11:08:56Z | https://github.com/kubernetes/kubernetes/issues/120298 | 1,875,083,288 | 120,298 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
now, we have a cluster with 255 nodes.
when we use hpa scale a deployment with `nodeAffinity` :
```
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: biz.type
operator: In
values:
... | Pod is Unschedulable with enough resource | https://api.github.com/repos/kubernetes/kubernetes/issues/120293/comments | 12 | 2023-08-31T07:29:45Z | 2024-04-04T05:55:28Z | https://github.com/kubernetes/kubernetes/issues/120293 | 1,874,947,498 | 120,293 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
I'm starting my kubelet via
```
ExecStart=/usr/bin/kubelet --container-runtime-endpoint=unix:///var/run/containerd/containerd.sock --make-iptables-util-chains=true --cloud-provider=external
```
In a kubelet 1.28, when running `kubeadm init` if i follow `journalctl -f -u kubelet.service`
`... | We Need finer grained PLEG Error reporting on 1.28 - Kubelet not running containers, but no logging about errors (other then "PLEG"), containerd looks OK. | https://api.github.com/repos/kubernetes/kubernetes/issues/120292/comments | 14 | 2023-08-31T07:10:56Z | 2024-03-31T13:10:10Z | https://github.com/kubernetes/kubernetes/issues/120292 | 1,874,919,419 | 120,292 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added/removed?
remove [`volumesNeedReportedInUse`](https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/volumemanager/reconciler/reconciler_common.go#L148)for reconstructed volumes and [corresponding reconciler code](https://github.com/kubernetes/kubernetes/blob/master/pkg/k... | Should we remove `volumesNeedReportedInUse` for reconstructed volumes? | https://api.github.com/repos/kubernetes/kubernetes/issues/120287/comments | 8 | 2023-08-31T03:41:26Z | 2024-05-31T12:32:26Z | https://github.com/kubernetes/kubernetes/issues/120287 | 1,874,733,300 | 120,287 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
```
...
wget https://dl.k8s.io/v1.26.3/bin/linux/arm64/kubelet
...
```
```
kubeadm init --config=init.yaml
...
[ERROR KubeletVersion]: couldn't get kubelet version: cannot execute 'kubelet --version': executable file not found in $PATH
...
```
```
localhost:~/bin# ./kubelet
-as... | downloadkubernetes bin file arm64 not found | https://api.github.com/repos/kubernetes/kubernetes/issues/120283/comments | 2 | 2023-08-31T01:44:45Z | 2023-08-31T01:53:47Z | https://github.com/kubernetes/kubernetes/issues/120283 | 1,874,617,314 | 120,283 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
When using wait with the timeout flag like `kubectl wait pods -n default --all --for condition=Ready --timeout=9s -v=9` and with a kubeconfig points to rancher which uses the reverse proxy to contact the kube-api, the timeout flag is ignored.
When debugging kubectl I noticed it is being stuck at... | Kubectl wait timeout flag is ignored when going through reverse proxy | https://api.github.com/repos/kubernetes/kubernetes/issues/120282/comments | 10 | 2023-08-31T01:22:20Z | 2024-03-27T12:24:23Z | https://github.com/kubernetes/kubernetes/issues/120282 | 1,874,589,912 | 120,282 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Hello,
I am testing backup and restore feature functionality with Velero through KOTS where we are taking full volume backup.
While `restoring` from the `backup` KOTS delete all resources (to restore it from backup) related to my application including (deploys,pods,sts,pvc,pv,sc) but as kubel... | immediate Storage Class deletion causing the pods to go in Termination stage | https://api.github.com/repos/kubernetes/kubernetes/issues/120277/comments | 8 | 2023-08-30T19:30:18Z | 2024-04-27T20:37:52Z | https://github.com/kubernetes/kubernetes/issues/120277 | 1,874,251,015 | 120,277 |
[
"kubernetes",
"kubernetes"
] | on a personal whim I was curious how many imports from `go.mod` may be archived projects.
I quickly hacked up [github.com/vbatts/is-archived](https://github.com/vbatts/is-archived) to find out such a thing.
Running this on kubernetes HEAD 3cf3702d1e638548ba9fcfd957ebae43e1a573cb found 7 projects that have been arch... | deal with archived code imports | https://api.github.com/repos/kubernetes/kubernetes/issues/120276/comments | 9 | 2023-08-30T18:49:38Z | 2025-01-15T19:05:53Z | https://github.com/kubernetes/kubernetes/issues/120276 | 1,874,201,555 | 120,276 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Pods usually have a 30s grace period when they do not define a terminationGracePeriodSeconds for termination. When NodeGraceful Shutdown is enabled the grace period gets set to the ShutdownGracePeriodSeconds, which is configured by the Kubelet. Let's say the Kubelet is configured with a ShutdownGr... | Node Graceful Shutdown is taking too long | https://api.github.com/repos/kubernetes/kubernetes/issues/120271/comments | 8 | 2023-08-30T16:45:12Z | 2023-09-08T13:32:00Z | https://github.com/kubernetes/kubernetes/issues/120271 | 1,874,038,309 | 120,271 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are flaking?
All e2e suites that allow concurrent execution.
`pull-kubernetes-e2e-gce`, `pull-kubernetes-e2e-kind-kms`, etc
### Which tests are flaking?
All tests that involve cluster-scoped resources. See example run.
### Since when has it been flaking?
since https://github.com/kubernet... | AdmissionWebhook MatchConditions cause flaking tests all around suite | https://api.github.com/repos/kubernetes/kubernetes/issues/120270/comments | 4 | 2023-08-30T16:25:53Z | 2024-02-20T19:49:50Z | https://github.com/kubernetes/kubernetes/issues/120270 | 1,874,011,615 | 120,270 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
In `unmountDetachDevices` (see [here](https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/volumemanager/reconciler/reconciler_common.go#L279-L316)) we have the following code that will ultimately dispatch `NodeUnstageVolume` RPCs to the underlying CSI plugin.
Here is a pseudo-code re... | Transient Driver Registration on the CSINode can Result in Missed NodeUnstageVolume RPCs During Pod Migration | https://api.github.com/repos/kubernetes/kubernetes/issues/120268/comments | 4 | 2023-08-30T15:59:04Z | 2023-09-13T00:44:31Z | https://github.com/kubernetes/kubernetes/issues/120268 | 1,873,966,419 | 120,268 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
In a GKE cluster version 1.26.7-gke.500
I have a few deployments. on a node pool
when i delete that node pool, or even all the node pools.
one pod, that is controlled by deployment keeps its running state on a node that does not exist
how can i avoid that?
### What did you expect to happe... | pod (deployment) keeps its running state on a node that doesn't exists. | https://api.github.com/repos/kubernetes/kubernetes/issues/120264/comments | 8 | 2023-08-30T14:59:07Z | 2024-03-29T19:53:08Z | https://github.com/kubernetes/kubernetes/issues/120264 | 1,873,859,639 | 120,264 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Some Pods are stuck in NotReady status after related node becomes not ready for few seconds and then is healthy again.
From kube-controller-manager logs, pods won't be evicted and are still running.
Give one Pod for example.
```
I0828 07:00:00.435121 1 controller_utils.go:140] Updating rea... | Pods are stuck in NotReady after node becomes notReady for few seconds | https://api.github.com/repos/kubernetes/kubernetes/issues/120256/comments | 7 | 2023-08-30T09:38:46Z | 2023-09-14T09:04:23Z | https://github.com/kubernetes/kubernetes/issues/120256 | 1,873,312,432 | 120,256 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
On node reboot, we see app container is started before initContainers could complete. Essentially both app and initContainers are running in parallel. We are using initContainers to ensure dependent drivers are loaded by other pod before the app container could start. But because app container is st... | App container is started before initContainers could complete on node reboot | https://api.github.com/repos/kubernetes/kubernetes/issues/120247/comments | 18 | 2023-08-30T02:30:53Z | 2023-09-06T01:34:56Z | https://github.com/kubernetes/kubernetes/issues/120247 | 1,872,779,817 | 120,247 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
While discussing the cost exceed case raised by @padlar, @liggitt suggested to surface whether the cost was dominated by the complexity of the expression / unbounded size of the things referenced in the expression, or number of times the expression is evaluated. Those information... | CEL cost system: differentiate where the cost comes from when exceed the limit | https://api.github.com/repos/kubernetes/kubernetes/issues/120240/comments | 5 | 2023-08-29T17:26:33Z | 2024-08-29T20:17:38Z | https://github.com/kubernetes/kubernetes/issues/120240 | 1,872,102,932 | 120,240 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
Variables field is added to ValidatingAdmissionPolicy in 1.28. Currently the usage of variables would expect a operator associated.
e.g. (thanks to Mariam Fahmy for raising the issue)
```
apiVersion: admissionregistration.k8s.io/v1beta1
kind: ValidatingAdmissionPolicy
metad... | ValidatingAdmissionPolicy: variable expression type not recognized as return type in expression that evaluates the variable | https://api.github.com/repos/kubernetes/kubernetes/issues/120239/comments | 5 | 2023-08-29T16:06:24Z | 2023-10-13T23:55:44Z | https://github.com/kubernetes/kubernetes/issues/120239 | 1,871,986,761 | 120,239 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
In my k8s cluster(1.22), the `workload-high` prioritylevelconfiguration has 40 shares and the `workload-low` has 100 shares. Is this correct? I think the `workload-high` should has more shares.
```
# kubectl get prioritylevelconfiguration
NAME TYPE ASSUREDCONCURRENCYSHARES Q... | should workload-high has more shares than workload-low? | https://api.github.com/repos/kubernetes/kubernetes/issues/120238/comments | 7 | 2023-08-29T15:15:19Z | 2024-02-20T19:49:46Z | https://github.com/kubernetes/kubernetes/issues/120238 | 1,871,895,082 | 120,238 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
Add the skip return status support for PreEnqueue, and won't run PreFilter if PreEnqueue returned a Skip status
### Why is this needed?
We have skip status for PreFilter to return, and when we skip prefilter, the following filter will not run, but we don't have skip choice for Pr... | Add the skip return status support for PreEnqueue, and won't run PreFilter if PreEnqueue returned a Skip status | https://api.github.com/repos/kubernetes/kubernetes/issues/120237/comments | 13 | 2023-08-29T14:33:16Z | 2024-01-31T02:59:32Z | https://github.com/kubernetes/kubernetes/issues/120237 | 1,871,813,327 | 120,237 |
[
"kubernetes",
"kubernetes"
] | Recently, as part of go1.21's [extended backwards compatibility changes](https://go.dev/blog/compat), the [runtime/metrics](https://pkg.go.dev/runtime/metrics) packages [exposes metrics](https://pkg.go.dev/runtime/metrics#hdr-Supported_metrics) on the usage of codepaths that depend on non-default values of GODEBUG sett... | Expose metrics for the `/godebug/non-default-behavior/*` go runtime metrics | https://api.github.com/repos/kubernetes/kubernetes/issues/120236/comments | 3 | 2023-08-29T14:04:15Z | 2024-01-08T06:20:48Z | https://github.com/kubernetes/kubernetes/issues/120236 | 1,871,749,699 | 120,236 |
[
"kubernetes",
"kubernetes"
] | When we use v1.28.0 with `SidecarContainers` feature,
(restartable init container = init container with container `restartPolicy` of `Always`)
a restartable init container, in a pod with `restartPolicy` set to `Never`, restarts when its `startupProbe` fails.
This is inconsistent with the [KEP](https://github.com... | Desired behavior of a startup failed restartable init container in a pod with `restartPolicy` set to `Never` | https://api.github.com/repos/kubernetes/kubernetes/issues/120234/comments | 18 | 2023-08-29T12:25:50Z | 2025-01-18T11:55:02Z | https://github.com/kubernetes/kubernetes/issues/120234 | 1,871,562,573 | 120,234 |
[
"kubernetes",
"kubernetes"
] | ### Which jobs are failing?
when I change kubelet cpumanager from none to static and do some testing for guaranteed pod ,I found some interesting things.
### Which tests are failing?
I deploy a simple pod with requests and limits all is 2cpu, after that.
kubelet cpu_manager_state like:
```bash
{"policyName":"... | cpuset.cpu changed for Guaranteed pod when use cpumanager static policy | https://api.github.com/repos/kubernetes/kubernetes/issues/120230/comments | 16 | 2023-08-29T09:31:19Z | 2023-09-04T08:21:23Z | https://github.com/kubernetes/kubernetes/issues/120230 | 1,871,260,943 | 120,230 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
While migrating webhooks to CEL for one of the `oneof` structs in our CRD api, I noticed that the empty object in CEL is interpreted wrongly.
Empty map object seems to be interpreted correctly. In the example below, both map and object are sent empty (`{}`) in json output.
### What did you expec... | CEL: `has` macro wrongly returns false for an empty struct object | https://api.github.com/repos/kubernetes/kubernetes/issues/120226/comments | 8 | 2023-08-29T07:37:17Z | 2023-10-25T21:48:42Z | https://github.com/kubernetes/kubernetes/issues/120226 | 1,871,078,190 | 120,226 |
[
"kubernetes",
"kubernetes"
] | Drop the `apimachinery` and `client-go` dependencies from https://github.com/kubernetes/kms/blob/master/go.mod to keep the import lighter.
/sig auth
/triage accepted
| [KMSv2] Drop the apimachinery and client-go dependencies from k8s.io/kms | https://api.github.com/repos/kubernetes/kubernetes/issues/120216/comments | 7 | 2023-08-28T16:45:52Z | 2023-10-23T17:52:04Z | https://github.com/kubernetes/kubernetes/issues/120216 | 1,870,103,804 | 120,216 |
[
"kubernetes",
"kubernetes"
] | Vulnerability-
Protobuf contains an out-of-bounds read vulnerability due to a lack of user input validation. A remote attacker could exploit this by providing protobuf with a crafted message in order to cause the system to crash.
the [fix](https://github.com/protocolbuffers/protobuf-go/commit/edaf511a7a37a90db2727b... | Update google.golang.org/protobuf to v1.31.0 | https://api.github.com/repos/kubernetes/kubernetes/issues/120217/comments | 9 | 2023-08-28T16:38:05Z | 2023-08-29T12:38:27Z | https://github.com/kubernetes/kubernetes/issues/120217 | 1,870,137,079 | 120,217 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
**KubeProxyConntrackConfiguration** should support UDP conntrack settings.
**KubeProxyConntrackConfiguration** supports TCP conntrack settings and setting conntrack timeout for UDP connections is currently a manual and tiresome process.
### Why is this needed?
Currently configu... | kube-proxy conntrack configuration should support UDP timeouts | https://api.github.com/repos/kubernetes/kubernetes/issues/120214/comments | 12 | 2023-08-28T16:27:06Z | 2023-10-11T23:59:56Z | https://github.com/kubernetes/kubernetes/issues/120214 | 1,870,077,837 | 120,214 |
[
"kubernetes",
"kubernetes"
] | The default permission of the log file to which audit logs are being written is 600. The permission of the compressed log file is the same as that of the log file to which audit logs are being written. The default permission is 600.
In some scenarios, for security purposes, you are advised not to have the write perm... | Is the audit log permission configurable? | https://api.github.com/repos/kubernetes/kubernetes/issues/120203/comments | 10 | 2023-08-28T08:20:06Z | 2024-04-18T02:10:32Z | https://github.com/kubernetes/kubernetes/issues/120203 | 1,869,252,498 | 120,203 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
`kubectl get appwrappers -n test-kfp lang-id-mc2 --watch -o yaml --show-managed-fields` produced the following (edited for relevance).
```
apiVersion: mcad.ibm.com/v1beta1
kind: AppWrapper
metadata:
annotations:
edge.kcp.io/downsynced: "true"
kcp.io/cluster: 2hiptwa4c04b197k
... | ManagedFields not updated when old manager clears status | https://api.github.com/repos/kubernetes/kubernetes/issues/120201/comments | 11 | 2023-08-28T06:34:00Z | 2024-09-13T01:44:57Z | https://github.com/kubernetes/kubernetes/issues/120201 | 1,869,075,458 | 120,201 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The current tests in WinKernel Kube proxy are dependent on Windows HNS environment. The tests expect HNS to be present and are executed by making real system calls to test environment. This makes testing difficult for the developer to write make test cases as it mandates to have a real Windows Node ... | Remove the dependency of HNS in Winkernel KubeProxy Tests by replacing with mock calls. | https://api.github.com/repos/kubernetes/kubernetes/issues/120200/comments | 6 | 2023-08-28T05:50:21Z | 2023-09-08T07:32:16Z | https://github.com/kubernetes/kubernetes/issues/120200 | 1,869,018,453 | 120,200 |
[
"kubernetes",
"kubernetes"
] | Hello Folks ,
I am using Server Version: v1.24.16-eks-2d98532, created a configMap with two data point which are ssl crt and key as shown below
```
apiVersion: v1
kind: ConfigMap
metadata:
name: ssl-keys
data:
server.crt: |
-----BEGIN CERTIFICATE-----
-----END CERTIFICATE-----
server.key:... | configmap voulme not mapping to a MountPath to an airflow POD | https://api.github.com/repos/kubernetes/kubernetes/issues/120190/comments | 4 | 2023-08-26T22:45:49Z | 2023-08-29T12:09:30Z | https://github.com/kubernetes/kubernetes/issues/120190 | 1,868,288,266 | 120,190 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
During one of our tests with daemonsets, we find new pods being scheduled despite having a podAntiAffinity rule set into the daemonset.
Here's our test case;
1. We installed older version of our app.
2. We apply a finalizer to a pod( rlinux-xs84n in this case).
3. We upgrade(rolling update) ... | podAntiAffinity in a daemonset is not accepted , k8s versions > 1.27.0 | https://api.github.com/repos/kubernetes/kubernetes/issues/120189/comments | 17 | 2023-08-26T17:53:34Z | 2024-04-26T19:26:53Z | https://github.com/kubernetes/kubernetes/issues/120189 | 1,868,190,770 | 120,189 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
The following errors happen every few mins in the syslog. I'm not sure what's going on.
```
kubelet_pods.go:897] "Unable to retrieve pull secret, the image pull may not succeed." pod="kube-system/local-volume-provisioner-g4shf" secret="" err="secret \"my-app-registry\" not found"
kubelet_po... | Pods in kube-system trying and failing to retrieve an image pull secret from another namespace | https://api.github.com/repos/kubernetes/kubernetes/issues/120188/comments | 9 | 2023-08-26T13:17:21Z | 2024-04-27T20:37:52Z | https://github.com/kubernetes/kubernetes/issues/120188 | 1,868,094,527 | 120,188 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
If a Pod is evicted because it violates EmptyDir storage limits, your pod will be evicted but it will be marked as completed.
### What did you expect to happen?
I would expect the pod to say `terminating` or `Failed` rather than completed. For a Job, the controller will take this Pod ... | Evicting Pod due to emptyDir violation returns as Completed | https://api.github.com/repos/kubernetes/kubernetes/issues/120182/comments | 21 | 2023-08-25T19:17:48Z | 2023-09-21T21:49:50Z | https://github.com/kubernetes/kubernetes/issues/120182 | 1,867,571,237 | 120,182 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
We have a custom resource with a name - `Foo`
We have a CronJob named - `Foo`
Now, we have a job whose owner refernce is set to the custom resource name - `Foo`. We have noticed that the job when stopped (suspended) basically causes an event/warning - "[UnexpectedJob](https://github.com/kubern... | job executions started manually for a cronjob are marked as unexpected | https://api.github.com/repos/kubernetes/kubernetes/issues/120181/comments | 12 | 2023-08-25T17:59:15Z | 2025-02-13T00:34:39Z | https://github.com/kubernetes/kubernetes/issues/120181 | 1,867,475,791 | 120,181 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Passing passwords via the command line is known to be insecure https://www.netmeister.org/blog/passing-passwords.html
### What did you expect to happen?
There is an option to pass the password via file like `docker login --password-stdin`
### How can we reproduce it (as minimally and precisely as... | kubectl create secret docker-registry --docker-password is insecure | https://api.github.com/repos/kubernetes/kubernetes/issues/120180/comments | 11 | 2023-08-25T16:54:34Z | 2024-03-27T22:31:06Z | https://github.com/kubernetes/kubernetes/issues/120180 | 1,867,388,855 | 120,180 |
[
"kubernetes",
"kubernetes"
] | 2.8.1 is identified as affected by https://github.com/advisories/GHSA-h8c3-8522-vxc6 and https://github.com/advisories/GHSA-hqxw-f8mx-cpmw; bumping to 2.8.2 avoids k/k triggering scanners on those CVEs.
When this PR was merged, an subsequent entry for 2.8.1 was left in go.sum which needs to be removed.
https://gi... | Remove unused "github.com/docker/distribution v2.8.1+incompatible" from go.sum | https://api.github.com/repos/kubernetes/kubernetes/issues/120178/comments | 8 | 2023-08-25T16:52:29Z | 2023-08-26T00:36:31Z | https://github.com/kubernetes/kubernetes/issues/120178 | 1,867,386,385 | 120,178 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Hello Team,
I hope this message finds you well. I wanted to bring to your attention an issue related to updating StatefulSets within the Kubernetes ecosystem.
Currently, when attempting to update a StatefulSet using the `kubectl apply` command, an error message is displayed that reads as follo... | Subject: Inaccurate Error Message for StatefulSet Updates in Kubernetes | https://api.github.com/repos/kubernetes/kubernetes/issues/120171/comments | 8 | 2023-08-25T10:05:42Z | 2024-03-27T13:24:23Z | https://github.com/kubernetes/kubernetes/issues/120171 | 1,866,756,977 | 120,171 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
i want to use this new feature : Kubernetes 1.27: In-place Resource Resize for Kubernetes Pods (alpha) (https://kubernetes.io/blog/2023/05/12/in-place-pod-resize-alpha/)
i do not have this directory(~/go/src/k8s.io/kubernetes)
<img width="696" alt="image" src="https://github.com/kubernetes/kube... | how to change the InPlacePodVerticalScaling flag while the k8s is installed by kubeadm | https://api.github.com/repos/kubernetes/kubernetes/issues/120170/comments | 6 | 2023-08-25T09:37:34Z | 2023-11-27T11:57:56Z | https://github.com/kubernetes/kubernetes/issues/120170 | 1,866,707,193 | 120,170 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
when node shutdown, and kubelet had enable GracefulShutdown feature. the pods in the node is terminated after, but it is confused that pods has different containerStatuses,
There are `OOMKilled, Completed and Error` containerStatus
```
cert-manager-cainjector-cc844b5f4-dn4kb 0/1 ... | graceful shutdown make pod containerStatuses different | https://api.github.com/repos/kubernetes/kubernetes/issues/120168/comments | 5 | 2023-08-25T03:45:43Z | 2023-08-29T05:48:12Z | https://github.com/kubernetes/kubernetes/issues/120168 | 1,866,222,396 | 120,168 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
LimitRanger doesn't use a correct calculation for pod resource requests if the sidecar feature gate is enabled.
### What did you expect to happen?
LimitRanger should properly account for sidecars.
### How can we reproduce it (as minimally and precisely as possible)?
Create a pod that without the... | LimitRanger doesn't consider sidecar resource requests | https://api.github.com/repos/kubernetes/kubernetes/issues/120163/comments | 5 | 2023-08-24T19:24:01Z | 2023-08-25T11:53:00Z | https://github.com/kubernetes/kubernetes/issues/120163 | 1,865,751,681 | 120,163 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
When I specifying node names in deployment like below.
```yaml
spec:
replicas: 2
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
nodeName: node1
hostNetwork: true
containers:
- name: nginx
... | Numerous incorrect pods caused by specifying node names | https://api.github.com/repos/kubernetes/kubernetes/issues/120160/comments | 8 | 2023-08-24T15:56:31Z | 2023-09-12T06:00:47Z | https://github.com/kubernetes/kubernetes/issues/120160 | 1,865,458,645 | 120,160 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
One common use case for HPA stabilization windows is to ensure that after some autoscaling action, the system has time to bring new resources online before potentially scaling again. For example, it could take a while for pods to start, for data to migrate from old pods to new po... | Support min time between autoscaling | https://api.github.com/repos/kubernetes/kubernetes/issues/120158/comments | 8 | 2023-08-24T15:47:52Z | 2024-03-27T11:24:23Z | https://github.com/kubernetes/kubernetes/issues/120158 | 1,865,442,439 | 120,158 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
Same issue as #105395, can't reopen it.
We've observed this issue and it was causing issues when we wanted to change the version of the software deployed while also scaling out the deployment. We have some data store configuration and database schema update scripts implemented as init containe... | Old version pods are created when deploying a new version with more replicas - Reopen old issue? | https://api.github.com/repos/kubernetes/kubernetes/issues/120157/comments | 8 | 2023-08-24T14:24:28Z | 2024-03-27T02:22:23Z | https://github.com/kubernetes/kubernetes/issues/120157 | 1,865,281,937 | 120,157 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
Feedback originally provided on slack: https://kubernetes.slack.com/archives/C0EG7JC6T/p1692869878675069?thread_ts=1692653423.769159&cid=C0EG7JC6T
It would be helpful to have additional fields on APIServer spans. This is the current set of attributes we provide:
https://gith... | Add additional span attributes to APIServer tracing spans | https://api.github.com/repos/kubernetes/kubernetes/issues/120156/comments | 4 | 2023-08-24T13:13:17Z | 2023-09-21T20:07:12Z | https://github.com/kubernetes/kubernetes/issues/120156 | 1,865,132,250 | 120,156 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
During kubelet startup, it checks all volumes against the desired state (`desired_state_of_world_populator.go`). One of these steps is a `Get` request [on the PVC](https://github.com/kubernetes/kubernetes/blob/v1.28.0/pkg/kubelet/volumemanager/populator/desired_state_of_world_populator.go#L487C21-L4... | kubelet restart during intermittent network issues unmounts volumes of running pods | https://api.github.com/repos/kubernetes/kubernetes/issues/120150/comments | 13 | 2023-08-24T09:13:39Z | 2025-02-20T18:26:02Z | https://github.com/kubernetes/kubernetes/issues/120150 | 1,864,740,473 | 120,150 |
[
"kubernetes",
"kubernetes"
] | ### What would you like to be added?
We have seen huge disk space on the node but metrics such kubernetes container -> ephemeral storage usage or kubernetes pod -> ephemeral pod storage usage ( in v1.22+) were not report the correct storage usage.
This is observed in the pod with image like below with VOLUME part [... | Report ephemeral storage usage for containers with VOLUME entry defined in Dockerfile | https://api.github.com/repos/kubernetes/kubernetes/issues/120149/comments | 7 | 2023-08-24T05:25:20Z | 2024-03-27T05:24:21Z | https://github.com/kubernetes/kubernetes/issues/120149 | 1,864,429,147 | 120,149 |
[
"kubernetes",
"kubernetes"
] | ### What happened?
We restart a device plugin frequently as part of our design.
When a device plugin goes down, kubelet [marks the devices it was managing as unhealthy](https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/cm/devicemanager/manager.go#L240), removing it from allocatable, and preventing ... | Race window between scheduler and device plugin restart in kubelet | https://api.github.com/repos/kubernetes/kubernetes/issues/120146/comments | 15 | 2023-08-24T00:20:34Z | 2025-02-17T19:42:03Z | https://github.com/kubernetes/kubernetes/issues/120146 | 1,864,179,096 | 120,146 |
Subsets and Splits
Unique Owner-Repo Count
Counts the number of unique owner-repos in the dataset, providing a basic understanding of diverse repositories.