Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
fixtodo:validate events on PVCs in integration volume binding test
**What this PR does / why we need it**:
fixtodo:validate events on PVCs in integration volume binding test
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes #
**Special notes for your reviewer**:
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 62937, 63105, 63031, 63174). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Revert "Revert "Revert revert of equivalence class hash calculation i…
…n scheduler""
This reverts commit 4386751b5d.
**What this PR does / why we need it**:
This re-introduces the change from https://github.com/kubernetes/kubernetes/pull/58555 which changes how the scheduler computes equivalence classes of pods. I believe we have fixed the flakiness observed previously (https://github.com/kubernetes/kubernetes/issues/61512, https://github.com/kubernetes/kubernetes/issues/62921). I have run the test in question a few dozen times without a failure.
```bash
make test-integration WHAT="./test/integration/scheduler" KUBE_TEST_ARGS="-run TestPreemptionStarvation" GOFLAGS="-v"
```
/ref https://github.com/kubernetes/kubernetes/issues/58222
**Special notes for your reviewer**:
I had to resolve several merge conflicts. I think I resolved them correctly, but keep an eye out for anything silly.
**Release note**:
```release-note
NONE
```
/sig scheduling
Automatic merge from submit-queue (batch tested with PRs 62937, 63105, 63031, 63174). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Remove unnecessary typer from create/update handlers
**What this PR does / why we need it**:
Refactoring to remove unnecessarily duplicated definition of an ObjectTyper for some of the handlers.
The patch handler also has an extra ObjectConvertor, but it is defined differently in both places so removing one would possibly have side effects.
```release-note
NONE
```
/sig api-machinery
Automatic merge from submit-queue (batch tested with PRs 62937, 63105, 63031, 63174). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
rest mappings cannot logically be object converters
A `RESTMapping` in the general sense cannot be a `ObjectConverter` since the conversions are compiled, but the RESTMappings are discovered. This starts isolating the bad assumptions into `kubectl` where they are used and removes the other bad `RESTMapping` use I found in the REST API installer that uses a mapping to determine scopes instead of using the metadata provided during API registration.
intersection of @kubernetes/sig-api-machinery-bugs and @kubernetes/sig-cli-maintainers
@sttts @pwittrock @soltysh
```release-note
NONE
```
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Fix dockershim e2e
**What this PR does / why we need it**:
Delete checkpoint file when GetCheckpoint fails due to corrupt checkpoint. Earlier, before checkpointmanager, [`GetCheckpoint` in dockershim was deleting corrupt checkpoint file implicitly](https://github.com/kubernetes/kubernetes/pull/56040/files#diff-9a174fa21408b7faeed35309742cc631L116). In checkpointmanager's `GetCheckpoint` this implicit deletion of corrupt checkpoint is not happening. Because of this few e2e tests are failing because these tests are testing this deletion.
Changes are being added to delete checkpoint file if found corrupted.
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes#62738
**Special notes for your reviewer**:
No new behavior is being introduced. Implicit deletion of corrupt checkpoint is being done explicitly.
**Release note**:
```release-note
None
```
/cc @dashpole @sjenning @derekwaynecarr
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Change seccomp annotation from "docker/default" to "runtime/default"
**What this PR does / why we need it**:
This PR changes seccomp annotation from "docker/default" to "runtime/default", so that it is can be applied to all kinds of container runtimes. This PR is a followup of [#1963](https://github.com/kubernetes/community/pull/1963).
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes#39845
**Special notes for your reviewer**:
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
collapse patch conflict retry onto GuaranteedUpdate
xref https://github.com/kubernetes/kubernetes/issues/63104
This PR builds on https://github.com/kubernetes/kubernetes/pull/62868
1. When the incoming patch specified a resourceVersion that failed as a precondition, the patch handler would retry uselessly 5 times. This PR collapses onto GuaranteedUpdate, which immediately stops retrying in that case.
2. When the incoming patch did not specify a resourceVersion, and persisting to etcd contended with other etcd updates, the retry would try to detect patch conflicts with deltas from the first 'current object' retrieved from etcd and fail with a conflict error in that case. Given that the user did not provide any information about the starting version they expected their patch to apply to, this does not make sense, and results in arbitrary conflict errors, depending on when the patch was submitted relative to other changes made to the resource. This PR changes the patch application to be performed on the object retrieved from etcd identically on every attempt.
fixes#58017
SMP is no longer computed for CRD objects
fixes#42644
No special state is retained on the first attempt, so the patch handler correctly handles the cached storage optimistically trying with a cached object first
/assign @lavalamp
```release-note
fixed spurious "unable to find api field" errors patching custom resources
```
Automatic merge from submit-queue (batch tested with PRs 62911, 63200). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
stop duplicating preferred version order
`GroupMeta` includes two fields for a preferred groupVersion, `.GroupVersion` and `.GroupVersion[0]`. This collapses onto the latter.
@kubernetes/sig-api-machinery-pr-reviews
lots of ripples, but eliminate of duplication is good.
/assign @sttts
/assign @cheftako
```release-note
NONE
```
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
remove hardcoded list of resources
**Release note**:
```release-note
NONE
```
Removes the hardcoded list of resources in cmdutil.ValidResourceTypeList (which was not being kept up to date) and instead suggests using the `kubectl api-resources` command in order to retrieve a discovery-based list of supported resources.
I prefer this approach over updating `cmdutil.ValidResourceTypeList` to be based on discovery in order to avoid potential calls to the server while building the help output of commands.
cc @soltysh
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Remove KUBE_API_VERSIONS
Fixes https://github.com/kubernetes/kubernetes/issues/63102
KUBE_API_VERSIONS is an attempt to control the available serialization of types. It pre-dates the idea that we'll have separate schemes, so it's not a thing that makes sense anymore.
Server-side we've had a very clear message about breaks in the logs for a year "KUBE_API_VERSIONS is only for testing. Things will break.".
Client-side it became progressively more broken as we moved to generic types for CRUD more than a year ago. What is registered doesn't matter when everything is unstructured.
We should remove this piece of legacy since it doesn't behave predictable server-side or client-side.
@smarterclayton @lavalamp
@kubernetes/sig-api-machinery-bugs
```release-note
KUBE_API_VERSIONS is no longer respected. It was used for testing, but runtime-config is the proper flag to set.
```
Automatic merge from submit-queue (batch tested with PRs 63009, 63062). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
clean up unused code fakeRL in requestinfo_test.go
**What this PR does / why we need it**:
clean up unused code fakeRL in requestinfo_test.go
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes #
**Special notes for your reviewer**:
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 63009, 63062). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Register Prometheus etcdmetrics only for apiserver
Removed automatic registration with `init` funciton and use `Register` function to register metrics for etcd storage only when requested.
**What this PR does / why we need it**: Prevents leaking etcd metrics to other k8s components
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Partially #63004
**Special notes for your reviewer**:
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 63186, 63133). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
eliminate alwaysAdmit admission in apiserver test
**What this PR does / why we need it**:
alwaysAdmit is of no meaning, previously if not specify an admission plugin, it will panic, but now since we add protection `if admit != nil`. So can safely remove it.
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 63186, 63133). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Add level to remote client glog.
Set log level for remote client glog.
There is no way to disable `glog.Infof()` when `logtostderr` is specified. Because of this, `crictl logs` will always output an annoying line now:
```
# crictl logs 8a7d0379367b7
I0426 08:34:13.679995 6949 remote_runtime.go:43] Connecting to runtime service unix:///run/containerd/containerd.sock
I0426 02:20:19.389765 1 main.go:76] opts: {{/usr/sbin/dnsmasq [-k --cache-size=1000 --no-negcache --log-facility=- --server=/cluster.local/127.0.0.1#10053 --server=/in-addr.arpa/127.0.0.1#10053 --server=/ip6.arpa/127.0.0.1#10053] true} /etc/k8s/dns/dnsmasq-nanny 10000000000}
I0426 02:20:19.390103 1 nanny.go:94] Starting dnsmasq [-k --cache-size=1000 --no-negcache --log-facility=- --server=/cluster.local/127.0.0.1#10053 --server=/in-addr.arpa/127.0.0.1#10053 --server=/ip6.arpa/127.0.0.1#10053]
```
This PR sets a level for the log, so that it won't show up in `crictl logs`.
Signed-off-by: Lantao Liu <lantaol@google.com>
**Release note**:
```release-note
none
```
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
wire printflags through additional cmds
**Release note**:
```release-note
NONE
```
Adds PrintFlag pattern to more commands.
cc @deads2k @soltysh
Automatic merge from submit-queue (batch tested with PRs 59367, 60007). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Do not schedule pod to the node under PID pressure.
Signed-off-by: Da K. Ma <klaus1982.cn@gmail.com>
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
part of #54313
**Release note**:
```release-note
Added CheckNodePIDPressurePredicate to checks if a pod can be scheduled on
a node reporting pid pressure condition.
```
Automatic merge from submit-queue (batch tested with PRs 59367, 60007). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
podtolerationrestriction: fix informer race in test
**What this PR does / why we need it**: This fixes test flakes in the PodTolerationRestriction admission controller unit tests. They seem to pass most of the time currently, but modifications I was making for #58818 changed timing and caused it to constantly break.
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*: n/a
**Special notes for your reviewer**: n/a
Sending this as a one-off because the changes for both of the admin controllers in #58818 require additional discussion. Thanks to @ericchiang for finding it and authoring the commit; I just rebased and sent the PR.
```release-note
NONE
```
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Upgrade Azure Go SDK to stable version
**What this PR does / why we need it**:
Kubernetes is using a beta version of Azure Go SDK now. If there are bugs in them, it's hard to upgrade because Azure Go SDK won't release new patches for pre-released SDK versions. We should upgrade Go SDK to stable version (e.g. v14.6.0)
Refer #62249
Refer Azure/azure-sdk-for-go#1586
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes#63056
**Special notes for your reviewer**:
This PR includes changes in #61972, but with a newer go-autorest version.
**Release note**:
```release-note
Upgrade Azure Go SDK to stable version (v14.6.0)
```
builds on #62868
1. When the incoming patch specified a resourceVersion that failed as a precondition,
the patch handler would retry uselessly 5 times. This PR collapses onto GuaranteedUpdate,
which immediately stops retrying in that case.
2. When the incoming patch did not specify a resourceVersion, and persisting to etcd
contended with other etcd updates, the retry would try to detect patch conflicts with
deltas from the first 'current object' retrieved from etcd and fail with a conflict error
in that case. Given that the user did not provide any information about the starting version
they expected their patch to apply to, this does not make sense, and results in arbitrary
conflict errors, depending on when the patch was submitted relative to other changes made
to the resource. This PR changes the patch application to be performed on the object retrieved
from etcd identically on every attempt.
fixes#58017
SMP is no longer computed for CRD objects
fixes#42644
No special state is retained on the first attempt, so the patch handler correctly handles
the cached storage optimistically trying with a cached object first
Automatic merge from submit-queue (batch tested with PRs 59965, 59115, 63076, 63059). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Upgrade dep json-iterator/go to fix base64 decode bug
**What this PR does / why we need it**:
upgrade dep `json-iterator/go` to fix base64 decode bug #62742
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes#62742
**Special notes for your reviewer**:
Just upgrade `json-iterator/go` to latest which includes base64 decode fix https://github.com/json-iterator/go/pull/266
No other code changes
**Release note**:
```release-note
None
```
Automatic merge from submit-queue (batch tested with PRs 59965, 59115, 63076, 63059). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Prepull etcd before an upgrade
If kubeadm ever has to upgrade etcd it should prepull the image so
there is less downtime during the upgrade when etcd versions change.
Fixeskubernetes/kubeadm#669
Signed-off-by: Chuck Ha <ha.chuck@gmail.com>
**What this PR does / why we need it**:
This PR Prepulls the etcd image during a `kubeadm upgrade apply`.
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixeskubernetes/kubeadm#669
**Special notes for your reviewer**:
constants.MasterComponents was not changed because it is used in many places where etcd does not need to be nor should it be a part of this slice.
**Release note**:
```release-note
NONE
```
/cc @kubernetes/sig-cluster-lifecycle-pr-reviews
Automatic merge from submit-queue (batch tested with PRs 59965, 59115, 63076, 63059). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
kubeadm: prompt for confirmation when resetting a master
Signed-off-by: Alexander Brand <alexbrand09@gmail.com>
**What this PR does / why we need it**:
This PR implements a confirmation prompt when running `kubeadm reset` on a master node. This is to prevent users from mistakenly resetting a master node.
**Which issue(s) this PR fixes**:
Fixes https://github.com/kubernetes/kubeadm/issues/673
**Special notes for your reviewer**:
I am somewhat torn on the approach on how to detect that kubeadm is running on a master node. I went with checking for the apiserver manfiest file on the local filesystem, as it seems like a simpler approach when compared to getting a k8s client, getting a list of nodes, finding the current node, and checking if it has the master taint. I am happy to rework if the latter is more desirable.
Sample runs:
```
# ./kubeadm reset
[Warning] Are you sure you want to reset this master node? Type the word "confirm" to continue: no
Aborted reset operation on master node
# ./kubeadm reset
[Warning] Are you sure you want to reset this master node? Type the word "confirm" to continue: confirm
[preflight] Running pre-flight checks.
[reset] Stopping the kubelet service.
[reset] WARNING: The kubelet service could not be stopped by kubeadm: [exit status 1]
[reset] WARNING: Please ensure kubelet is stopped manually.
[reset] Unmounting mounted directories in "/var/lib/kubelet"
........
# ./kubeadm reset
[Warning] Are you sure you want to reset this master node? Type the word "confirm" to continue:
Aborted reset operation on master node
# ./kubeadm reset --confirm
[preflight] Running pre-flight checks.
[reset] Stopping the kubelet service.
[reset] WARNING: The kubelet service could not be stopped by kubeadm: [exit status 1]
[reset] WARNING: Please ensure kubelet is stopped manually.
[reset] Unmounting mounted directories in "/var/lib/kubelet"
........
```
**Release note**:
```release-note
kubeadm: prompt the user for confirmation when resetting a master node
```
Automatic merge from submit-queue (batch tested with PRs 59965, 59115, 63076, 63059). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
fix help message of kubeconfig-dir option(kubeadm alpha phase kubeconfig all)
**What this PR does / why we need it**:
This patch will fix wrong help message.
The command is kubeadm alpha phase kubeconfig [command]
The help message is for --kubeconfig-dir option.
kubeconfig-dir is not port.(It is directory)
So, I fixed the message.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #
**Special notes for your reviewer**:
**Release note**:
`NONE`
/sig cluster-lifecycle
Pengfei Ni