This change adds the `--discovery-token-ca-cert-hash` and `--discovery-token-unsafe-skip-ca-verification` flags for `kubeadm join` and corresponding fields on the kubeadm NodeConfiguration struct. These flags configure enhanced TLS validation for token-based discovery.
The enhanced TLS validation works by pinning the public key hashes of the cluster CA. This is done by connecting to the `cluster-info` endpoint initially using an unvalidated/unsafe TLS connection. After the cluster info has been loaded, parsed, and validated with the existing symmetric signature/MAC scheme, the root CA is validated against the pinned public key set. A second request is made using validated/safe TLS using the newly-known CA and the result is validated to make sure the same `cluster-info` was returned from both requests.
This validation prevents a class of attacks where a leaked bootstrap token (such as from a compromised worker node) allows an attacker to impersonate the API server.
This change also update `kubeadm init` to print the correct `--discovery-token-ca-cert-hash` flag in the example `kubeadm join` command it prints at the end of initialization.
This change adds a `k8s.io/kubernetes/cmd/kubeadm/app/util/pubkeypin` package which implements x509 public key pinning in the style of RFC7469. This is the public key hash format used by the new `kubeadm join --discovery-token-ca-cert-hash` flag.
Hashes are namespaced with a short type, with "sha256" being the only currently-supported format. Type "sha256" is a hex-encoded SHA-256 hash over the Subject Public Key Info (SPKI) object in DER-encoded ASN.1.
Automatic merge from submit-queue
Add Cluster Autoscaler scalability test suite
This suite is intended for manually testing Cluster Autoscaler on large clusters. It isn't supposed to be run automatically (at least for now).
It can be run on Kubemark (with #50440) with the following setup:
- start Kubemark with NUM_NODES=1 (as we require there to be exactly 1 replica per hollow-node replication controller in this setup)
- set kubemark-master machine type manually to appropriate type for the Kubemark cluster size. Maximum Kubemark cluster size reached in test run is defined by maxNodes constant, so for maxNodes=1000, please upgrade to n1-standard-32. Adjust if modifying maxNodes.
- start Cluster Autoscaler pod in the external cluster using image built from version with Kubemark cloud provider (release pending)
- for grabbing metrics from ClusterAutoscaler (with #50382), add "--include-cluster-autoscaler=true" parameter in addition to regular flags for gathering components' metrics/resource usage during e2e tests
cc @bskiba
Automatic merge from submit-queue
Make ClusterID required for AWS.
**What this PR does / why we need it**:
Makes ClusterID required for AWS and provides a flag to run in un-tagged mode
fixes#48954
**Release note**:
```release-note
A cluster using the AWS cloud provider will need to label existing nodes and resources with a ClusterID or the kube-controller-manager will not start. To run without a ClusterID pass --allow-untagged-cloud=true to the kube-controller-manager on startup.
```
Automatic merge from submit-queue (batch tested with PRs 45186, 50440)
Add functionality needed by Cluster Autoscaler to Kubemark Provider.
Make adding nodes asynchronous. Add method for getting target
size of node group. Add method for getting node group for node.
Factor out some common code.
**Release note**:
```
NONE
```
Automatic merge from submit-queue (batch tested with PRs 45186, 50440)
Retry fed-svc creation on diff NodePort during e2e tests
**What this PR does / why we need it**:
Currently in federated end2end tests, the creation of services are
done with a randomize NodePort selection take is causing e2e test
flakes if the creation of a federated service failed if the port is
not available.
Now the util.CreateService(...) function is retrying to create the
service on different nodePort in case of error. The method retry until
success or all possible NodePorts have been tested and also failed.
**Which issue this PR fixes**
fixes#44018
Automatic merge from submit-queue (batch tested with PRs 50306, 49624)
Add daemonset to all categories
**What this PR does / why we need it**:
We could get daemonset resource by running command `kubectl get all`.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #
fix https://github.com/kubernetes/kubernetes/issues/49620
Automatic merge from submit-queue (batch tested with PRs 50306, 49624)
simplify logic around LB deletion for servicecontroller
**What this PR does / why we need it**: simplify logic around LB deletio
**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**: discovered when aswering to #50299
**Release note**:
```release-note
None
```
Make adding nodes asynchronous. Add method for getting target
size of node group. Add method for getting node group for node.
Factor out some common code.
Automatic merge from submit-queue (batch tested with PRs 50386, 50374, 50444, 50382)
Add grabbing Cluster Autoscaler metrics in e2e tests
This adds:
- collecting metrics from Cluster Autoscaler before & after e2e test run
- --include-cluster-autoscaler opt-in flag
- passing external cluster client to MetricsGrabber (required for Kubemark setup, as Cluster Autoscaler doesn't run on master in this case)
Automatic merge from submit-queue (batch tested with PRs 50386, 50374, 50444, 50382)
jsonpath: fix comments
Minor fix to the comments. And avoid a named return value.
**Release note**:
```release-note
NONE
```
/cc @sttts
Automatic merge from submit-queue (batch tested with PRs 50386, 50374, 50444, 50382)
wires ban flunder admission plugin to the sample server
**What this PR does / why we need it**:
this PR wires ban flunder admission plugin to the sample server.
**Release note**:
```
NONE
```
Automatic merge from submit-queue (batch tested with PRs 50386, 50374, 50444, 50382)
Add explicit API kind and version to the audit policy file on GCE
Adds an explicit API version and kind to the audit policy file in GCE configuration scripts. It's a prerequisite for https://github.com/kubernetes/kubernetes/pull/49115
/cc @tallclair @piosz
Most types now have valid rest mappings because
NewDefaultRESTMapperFromScheme no longer ignores certain import
paths. Thus we can no longer use the lack of a valid REST mapping
as an indicator for when to use kindWhiteList. Thus kindWhiteList
now serves as a whitelist for all kinds and not just those that
formally had no mapping. This does mean that we could whitelist
kinds due to a name conflict, but that is unlikely as names such as
GetOptions are not appropriate for new objects.
Signed-off-by: Monis Khan <mkhan@redhat.com>
Automatic merge from submit-queue
Update mrubin to matchstick in OWNERS
**What this PR does / why we need it**:
per https://github.com/kubernetes/kubernetes/issues/50048#issuecomment-320000920
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #
associated with #50048
**Special notes for your reviewer**:
/assign @matchstick
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue
[Federation] Implement leader election for controller-manager
**What this PR does / why we need it**:
- Add cluster-selector for namespace
- Add support for creating federation-only objects. - Ref #44631
- Implements leader election for controller-manager.
Ref: #44283
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#44490
**Special notes for your reviewer**:
This PR also fixes the issue #44490, which is about delay in initializing controller-manager due to unavailability of api-server.
**Release note**:
```release-note
federation: Support for leader-election among federation controller-manager instances introduced.
```
/cc @kubernetes/sig-federation-pr-reviews
Automatic merge from submit-queue
fix error message for scale
**What this PR does / why we need it**:
ref: https://github.com/kubernetes/kubernetes/blob/master/cmd/genutils/genutils_test.go#L33
**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**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 49725, 50367, 50391, 48857, 50181)
Use 'Infof' instead of 'Errorf' for a debug log
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #
#50167
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 49725, 50367, 50391, 48857, 50181)
Add e2e test for privileged containers
**What this PR does / why we need it**:
This PR adds node e2e test for privileged containers.
**Which issue this PR fixes**
Part of #44118.
**Special notes for your reviewer**:
**Release note**:
```release-note
NONE
```
/assign @Random-Liu
Automatic merge from submit-queue (batch tested with PRs 49725, 50367, 50391, 48857, 50181)
New get-kube.sh option: KUBERNETES_SKIP_RELEASE_VALIDATION
**What this PR does / why we need it**:
This is an alternative solution to https://github.com/kubernetes/kubernetes/pull/49884. The goal is to be able to pull releases that were built by bazel jobs (both presubmit and postsubmit builds), which currently fail our regex validation against the version string.
This implementation is a simple "I know what I'm doing" breakglass option to turn regex validation off, whereas https://github.com/kubernetes/kubernetes/pull/49884 was to extend our validation to support the new formats of bazel build jobs. I'm testing the waters to see if this is a more palatable solution.
**Release note**:
```release-note
New get-kube.sh option: KUBERNETES_SKIP_RELEASE_VALIDATION
```
CC @BenTheElder @fejta @ixdy
Automatic merge from submit-queue (batch tested with PRs 49725, 50367, 50391, 48857, 50181)
Don't call one of pointless conversions
@kubernetes/sig-federation-pr-reviews
Automatic merge from submit-queue
Azure: Allow VNet to be in a separate Resource Group
**What this PR does / why we need it**:
This PR allows Kubernetes in an Azure context to use a VNet which is not in the same Resource Group as Kubernetes.
We need this because currently Azure Cloud Provider driver assumes that it should have a VNet for himself but if there is one thing that should be shared amongst Azure resources it's a VNet cause, well, things might want to talk to each other in a private network, don't you think ?
I guess this should we backported down to 1.6 branch.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*:
fixes#49577
**Release note**:
```release-note
NONE
```
@kubernetes/sig-azure
@kubernetes/sig-azure-pr-reviews
Automatic merge from submit-queue (batch tested with PRs 49642, 50335, 50390, 49283, 46582)
Admit sysctls for other runtime.
Fixes https://github.com/kubernetes/kubernetes/issues/50343.
Admit sysctl for other runtimes.
/cc @mikebrow @yujuhong @feiskyer @sttts
Automatic merge from submit-queue (batch tested with PRs 49642, 50335, 50390, 49283, 46582)
Improve GC discovery sync performance
Improve GC discovery sync performance by only syncing when discovered
resource diffs are detected. Before, the GC worker pool was shut down
and monitors resynced unconditionally every sync period, leading to
significant processing delays causing test flakes where otherwise
reasonable GC timeouts were being exceeded.
Related to https://github.com/kubernetes/kubernetes/issues/49966.
/cc @kubernetes/sig-api-machinery-bugs
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 49642, 50335, 50390, 49283, 46582)
Add rbac.authorization.k8s.io/v1
xref https://github.com/kubernetes/features/issues/2
Promotes the rbac.authorization.k8s.io/v1beta1 API to v1 with no changes
```release-note
The `rbac.authorization.k8s.io/v1beta1` API has been promoted to `rbac.authorization.k8s.io/v1` with no changes.
The `rbac.authorization.k8s.io/v1alpha1` version is deprecated and will be removed in a future release.
```
Automatic merge from submit-queue (batch tested with PRs 49615, 49321, 49982, 49788, 50355)
Simplify hack/verify-flags-underscore.py
**What this PR does / why we need it**:
This PR removes the need for `hack/verify-flags/known-flags.txt` and verify-flags-underscore.py will always parse the flags from go files to check if they have underscore.
It is much faster compared to earlier checks and it does its job to check for underscore in flags.
Now:
```
# time ./hack/verify-flags-underscore.py
real 0m1.638s
user 0m1.560s
sys 0m0.076s
```
Before:
```
# time ./hack/verify-flags-underscore.py
real 0m22.585s
user 0m22.464s
sys 0m0.112s
```
It has become a pain to keep adding new flag to `known-flags.txt` whenever a new flag is introduced. with this PR this is step is not required anymore.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#40329#50319
**Special notes for your reviewer**:
**Release note**:
```
NONE
```
/cc @fejta @mtaufen
Automatic merge from submit-queue (batch tested with PRs 49615, 49321, 49982, 49788, 50355)
csr: add resync to csr approver
fixes https://github.com/kubernetes/kubernetes/issues/49787
```release-note
Fix an issue where if a CSR is not approved initially by the SAR approver is not retried.
```
Automatic merge from submit-queue (batch tested with PRs 49615, 49321, 49982, 49788, 50355)
kubeadm: Move all node bootstrap token related code in one phase package
**What this PR does / why we need it**:
Part of the phases refactoring.
Moves everything Node Bootstrap Token-related into its own package.
In the future there will be a `phases/bootstraptoken/master` pkg as well.
The generic bootstrap token client functions should be moved to client go eventually https://github.com/kubernetes/client-go/issues/114
**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**:
I'll yet add the CLI interface for this tomorrow.
Not sure if this compiles currently, but I'm uploading this now for initial review.
**Release note**:
```release-note
NONE
```
@kubernetes/sig-cluster-lifecycle-pr-reviews @mattmoyer
Automatic merge from submit-queue (batch tested with PRs 49615, 49321, 49982, 49788, 50355)
Fix unused Secret export logic.
**What this PR does / why we need it**:
The strategy used for the secret store defined custom export logic, and
had accompanying unit tests. However the secret storage did not actually
wire this up by setting an ExportStrategy and thus the code was never
used in the real world.
This change fixes the missing assignment and adds testing at a higher
level to ensure any uses of the generic registry.Store that we expect to
have an ExportStrategy do, and no others.
Several other strategies in the RBAC package also appeared to have
unwired Export logic, however their implementations were all empty
leading me to believe that these are not considered exportable. The
empty methods have now been removed.
**Which issue this PR fixes**: fixes#49042
**Release note**:
```release-note
```
Automatic merge from submit-queue (batch tested with PRs 49615, 49321, 49982, 49788, 50355)
remove redundant comment
**What this PR does / why we need it**:
there has redundant comment and delete it!
**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**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 50300, 50328, 50368, 50370, 50372)
Bugfix: set resources only for fluentd-gcp container.
There is more than one container in fluentd-gcp deployment. Previous
implementation was setting resources for all containers, not just
the fluent-gcp one.
**What this PR does / why we need it**:
Bugfix; https://github.com/kubernetes/kubernetes/pull/49009 without this is eating more resources.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#50366
**Special notes for your reviewer**:
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 50300, 50328, 50368, 50370, 50372)
correct the allocated element number of pod selectable field set
a little fix to correct the allocated element number of pod selectable field set
Automatic merge from submit-queue (batch tested with PRs 50300, 50328, 50368, 50370, 50372)
Reduce hollow-kubelet cpu request
Fixes https://github.com/kubernetes/kubernetes/issues/50366
This should make kubemark-500 fit in 6 nodes again. Checked that it should be enough.
cc @kubernetes/sig-scalability-misc
Automatic merge from submit-queue (batch tested with PRs 50300, 50328, 50368, 50370, 50372)
Add unit tests for GenerateLink
Fixes https://github.com/kubernetes/kubernetes/issues/48321
cc @lavalamp
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 50300, 50328, 50368, 50370, 50372)
Clean validation_test go file
When i wrote test cases for feature gate of local storage quota, found some unused vars and useless code, remove them
**Release note**:
```release-note
NONE
```
