Automatic merge from submit-queue
Add a new API version apps/v1beta2
xref: #49135
This PR adds a new API version `apps/v1beta2` which contains a copy (of types, conversions, and defaults) of `apps/v1beta1` StatefulSet, Deployment, and their subresources. Note that `apps/v1beta2` is still WIP and we will make breaking changes to it before releasing 1.8.
Moving core controllers (StatefulSet, Deployment, ReplicaSet, DaemonSet) to `apps/v1beta2` is the first step of moving them to `apps/v1` (GA).
This PR is a starting point for DaemonSet and ReplicaSet to move from `/extensions` to `/apps` and for Deployment and StatefulSet to make some breaking changes (e.g. new defaults and/or remove deprecated fields).
```release-note
Add a new API version apps/v1beta2
```
Automatic merge from submit-queue
Use specified ServerName in aggregator TLS validation
Fixes#49354
The aggregator sets a ServerName in the proxier tlsConfig, but the code path handling websocket upgrade requests did not honor it, and instead tried to verify TLS using the dialed host
* Honors ServerName if already set in tls.Config
* Adds unit tests for upgrade functionality via the aggregator
* Fixes mutation of shared tlsConfig.ServerName in spdy roundtripper
```release-note
Websocket requests to aggregated APIs now perform TLS verification using the service DNS name instead of the backend server's IP address, consistent with non-websocket requests.
```
Automatic merge from submit-queue
Reduce hollow proxy mem/node
As likely expected, kubemark-scale failed to even start with n1-standard-8 nodes. Because 1/3rd of our hollow nodes didn't even get scheduled due to their requests:
```
I0720 17:45:08.139] Found only 3325 ready hollow-nodes while waiting for 5000.
I0720 17:45:20.435] 3326 hollow-nodes are reported as 'Running'
I0720 17:45:20.442] 1675 hollow-nodes are reported as NOT 'Running'
```
If we want to experiment with smaller nodes anyway, then this change is needed. Though we most likely will end up OOM'ing.
Explanation for new value:
We have 62.5 hollow-node / real-node
=> mem available per hollow node = 30GB / 62.5 = 480MB
minus 100MB (kubelet)
minus 20MB (npd)
=> 360MB for proxy should be = 100MB + 5000*(mem/node)
=> 50KB mem/node (with some slight slack)
cc @kubernetes/sig-scalability-misc
Automatic merge from submit-queue
Add an integration test library and some integration tests for scheduler
**What this PR does / why we need it**:
1. Add an integration test library (utils.go) for scheduler testing.
2. Cleaned up some of the tests in scheduler_test.go with the new integration test library.
3. Add priority_test.go with a couple of examples on how to test scheduler priority function in integration tests.
**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
```
ref/ #48176
@kubernetes/sig-scheduling-pr-reviews
@davidopp @k82cn @vikaschoudhary16
Automatic merge from submit-queue
Auto-calculate master disk and root disk sizes in GCE
@gmarek PR https://github.com/kubernetes/kubernetes/pull/49282 didn't fix the issue because MASTER_DISK_SIZE was defaulting to 20GB in config-test.sh before being calculated inside get-master-disk-size() where you use pre-existing value if any.
It should be fixed by this now.
Automatic merge from submit-queue
word spell error
**What this PR does / why we need 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 49328, 49285, 49307, 49127, 49163)
Cleanup storage e2e test names
**What this PR does / why we need it**:
Some test names had redundant [sig-storage] tags. Also, some tests still had [Volume] tag. This PR removes those tags.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #
**Release note**:
```release-note
NONE
```
/release-note-none
/sig storage
Automatic merge from submit-queue (batch tested with PRs 49328, 49285, 49307, 49127, 49163)
Make definite mount timeout for glusterfs volume mount.
The `backup-volfile-servers` mount option allows to specify more than
one server to be contacted in single mount command. With this option in place,
it is not required to iterate over all the servers in the addrlist.
A mount attempt with this option will fetch all the servers mentioned in
the list, Reference # https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/sect-Native_Client.html
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
Automatic merge from submit-queue (batch tested with PRs 49328, 49285, 49307, 49127, 49163)
fix mutation in statefulset sync
Original issue: https://github.com/openshift/origin/issues/15324
Seems like something inside `UpdateStatefulSet` mutates the StatefulSet. We might need to dig deeper to find the exact location where the non-copy mutation happens.
Automatic merge from submit-queue (batch tested with PRs 49328, 49285, 49307, 49127, 49163)
kubeadm: don't customize etcd selinux label
The original change that added the unconfined label included a comment
indicating it won't be needed in the future.
See: https://github.com/kubernetes/kubernetes/pull/33555#issuecomment-251126908
That time is now. https://github.com/kubernetes/kubernetes/pull/33663
has landed and means we no longer have to go out of our way to make that
work.
Removing the label also increases security since there wasn't really a
good reason for etcd to be run with such broad selinux privileges.
This also will allow kubeadm to avoid errors on distros without an spc_t
type, such as Gentoo and Container Linux (at the time of writing at
least).
Fixes https://github.com/kubernetes/kubeadm/issues/269
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 48565, 49172)
On GCE check whether NODE_LOCAL_SSDS=0 and handle this case appropriately
**What this PR does / why we need it**: Presently if you are using a mac and GCE and specify NODE_LOCAL_SSDS=0, or use the default, you end up with 2 local SSDs.
**Which issue this PR fixes** : fixes https://github.com/kubernetes/kubernetes/issues/49171
**Special notes for your reviewer**:
I've discovered that this issue is due to b353792f9c/cluster/gce/util.sh (L579)
If NODE_LOCAL_SSDS=0, this evaluates to $(seq 0)
```
$ for i in $(seq 0); do echo $i; done
1
0
```
From man seq on mac osx
```
The seq utility prints a sequence of numbers, one per line (default), from first (default 1),
to near last as possible, in increments of incr (default 1).When first is larger than last the
default incr is -1.
```
This was run on mac with the seq manpage indicating it comes from BSD Feb 19 2010.
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 48565, 49172)
Use $(location) to find generated output paths.
Along with the build-infra PR https://github.com/kubernetes/repo-infra/pull/19, this allows Kubernetes binaries (e.g. hyperkube) to be built as external dependencies in other Bazel projects.
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 49222, 49333, 48708, 49337)
Fix issue in installing containerized mounter
Fix PR #49335
PR #49157 causes failure when installing containerized mounter. This
PR is a fix for it
Automatic merge from submit-queue (batch tested with PRs 49222, 49333, 48708, 49337)
Allow to override build date
See https://reproducible-builds.org/ for why this is good
and https://reproducible-builds.org/specs/source-date-epoch/
for the definition of this variable.
Background: issue #48710
when trying to build the kubernetes package
for the openSUSE Linux distribution
```
/usr/bin/hyperkube
/usr/bin/kubeadm
/usr/src/kubernetes/platforms/linux/amd64/e2e.test
```
had a date+time string embedded in them in the format
2018-08-15T19:08:34Z
which meant that each build produced new binaries
even when nothing actually changed.
Those then had to be synced to mirrors, wasting their and our users' bandwidth.
Automatic merge from submit-queue (batch tested with PRs 49222, 49333, 48708, 49337)
glbc: change the label of the l7-lb-controller pod
This ensures that the default http backend service doesn't include this
pod as its endpoint. This fixes#49159
Automatic merge from submit-queue (batch tested with PRs 49330, 49252, 49262, 49278, 49334)
Add project to pd delete node gcloud command
**What this PR does / why we need it**: Add `--project=` to `gcloud compute instances list` calls from `Pod Disks should be able to detach from a node which was deleted`.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: Fixes #https://github.com/kubernetes/kubernetes/issues/49185
**Special notes for your reviewer**:
CC @kubernetes/sig-storage-pr-reviews
Automatic merge from submit-queue (batch tested with PRs 49330, 49252, 49262, 49278, 49334)
Simplify master-worker relation missing message
**What this PR does / why we need it**: Simplify messaging of missing relation in Juju deployments
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes https://github.com/juju-solutions/bundle-canonical-kubernetes/issues/309
**Special notes for your reviewer**:
**Release note**:
```NONE
```
Automatic merge from submit-queue (batch tested with PRs 49330, 49252, 49262, 49278, 49334)
Enable garbage collector e2e tests
These tests are not running in pre-submit: see 753266cb7d/jobs/config.json (L9207)
Automatic merge from submit-queue
Bump up gce minNodesHealthCheckVersion due to known issues
**What this PR does / why we need it**: There are some known issues in previous 1.7 versions causing kube-proxy not correctly responding healthz traffic.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: From #49263.
**Special notes for your reviewer**:
/assign @nicksardo @freehan
cc @bowei @thockin
**Release note**:
```release-note
GCE Cloud Provider: New created LoadBalancer type Service will have health checks for nodes by default if all nodes have version >= v1.7.2.
```
informerFactory has been started already but the
factory.equivalencePodCache field has not been initialized yet. So when
Service informer calls onServiceAdd(), it tries to access this field
and crashes with npe.
The fix is to use informerFactory2.
panic: runtime error: invalid memory address or nil pointer dereference [recovered]
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x14c2c2f]
goroutine 11688 [running]:
k8s.io/kubernetes/vendor/k8s.io/apimachinery/pkg/util/runtime.HandleCrash(0x0, 0x0, 0x0)
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/vendor/k8s.io/apimachinery/pkg/util/runtime/runtime.go:58 +0x126
panic(0x28f5f00, 0x61621c0)
/usr/local/go/src/runtime/panic.go:489 +0x2cf
k8s.io/kubernetes/plugin/pkg/scheduler/core.(*EquivalenceCache).InvalidateCachedPredicateItemOfAllNodes(0x0, 0xc4202a9020)
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/plugin/pkg/scheduler/core/equivalence_cache.go:141 +0x5f
k8s.io/kubernetes/plugin/pkg/scheduler/factory.(*ConfigFactory).onServiceAdd(0xc424ad5950, 0x2cf6fc0, 0xc4249c05a0)
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/plugin/pkg/scheduler/factory/factory.go:347 +0x53
k8s.io/kubernetes/plugin/pkg/scheduler/factory.(*ConfigFactory).(k8s.io/kubernetes/plugin/pkg/scheduler/factory.onServiceAdd)-fm(0x2cf6fc0, 0xc4249c05a0)
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/plugin/pkg/scheduler/factory/factory.go:247 +0x3e
k8s.io/kubernetes/vendor/k8s.io/client-go/tools/cache.ResourceEventHandlerFuncs.OnAdd(0xc423a47310, 0xc423a47320, 0xc423a47330, 0x2cf6fc0, 0xc4249c05a0)
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/vendor/k8s.io/client-go/tools/cache/controller.go:195 +0x49
k8s.io/kubernetes/vendor/k8s.io/client-go/tools/cache.(*ResourceEventHandlerFuncs).OnAdd(0xc423938ce0, 0x2cf6fc0, 0xc4249c05a0)
<autogenerated>:54 +0x73
k8s.io/kubernetes/vendor/k8s.io/client-go/tools/cache.(*processorListener).run(0xc424a698b0)
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/vendor/k8s.io/client-go/tools/cache/shared_informer.go:544 +0x287
k8s.io/kubernetes/vendor/k8s.io/client-go/tools/cache.(*processorListener).(k8s.io/kubernetes/vendor/k8s.io/client-go/tools/cache.run)-fm()
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/vendor/k8s.io/client-go/tools/cache/shared_informer.go:380 +0x2a
k8s.io/kubernetes/vendor/k8s.io/apimachinery/pkg/util/wait.(*Group).Start.func1(0xc424a87868, 0xc423a47370)
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:71 +0x4f
created by k8s.io/kubernetes/vendor/k8s.io/apimachinery/pkg/util/wait.(*Group).Start
/go/src/k8s.io/kubernetes/_output/local/go/src/k8s.io/kubernetes/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:72 +0x62
Automatic merge from submit-queue
remove redundant param in e2e_node/remote
**What this PR does / why we need it**:
* remove redundant param in e2e_node/remote/remote.go
* fix a small typo
**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
Remove redundent bug-fix item in CHANGELOG
**What this PR does / why we need it**:
Just remove redundent bug-fix items in CHANGELOG.
Automatic merge from submit-queue (batch tested with PRs 49316, 46117, 49064, 48073, 49323)
Enable RBAC by default in hack/local-up-cluster.sh
**What this PR does / why we need it**:
Since hack/local-up-cluster.sh checks https API port after commit 413ab26df9, RBAC should be enabled by default to avoid TLS port access denied.
**Which issue this PR fixes** : Fixes#49319
**Special notes for your reviewer**:
@liggitt
**Release note**:
```release-note
hack/local-up-cluster.sh now enables RBAC authorization by default
```
Automatic merge from submit-queue (batch tested with PRs 49316, 46117, 49064, 48073, 49323)
add e2e tests for the bootstrapsigner and tokencleaner controllers, integration testing for bootstrap token auth
**What this PR does / why we need it**:
Add e2e test for bootstrap signer
**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
```
Automatic merge from submit-queue (batch tested with PRs 49316, 46117, 49064, 48073, 49323)
Test Ubuntu image using GKE image spec on master
Ref: https://github.com/kubernetes/kubernetes/issues/46891
This PR changes the files referenced in test-infra for running Ubuntu image tests against GKE system spec on master.
The two properties files are shared by the tests against all k8s branches but the `SYSTEM_SPEC_NAME` is only available on master. This should be fine because the tests in the non master branches will just ignore the unknown env variable.
**Release note**:
```
None
```
/assign @yujuhong
Automatic merge from submit-queue (batch tested with PRs 49316, 46117, 49064, 48073, 49323)
Modular extensions for kube scheduler perf testing framework
**What this PR does / why we need it**:
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#45973
**Special notes for your reviewer**:
It is not same as the existing one, the previous one has a single nodeaffinity key with multiple values. This one has multiple keys, values.
**Release note**:
```
NONE
```
Automatic merge from submit-queue
Use custom port for node-problem-detector
It fixes https://github.com/kubernetes/kubernetes/issues/49263
```release-note
Use port 20256 for node-problem-detector in standalone mode.
```
