Automatic merge from submit-queue (batch tested with PRs 57550, 60089). 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 subnet size restriction for IPv6
RangeSize was restricting IPv6 subnets to a /66 due to the
logic using a uint64. This is not practical for IPv6.
This change removes the /64 restriction, but also sets a limit
on the range that can be allocated, so that the bitmap will not grow too large.
**What this PR does / why we need it**:
This PR removes the /66 restriction in ipallocator for IPv6. It is not practical to restrict
IPv6 to /66. Currently a /64 or /48 is not allowed. The problem with removing the restriction is
the bitmap that tracks the subnets can grow really large, so a limit
on the max size of the subnet was set to 65536.
Setting the max size will have a side-effect with larger subnets that the allocator
will allocate in a smaller section of IP's, this will need to be addressed in a follow-on PR.
**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#60081
**Special notes for your reviewer**:
**Release note**:
```release-note-none
```
The registry abstraction is unnecessary and adds direct coupling to the
core types. By using a wrapper, we carry through the default
implementations of the non-mutating operations. The DeleteCollection
method is explicitly patched out since it cannot be correctly
implemented on the storage currently.
As a result, TableConvertor is now exposed.
A few other minor refactorings
* Corrected the case of some variables
* Used functions instead of methods for several helper methods
* Removed the legacy Deleter - service was the only remaining consumer
RangeSize was restricting IPv6 subnets to a /66 due to the
logic using a uint64. This is not practical for IPv6.
This change removes the /64 restriction, but also sets a limit
on the range that can be allocated, so that the bitmap will not grow too large.
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 portallocator comments
**What this PR does / why we need it**:
fix portallocator comments
**Which issue(s) this PR fixes** :
**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 ExternalTrafficLocalOnly from kube_feature gate
*What this PR does / why we need it**:
This PR is for v1.10.
External Source IP Preservation (ESIPP) had been promoted to GA since 1.7. Following the proposal on https://github.com/kubernetes/kubernetes/issues/46404#issuecomment-303939180, we should be able to remove it from feature gate now.
Added release note to announce this.
Also ref the previous attempt: https://github.com/kubernetes/kubernetes/pull/45857.
**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#56645
**Special notes for your reviewer**:
**Release note**:
```release-note
"ExternalTrafficLocalOnly" has been removed from feature gate. It has been a GA feature since v1.7.
```
Automatic merge from submit-queue (batch tested with PRs 52767, 55065, 55148, 56228, 56221). 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 DefaultGarbageCollectionPolicy to DeleteDependents for workloads controllers
**What this PR does / why we need it**:
As part of the apps/v1 GA effort (kubernetes/features#353) for v1.9. For core controllers, like `Deployment`, `DaemonSet`, `ReplicaSet`, and `StatefulSet`, changing the `DefaultGarbageCollectionPolicy` from `OrphanDependents` to `DeleteDependents` will make these objects consistent with the default behavior for all new objects.
For legacy API versions, the `DefaultGarbageCollectionPolicy` remains `OrphanDependents`.
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
ref #55027
**Special notes for your reviewer**:
/cc @enisoc @caesarxuchao @kow3ns
/assign @kubernetes/sig-apps-api-reviews
**Release note**:
```release-note
The default garbage collection policy for Deployment, DaemonSet, StatefulSet, and ReplicaSet has changed from OrphanDependents to DeleteDependents when the deletion is requested through an `apps/v1` endpoint. Clients using older endpoints will be unaffected. This change is only at the REST API level and is independent of the default behavior of particular clients (e.g. this does not affect the default for the kubectl `--cascade` flag).
If you upgrade your client-go libs and use the `AppsV1()` interface, please note that the default garbage collection behavior is changed.
```
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>.
Check dup NodePort with protocols when update services
**What this PR does / why we need it**:
As the title says.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#48579fixes: #54898fixes: #55327
**Special notes for your reviewer**:
/assign @freehan
/cc @cblecker
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 53796, 54666, 54516, 54625, 54704). 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 TestCRD Flake
**What this PR does / why we need it**:
The DestroyFunc functions returned by generic.NewRawStorage is never
called when we do a StartTestServer() in the test suite. For a quick
hack for now, added TrackStorageCleanup/RegisterStorageCleanup and
CleanupStorage. Note that unless TrackStorageCleanup is called (which
is called only from the test suite) the other two methods are
no-ops essentially. So no change in behavior at runtime. This vastly
brings down the number of goroutines that are left behind when this
test is executed and should reduce if not eliminate the flakiness
of TestCRD
**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
```
The DestroyFunc functions returned by generic.NewRawStorage is never
called when we do a StartTestServer() in the test suite. For a quick
hack for now, added TrackStorageCleanup/RegisterStorageCleanup and
CleanupStorage. Note that unless TrackStorageCleanup is called (which
is called only from the test suite) the other two methods are
no-ops essentially. So no change in behavior at runtime. This vastly
brings down the number of goroutines that are left behind when this
test is executed and should reduce if not eliminate the flakiness
of TestCRD
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>.
Updates RangeSize error message and tests for IPv6.
**What this PR does / why we need it**:
Updates the RangeSize function's error message and tests for IPv6. Converts RangeSize unit test to a table test and tests for success and failure cases. This is needed to support IPv6. Previously, it was unclear whether RangeSize supported IPv6 CIDRs. These updates make IPv6 support explicit.
**Which issue this PR fixes**
Partially fixes Issue #1443
**Special notes for your reviewer**:
/area ipv6
**Release note**:
```NONE
```
Automatic merge from submit-queue (batch tested with PRs 51824, 50476, 52451, 52009, 52237)
fix issue(#47976)Invalid value error when creating service from expor…
…ted config
**What this PR does / why we need it**:
close issue #47976
**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
```
Currently setting watch cache size for a given resource does not disable
the watch cache. This commit adds a new `default-watch-cache-size` flag
to map to the existing field, and refactors how watch cache sizes are
calculated to bring all of the code into one place. It also adds debug
logging to startup to allow us to verify watch cache enablement in
production.
Automatic merge from submit-queue (batch tested with PRs 51682, 51546, 51369, 50924, 51827)
Clear values for disabled alpha fields
Fixes#51831
Before persisting new or updated resources, alpha fields that are disabled by feature gate must be removed from the incoming objects.
This adds a helper for clearing these values for pod specs and calls it from the strategies of all in-tree resources containing pod specs.
Addresses https://github.com/kubernetes/community/pull/869
Automatic merge from submit-queue (batch tested with PRs 50889, 51347, 50582, 51297, 51264)
support fieldSelector spec.schedulerName
**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#49190
**Special notes for your reviewer**:
/assign @davidopp @bsalamat
/cc @lavalamp
**Release note**:
```release-note
add fieldSelector spec.schedulerName
```
Automatic merge from submit-queue (batch tested with PRs 51148, 50816, 49741, 50858, 51223)
Add tests around TableConvert and server side printing
Also wire in more table printers
@fabianofranz
Automatic merge from submit-queue (batch tested with PRs 50711, 50742, 50204)
Fix comment of limitranges
**What this PR does / why we need it**:
The comment of limitrages' API seems to be copied from
pkg/registry/autoscaling/horizontalpodautoscaler/storage/storage.go
with the other implementation code.
It is a little difficult to understand what is the API, then this
PR fixes it.
Automatic merge from submit-queue
Use nodePortOp for allocating healthCheck nodePort
**What this PR does / why we need it**: Allocate healthCheck nodePort via nodePortOp so that we won't leak port on failure.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#49999
**Special notes for your reviewer**:
/assign @xiangpengzhao @thockin
**Release note**:
```release-note
NONE
```
The comment of limitrages' API seems to be copied from
pkg/registry/autoscaling/horizontalpodautoscaler/storage/storage.go
with the other implementation code.
It is a little difficult to understand what is the API, then this
PR fixes it.
Automatic merge from submit-queue
Remove deprecated ESIPP beta annotations
**What this PR does / why we need it**:
Remove deprecated ESIPP beta annotations.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#50187
**Special notes for your reviewer**:
/assign @MrHohn
/sig network
**Release note**:
```release-note
Beta annotations `service.beta.kubernetes.io/external-traffic` and `service.beta.kubernetes.io/healthcheck-nodeport` have been removed. Please use fields `service.spec.externalTrafficPolicy` and `service.spec.healthCheckNodePort` instead.
```
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 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
Support exec/attach/portforward in `kubectl proxy`
Use the UpgradeAwareProxy shared code in kubectl proxy. Provide a separate transport for those requests that does not have HTTP/2 enabled. Refactor the code to be a bit cleaner in places and to better separate changes.
Fixes#32026
```release-note
`kubectl proxy` will now correctly handle the `exec`, `attach`, and `portforward` commands. You must pass `--disable-filter` to the command in order to allow these endpoints.
```
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.
Automatic merge from submit-queue (batch tested with PRs 50087, 39587, 50042, 50241, 49914)
convert default predicates to use the default
Builds on https://github.com/kubernetes/kubernetes/pull/50019 (lgtm'd already)
This converts the already default field selectors to use the default value. I'll let CI point out the unit test failures for me to chase.
@kubernetes/sig-api-machinery-misc
Automatic merge from submit-queue (batch tested with PRs 50091, 50231, 50238, 50236, 50243)
add fieldSelector podIP
**What this PR does / why we need it**:
Currently the `fieldSelector` for `Pod` only support `metadata.name`, `metadata.namespace`, `spec.nodeName`, `spec.restartPolicy` and `status.phase`.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#50075
**Special notes for your reviewer**:
/cc @wojtek-t @caesarxuchao @justinsb @LiliC
**Release note**:
```release-note
add fieldSelector podIP
```
Automatic merge from submit-queue (batch tested with PRs 49409, 49352, 49266, 48418)
Use helper to init ClusterIP and NodePort in Create of service
**What this PR does / why we need it**:
Make service `Create` more readable and testable.
- use `initClusterIP` introduced in #46197 to init ClusterIP allocation in service `Create`
- add a new helper `initNodePort` to init NodePort allocation in service `Create`
- TBD: add test case for `initNodePort`. This will cover the NodePort allocation process in `Create`. If this PR makes sense, I will write a test case later.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes#35354 (not directly. #35354 was fixed by #46197. The idea of this PR is from https://github.com/kubernetes/kubernetes/pull/46197#discussion_r120910077)
**Special notes for your reviewer**:
/cc @thockin @freehan
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 49055, 49128, 49132, 49134, 49110)
add svc and netpol to discovery
Fixes https://github.com/kubernetes/kubernetes/issues/48962
one shortname was missing entirely, the other was on a storage not actually used as storage.
@ncdc
Add support for creating resources that are not immediately visible to
naive clients, but must first be initialized by one or more privileged
cluster agents. These controllers can mark the object as initialized,
allowing others to see them.
Permission to override initialization defaults or modify an initializing
object is limited per resource to a virtual subresource "RESOURCE/initialize"
via RBAC.
Initialization is currently alpha.
Automatic merge from submit-queue (batch tested with PRs 44931, 44808)
Update to use Semantic.DeepEqual in regsitry
**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 https://github.com/kubernetes/kubernetes/issues/43402
**Release note**:
```
NONE
```
/assign @janetkuo
Automatic merge from submit-queue (batch tested with PRs 44837, 44779, 44492)
Default ObjectNameFunc for all REST Stores
All `Store`s in Kubernetes follow the same logic for determining the name of an object. This change makes it so that `CompleteWithOptions` defaults the `ObjectNameFunc` if it is not specified. Thus a user does not need to remember to use `ObjectMeta.Name`. Using the wrong field as the name can lead to an object which has a name that bypasses normal object name validation.
Signed-off-by: Monis Khan <mkhan@redhat.com>
cc @liggitt @soltysh for review
**Release note**:
```
NONE
```
Automatic merge from submit-queue
etcd component status check should include credentials
- [x] Add TLS credentials into `pkg/genericapiserver.Backend`.
- [x] Add TLS credentials into `pkg/registry/core/componentstatus.Server`.
- [x] `pkg/probe/http.httpProber` should accept the TLS credentials.
Now it is working.
```console
$ kubectl get cs
NAME STATUS MESSAGE ERROR
scheduler Healthy ok
controller-manager Healthy ok
etcd-0 Healthy {"health": "true"}
```
Fixes https://github.com/kubernetes/kubernetes/issues/27343.
All Stores in Kubernetes follow the same logic for determining the name
of an object. This change makes it so that CompleteWithOptions defaults
the ObjectNameFunc if it is not specified. Thus a user does not need to
remember to use ObjectMeta.Name. Using the wrong field as the name can
lead to an object which has a name that bypasses normal object name
validation.
Signed-off-by: Monis Khan <mkhan@redhat.com>
etcd has support for client-cert-auth, which can be configured via the flag `--ca-file`,
when that is enabled, all the client requests must present with a client certificate,
however, the current component status check uses a single transport for all of the checks,
this is wrong, the checks should be different for each of different component, and make
each of them use different transport(tls configurations).
Automatic merge from submit-queue
Change an Errorf to a Fatalf to prevent a panic.
In #37982 we see that if this error is hit then we will panic on the next line. Lets use `Fatalf` where appropriate :)
This does *not* fix the flake, since that's caused by the test assuming that a particular port is free.
Automatic merge from submit-queue (batch tested with PRs 42379, 42668, 42876, 41473, 43260)
Don't use global rand for nodeport and IP allocs
Seed it at construction.
The result of this was that an update to a Service would release the
NodePort temporarily (the repair loop would fix it in a minute). During
that window, another Service could get allocated that Port.
Automatic merge from submit-queue
Add websocket support for port forwarding
#32880
**Release note**:
```release-note
Port forwarding can forward over websockets or SPDY.
```
- adjust ports to int32
- CRI flows the websocket ports as query params
- Do not validate ports since the protocol is unknown
SPDY flows the ports as headers and websockets uses query params
- Only flow query params if there is at least one port query param
- split out port forwarding into its own package
Allow multiple port forwarding ports
- Make it easy to determine which port is tied to which channel
- odd channels are for data
- even channels are for errors
- allow comma separated ports to specify multiple ports
Add portfowardtester 1.2 to whitelist
Automatic merge from submit-queue (batch tested with PRs 40638, 40742, 40710, 40718, 40763)
move pkg/storage to apiserver
Mechanical move of `pkg/storage` (not sub packages) to `k8s.io/apiserver`.
@sttts
Automatic merge from submit-queue (batch tested with PRs 40111, 40368, 40342, 40274, 39443)
Eliminate "Unknown service type: ExternalName"
When creating an ExternalName service, rest.go still generate the warning message "Unknown service type: ExternalName". This should be eliminated as this type of service is supported now.
Automatic merge from submit-queue (batch tested with PRs 38739, 40480, 40495, 40172, 40393)
Rename controller pkg/registry/core/controller to pkg/registry/core/r…
…eplicationcontroller
**What this PR does / why we need it**:
Rename controller pkg/registry/core/controller to pkg/registry/core/replicationcontroller
This will clarify the purpose of the controller since intent is replicationcontroller
Please refer to
https://github.com/kubernetes/kubernetes/issues/17648
**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**:
NONE
**Release note**:
```NONE
```
Automatic merge from submit-queue
move pkg/fields to apimachinery
Purely mechanical move of `pkg/fields` to apimachinery.
Discussed with @lavalamp on slack. Moving this an `labels` to apimachinery.
@liggitt any concerns? I think the idea of field selection should become generic and this ends up shared between client and server, so this is a more logical location.
Automatic merge from submit-queue
run staging client-go update
Chasing to see what real problems we have in staging-client-go.
@sttts you get similar results?
Automatic merge from submit-queue
replace global registry in apimachinery with global registry in k8s.io/kubernetes
We'd like to remove all globals, but our immediate problem is that a shared registry between k8s.io/kubernetes and k8s.io/client-go doesn't work. Since client-go makes a copy, we can actually keep a global registry with other globals in pkg/api for now.
@kubernetes/sig-api-machinery-misc @lavalamp @smarterclayton @sttts
Automatic merge from submit-queue (batch tested with PRs 38354, 38371)
Add GetOptions parameter to Get() calls in client library
Ref #37473
This PR is super mechanical - the non trivial commits are:
- Update client generator
- Register GetOptions in batch/v2alpha1 group
Automatic merge from submit-queue (batch tested with PRs 35300, 36709, 37643, 37813, 37697)
[etcd] test cleanup: remove unnecessary AddPrefix()
What?
Remove etcdtest.AddPrefix() in tests. They will be automatically prepended in etcd storage.
Why?
ref: #36290#36374
After the change, it will double prepend.
Automatic merge from submit-queue
Reuse fields and labels
This should significantly reduce memory allocations in apiserver in large cluster.
Explanation:
- every kubelet is refreshing watch every 5-10 minutes (this generally is not causing relist - it just renews watch)
- that means, in 5000-node cluster, we are issuing ~10 watches per second
- since we don't have "watch heartbets", the watch is issued from previously received resourceVersion
- to make some assumption, let's assume pods are evenly spread across pods, and writes for them are evenly spread - that means, that a given kubelet is interested in 1 per 5000 pod changes
- with that assumption, each watch, has to process 2500 (on average) previous watch events
- for each of such even, we are currently computing fields.
This PR is fixing this problem.
Automatic merge from submit-queue
switch bootstrap controller to use a client where possible
While looking at https://github.com/kubernetes/kubernetes/issues/37040, I found more places where we can use a normal client instead of a direct to etcd connection.
@wojtek-t you made similar changes in the same controller.
