The functionality to bootstrap node certificates is ready but is blocked
by a seperable issue discussed in:
https://github.com/kubernetes/community/pull/1982. The functionality
could be useful for power users who want to write their own approvers if
the feature could be promoted to beta. In it's current state this
feature doesn't help anybody.
I propose that we remove automated approval of node serving certificates
for now and work towards getting the node functionality to beta.
Automatic merge from submit-queue (batch tested with PRs 60476, 62462, 61391, 62535, 62394). 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 "git: Use VolumeHost.GetExec() to execute stuff in volume plugins"
This reverts commit c578542ad7 (PR #51098). The PR added support for containerized git, on the other hand it required git 1.8.5. This breaks git volumes on older distros (CentOS 7, Ubuntu 14.04) that have old git.
Git volumes are getting deprecated (https://github.com/kubernetes/kubernetes/issues/60999) so we should restore it to the last working state and not touch it any longer.
**Release note**:
```release-note
gitRepo volumes in pods no longer require git 1.8.5 or newer, older git versions are supported too now.
```
I'd like to cherry-pick it into 1.10.
/sig storage
Automatic merge from submit-queue (batch tested with PRs 60476, 62462, 61391, 62535, 62394). 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>.
Use function aws.Int64Value replace of deprecated function orZero
**What this PR does / why we need it**:
```
// orZero returns the value, or 0 if the pointer is nil
// Deprecated: prefer aws.Int64Value
func orZero(v *int64) int64 {
return aws.Int64Value(v)
}
```
Use function aws.Int64Value replace of deprecated function orZero and remove unused orZero .
**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 60476, 62462, 61391, 62535, 62394). 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>.
Private mount propagation
This PR changes the default mount propagation from "rslave" (newly added in 1.10) to "private" (default in 1.9 and before). "rslave" as default causes regressions, see below.
Value `"None"` has to be added to `MountPropagationMode` enum in API ("I don't want any propagation at all"), which translates to "private" on Linux. [We did not have use cases for it](https://github.com/kubernetes/community/pull/659#discussion_r131454319), but we have them now.
**Which issue(s) this PR fixes**
Fixes#62397, fixes#62396
**Special notes for your reviewer**:
CRI already has an option for private mount propagation in volumes, however it's called "PRIVATE", while Kubernetes API value is "None". I did not change PRIVATE to NONE to keep the interface stable. See `kubelet_pods.go`.
**Release note**:
```release-note
Default mount propagation has changed from "HostToContainer" ("rslave" in Linux terminology) to "None" ("private") to match the behavior in 1.9 and earlier releases. "HostToContainer" as a default caused regressions in some pods.
```
/sig storage
/sig node
Automatic merge from submit-queue (batch tested with PRs 62467, 62482, 62211). 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>.
Improve performance of affinity/anti-affinity predicate by 20x in large clusters
**What this PR does / why we need it**:
Improves performance of affinity/anti-affinity predicate by over 20x in large clusters. Performance improvement is smaller in small clusters, but it is still very significant and is about 4x. Also, before this PR, performance of the predicate was dropping quadratically with increasing size of nodes and pods. As the results shows, the slow down is now linear in larger clusters.
Affinity/anti-affinity predicate was checking all pods of the cluster for each node in the cluster to determine feasibility of affinit/anti-affinity terms of the pod being scheduled. This optimization first finds all the pods in a cluster that match the affinity/anti-affinity terms of the pod being scheduled once and stores the metadata. It then only checks the topology of the matching pods for each node in the cluster.
This results in major reduction of the search space per node and improves performance significantly.
Below results are obtained by running scheduler benchmarks:
```
make test-integration WHAT=./test/integration/scheduler_perf KUBE_TEST_ARGS="-run=xxx -bench=.*BenchmarkSchedulingAntiAffinity"
```
```
AntiAffinity Topology: Hostname
before: BenchmarkSchedulingAntiAffinity/500Nodes/250Pods-12 37031638 ns/op
after: BenchmarkSchedulingAntiAffinity/500Nodes/250Pods-12 10373222 ns/op
before: BenchmarkSchedulingAntiAffinity/500Nodes/5000Pods-12 134205302 ns/op
after: BenchmarkSchedulingAntiAffinity/500Nodes/5000Pods-12 12000580 ns/op
befor: BenchmarkSchedulingAntiAffinity/1000Nodes/10000Pods-12 498439953 ns/op
after: BenchmarkSchedulingAntiAffinity/1000Nodes/10000Pods-12 24692552 ns/op
AntiAffinity Topology: Region
before: BenchmarkSchedulingAntiAffinity/500Nodes/250Pods-12 60003672 ns/op
after: BenchmarkSchedulingAntiAffinity/500Nodes/250Pods-12 13346400 ns/op
before: BenchmarkSchedulingAntiAffinity/1000Nodes/10000Pods-12 600085491 ns/op
after: BenchmarkSchedulingAntiAffinity/1000Nodes/10000Pods-12 27783333 ns/op
```
**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 #
ref/ #56032#47318#25319
**Release note**:
```release-note
improve performance of affinity/anti-affinity predicate of default scheduler significantly.
```
/sig scheduling
Automatic merge from submit-queue (batch tested with PRs 62467, 62482, 62211). 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>.
Ensure resources created by `run --rm` are cleaned up
**Release note**:
```release-note
NONE
```
Resources created by `kubectl run --rm ...` are now cleaned up, even in the event of an error.
Relevant downstream issue: https://github.com/openshift/origin/issues/13276
cc @soltysh
Automatic merge from submit-queue (batch tested with PRs 62467, 62482, 62211). 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 nsenter GetFileType issue in containerized kubelet
**What this PR does / why we need it**:
UnmountDevice would fail in containerized kubelet in v1.9.x, and in the end, pod with volume mount could not be scheduled from one node to another since the original volume will never be released. This PR fixed this issue.
In [nsenter_mount.GetFileType func](https://github.com/kubernetes/kubernetes/blob/master/pkg/util/mount/nsenter_mount.go#L238), return error of following code will be different than [mount_linux.GetFileType func](https://github.com/kubernetes/kubernetes/blob/master/pkg/util/mount/mount.go#L347)
```
outputBytes, err := mounter.ne.Exec("stat", []string{"-L", `--printf "%F"`, pathname}).CombinedOutput()
```
Return error and output would be like following in nsenter_mount.GetFileType func:
error: `exit status 1`
output: `/usr/bin/stat: cannot stat '2': No such file or directory`
This PR makes the return error consistent as mount_linux.GetFileType func, and finally makes [isDeviceOpened func](https://github.com/kubernetes/kubernetes/blob/master/pkg/volume/util/operationexecutor/operation_generator.go#L1340) work.
**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#62282
**Special notes for your reviewer**:
/assign @dixudx
**Release note**:
```
fix nsenter GetFileType issue in containerized kubelet
```
/sig node
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>.
Add msau42 to approvers for volume scheduling
**What this PR does / why we need it**:
Add me as an approver for the volume scheduling code
**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. 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>.
The err has checked in TearDownAt func/kind bug
**What this PR does / why we need it**:
The err has checked in TearDownAt func/kind bug
**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 61608, 62304). 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 isNotDir error check
**What this PR does / why we need it**:
This check was supposed to handle the "subpath file" scenario, but:
1. It's wrong (should have been !)
2. It's not needed anymore. `IsLikelyNotMountPoint` was fixed to handle file mounts via https://github.com/kubernetes/kubernetes/pull/58433
**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. 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>.
PSP: move internal types to policy API group
**What this PR does / why we need it**:
This is a part of the PSP migration from extensions to policy API group. This PR moves internal types to the its final destination.
**Which issue(s) this PR fixes**:
Addressed to https://github.com/kubernetes/features/issues/5
Automatic merge from submit-queue (batch tested with PRs 62146, 62450). 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>.
Ensure expected load balancer is selected for Azure
**What this PR does / why we need it**:
Azure cloud provider is always selecting the last element of LB list. The reason is: getServiceLoadBalancer() refers the pointer of an local variable within for loop:
e7ed9b408a/pkg/cloudprovider/providers/azure/azure_loadbalancer.go (L202-L206)
This is a common mistake as golang is actually reusing same variable within the loop. This PR fixes this issue.
**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#62449
**Special notes for your reviewer**:
**Release note**:
```release-note
Ensure expected load balancer is selected for Azure
```
/sig azure
/kind bug
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>.
Move all create subcommands to its own subdirectory
**What this PR does / why we need it**:
Similarly to `kubectl set` subcommands move al `create` subcommands to its own subdirectory given that there are that many of them.
/assign @juanvallejo
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 58178, 62491, 60852). 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>.
use standard interface functions for printers
Updates the printObj funcs to be the standard resource printer interface we have.
@kubernetes/sig-cli-misc
Automatic merge from submit-queue (batch tested with PRs 62324, 61459, 62475, 62476, 61914). 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>.
Return error in mount_unsupported for unsupported platforms
**What this PR does / why we need it**:
Presently, `util/mount_unsupported.go` does not return any errors. For unsupported platforms, this hides failures. This PR returns errors, thereby properly informing users attempting to run on an unsupported platform.
**Which issue(s) this PR fixes** :
Fixes https://github.com/kubernetes/kubernetes/issues/61919
**Special notes for your reviewer**:
There are a few methods that simply call through to other methods. I did not return errors from those methods.
I've also updated an error check and message in `volume/fc/fc_test.go`, since it was ignoring an error on unsupported platforms.
**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>.
unused function clean
**What this PR does / why we need it**:
clean unused function in file
pkg/controller/volume/persistentvolume/scheduler_binder_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 62430, 62470). 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>.
make mikedanese owner of CertRotation features
I've taken these features over from @jcbsmpsn
```release-note
NONE
```
@kubernetes/sig-auth-pr-reviews
When kubernetes creates an iptable rule for "from-non-local public-port"
rule the rule gets created with two identical comment section.
The function `iptablesNonLocalNodePortArgs` creates a list of arguments
for the rule from iptablesCommonPortalArgs function. This function
already appends the arguments for the rules comments and therefore does
not require appending the comment again.
A flaky test exposed a race condition where shutting down one server
instance broke the startup of the next instance when using the same
socket path. Commit 1325c2f8be removed the reuse of the same socket
path and thus avoided the issue.
But the real fix is to ensure that the listening socket is really
closed once Stop returns. Two solutions were proposed in
https://github.com/grpc/grpc-go/issues/1861:
- waiting for the goroutine to complete
- closing the socket
The former is done here because it's cleaner to not keep lingering
goroutines. While at it, the Stop methods are made idempotent (similar
to e.g. Close on a socket) and no longer crash when called without
prior Start.
Fixes https://github.com/kubernetes/kubernetes/issues/59488
Automatic merge from submit-queue (batch tested with PRs 62455, 62465, 62427, 62416, 62411). 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>.
kuberuntime: logs: reduce logging level on waitLogs msg
Lots of occurrences of this msg coming from `waitLogs()`:
```
E0411 13:17:04.589338 7645 logs.go:383] Container "4fbf541ed1900c4670216a6a1ecf752cd07ac430f5547c5497fbc4b78e564b78" is not running (state="CONTAINER_EXITED")
E0411 14:02:18.168502 7645 logs.go:383] Container "dba4c535666d05310889965418592727047320743a233e226e2266b399836150" is not running (state="CONTAINER_EXITED")
E0411 14:02:41.342645 7645 logs.go:383] Container "a946289b36fe3c375c29dce020005424f3b980237892253d42b8bd8bfb595756" is not running (state="CONTAINER_EXITED")
E0411 14:02:49.907317 7645 logs.go:383] Container "e1d6014330e7422c03ae6db501d4fb296a4501355517cb60e2f910f54741361d" is not running (state="CONTAINER_EXITED")
```
Added in https://github.com/kubernetes/kubernetes/pull/55140
This message prints whenever something is watching the log when the container dies.
The comment right after the error msg say "this is normal" and thus should not be logged at Error level.
@derekwaynecarr @feiskyer @Random-Liu
Automatic merge from submit-queue (batch tested with PRs 62273, 62461). 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>.
allow higher burst for discovery
Discovery makes a lot of consecutive (maybe someday parallel) calls. One for every group and another for every version. Add a few CRDs and it's pretty easy to get to 50 calls.
This targets the burst on kubectl's discovery client only.
Automatic merge from submit-queue (batch tested with PRs 62273, 62461). 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>.
Don't log when error returned
**What this PR does / why we need it**:
Both logging and returning an error is an antipattern. If the caller wants it logged they will log it. And in this case it will be logged twice which is very confusing for debugging.
**Release note**:
```release-note
NONE
```
/kind cleanup
/sig api-machinery
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>.
Add generators for apps/v1 deployments
**What this PR does / why we need it**:
- Adds basic generator `DeploymentBasicAppsGeneratorV1` that is used on running `kubectl create deployment`.
- Changes default basic generator for `create deployment` to `deployment-basic/apps.v1`.
- Don't fall back in case generator is not available for `create deployment`, instead, return an error.
**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#61205
**Special notes for your reviewer**:
**Release note**:
```release-note
Add generators for `apps/v1` deployments.
```
Automatic merge from submit-queue (batch tested with PRs 62425, 62212, 60824, 62383, 62384). 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 the network plugin code
Moves the `pkg/kubelet/network` package down to `dockershim` since network
setup is done by the runtime. Also remove various interfaces/code added only for
backward compatibility with non-CRI runtimes, which are now removed from the
codebase.
**Release note**:
```release-note
NONE
```
