Commit Graph

25744 Commits

Author SHA1 Message Date
Kubernetes Submit Queue
2c9ffc0e88 Merge pull request #53764 from FengyunPan/unique-sg
Automatic merge from submit-queue (batch tested with PRs 56520, 53764). 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 service.UID into security group name

Related to: #53714 

**Release note**:
```release-note
NONE
```
2017-11-28 19:13:14 -08:00
Kubernetes Submit Queue
e9cf80f7c9 Merge pull request #56520 from oracle/for/upstream/master/revert-55336
Automatic merge from submit-queue (batch tested with PRs 56520, 53764). 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 "Merge pull request #55336 from oracle/for/upstream/master/53462"

This reverts commit ccb15fb498, reversing
changes made to 4904037645.



**What this PR does / why we need it**:

Reverting this PR due to the discussion https://github.com/kubernetes/kubernetes/pull/56448#discussion_r153508837 and https://github.com/kubernetes/kubernetes/pull/55336#discussion_r153652468. 

**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #56443

**Special notes for your reviewer**:

**Release note**:

```release-note
NONE
```

/cc @thockin @luxas @wlan0 @MrHohn

/priority critical-urgent
2017-11-28 19:13:11 -08:00
Kubernetes Submit Queue
4480204070 Merge pull request #55026 from dashpole/network_mock_docker
Automatic merge from submit-queue (batch tested with PRs 55893, 55906, 55026). 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>.

[Test Fix] Mock docker network dependencies and fix filtering bug

This PR only affects the mocked docker runtime, and has no impact on the kubelet.

Issue #53327 

When kubernetes creates a pod using the docker shim, it creates a container which contains the pod's network namespace, and then creates containers which specify that namespace.
The current mocked docker does not mock this interaction, and thus allows a container to be created even when the container whose network it is joining does not exist.
This allows the mocked kubelet to end up in a state where the pod does not exist, but a container in the pod does, and this breaks pod deletion.

This fixes the above by only allowing containers to be started if the container whose network it is trying to join is running.

Additionally, this PR fixes a filtering bug where we were incorrectly comparing docker container statuses.

/assign @shyamjvs 
can you test this to see if it fixes the issue?
/assign @Random-Liu 
for approval after @shyamjvs confirms this works.
2017-11-28 18:24:56 -08:00
Kubernetes Submit Queue
8cc6729d90 Merge pull request #55906 from aveshagarwal/master-scheduler-limits
Automatic merge from submit-queue (batch tested with PRs 55893, 55906, 55026). 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>.

Implement a priority function that considers pod's resource limits  

This PR implement a new priority function `ResourceLimitsPriorityMap` (disabled by default and behind alpha feature gate and not part of the scheduler's default priority functions list) that assigns a lowest possible score of 1 to a node that satisfies one or both of input pod's cpu and memory limits, mainly to break ties between nodes with same scores.

@kubernetes/sig-scheduling-pr-reviews @sjenning @derekwaynecarr 

**Release note**:

```release-note
A new priority function `ResourceLimitsPriorityMap` (disabled by default and behind alpha feature gate and not part of the scheduler's default priority functions list) that assigns a lowest possible score of 1 to a node that satisfies one or both of input pod's cpu and memory limits, mainly to break ties between nodes with same scores.
```
2017-11-28 18:24:52 -08:00
Kubernetes Submit Queue
6503f39bb1 Merge pull request #55893 from gnufied/aws-detach-fix-for-stopped-nodes
Automatic merge from submit-queue (batch tested with PRs 55893, 55906, 55026). 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>.

AWS: Implement fix for detaching volume from stopped instances

We should detach volume from stopped instanes.

Fixes https://github.com/kubernetes/kubernetes/issues/55892

```release-note
AWS: Fix detaching volume from stopped nodes.
```
2017-11-28 18:24:49 -08:00
Josh Horwitz
04c45e10db Revert "Merge pull request #55336 from oracle/for/upstream/master/53462"
This reverts commit ccb15fb498, reversing
changes made to 4904037645.
2017-11-28 19:04:42 -05:00
David Ashpole
63f7836d39 mock container networking and fix filtering bug 2017-11-28 15:32:18 -08:00
Kubernetes Submit Queue
8226973ae8 Merge pull request #52144 from andyxning/fix_network_value_for_stats_summary
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 network value for stats summary for multiple network interfaces

This PR is part of [Heapster #1788](https://github.com/kubernetes/heapster/pull/1788). 

The original reason is when there are more than one none `lo`, `docker0`, `veth` network interfaces instead of just one `eth0`, the network interface value is only partial and does not correct. For now, summary stats api only gets the eth0 network interface values.

The original issues about this can be find in [Heapster #1058](https://github.com/kubernetes/heapster/issues/1058) and [Cadvisor #1593](https://github.com/google/cadvisor/issues/1593).

```release-note
Fix stats summary network value when multiple network interfaces are available.
```

/cc @DirectXMan12 @piosz @xiangpengzhao @vishh @timstclair
2017-11-28 14:59:08 -08:00
Kubernetes Submit Queue
51033c4dec Merge pull request #56437 from dims/fix-unit-tests-that-need-openstack
Automatic merge from submit-queue (batch tested with PRs 56446, 56437). 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 unit tests that need openstack

**What this PR does / why we need it**:
Currently the unit tests that depend that they be on running inside an openstack vm fail as no one seem to have run them for a while.

**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 #56437

**Special notes for your reviewer**:

**Release note**:

```release-note
NONE
```
2017-11-28 13:08:13 -08:00
Kubernetes Submit Queue
7ce780d52e Merge pull request #56446 from ironcladlou/gc-test-flakes
Automatic merge from submit-queue (batch tested with PRs 56446, 56437). 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 GC sync race condition

Remove faulty diff detection logic from GC sync which leads to a race
condition: If the GC's discovery client is returning a fully up to date
view of server resources during the very first GC sync, the sync
function will never sync monitors or reset the REST mapper unless
discovery changes again. This causes REST mapping to fail for any custom
types already present in discovery.

Fixes https://github.com/kubernetes/kubernetes/issues/56262.

```release-note
NONE
```

/cc @liggitt @caesarxuchao
2017-11-28 13:08:10 -08:00
Kubernetes Submit Queue
6851e6abde Merge pull request #56344 from gmarek/gce-lb-fix-19
Automatic merge from submit-queue (batch tested with PRs 56344, 56473). 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 panic in GCE loadbalancer library
2017-11-28 04:17:30 -08:00
Kubernetes Submit Queue
52bd638e1a Merge pull request #56356 from tnozicka/fix-statefulset-creating-controllerrevision-multiple-times
Automatic merge from submit-queue (batch tested with PRs 56356, 56435). 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>.

Wait for controllerrevision informer to sync on statefulset controller startup

Fixes: https://github.com/kubernetes/kubernetes/issues/56355

Needs to be backported to at least 1.8, 1.9 and triaged for impact on 1.9 release

/cc @smarterclayton @kow3ns @enisoc 

**Release note**:
```release-note
NONE
```
2017-11-27 16:53:09 -08:00
Kubernetes Submit Queue
1ee9d20534 Merge pull request #56405 from shyamjvs/improve-error-handling-cidr-allocator
Automatic merge from submit-queue (batch tested with PRs 56094, 52910, 55953, 56405, 56415). 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>.

Requeue failed updates for retry in CIDR allocator

Split from https://github.com/kubernetes/kubernetes/pull/56352
Ref https://github.com/kubernetes/kubernetes/issues/52292

/cc @wojtek-t 
/kind bug
/priority critical-urgent
```release-note
NONE
```

cc @kubernetes/sig-network-misc
2017-11-27 16:04:28 -08:00
Kubernetes Submit Queue
ba4b07a431 Merge pull request #52910 from FengyunPan/volume-v3
Automatic merge from submit-queue (batch tested with PRs 56094, 52910, 55953, 56405, 56415). 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>.

Support VolumeV3 for OpenStack cloud Provider

Currently OpenStack supports Cinder v3 API, let Kubernetes support
it too.

Fix #52877

**Release note**:
```release-note
OpenStack cloud provider supports Cinder v3 API.
```
2017-11-27 16:04:22 -08:00
Dan Mace
a62d07ce2a Add a GC deadlock note 2017-11-27 16:50:29 -05:00
Dan Mace
eeeabce831 Add more GC sync logging 2017-11-27 16:47:37 -05:00
Davanum Srinivas
8300dd8901 Find a server name and router id to test TestRoutes 2017-11-27 16:32:10 -05:00
Dan Mace
9b2886df29 Ensure sync failures are correctly retried
Only track the last synced resources when all preceding steps have
completed to ensure that failures will be correctly retried.
2017-11-27 16:29:18 -05:00
Dan Mace
d89f58fcc8 Fix GC sync race condition
Remove faulty diff detection logic from GC sync which leads to a race
condition: If the GC's discovery client is returning a fully up to date
view of server resources during the very first GC sync, the sync
function will never sync monitors or reset the REST mapper unless
discovery changes again. This causes REST mapping to fail for any custom
types already present in discovery.
2017-11-27 16:11:47 -05:00
Davanum Srinivas
076f4da28b Allow volume test to run outside an actual OpenStack VM 2017-11-27 14:03:04 -05:00
Davanum Srinivas
f5299fe7e1 Skip TestRoutes if extension or router-id is missing 2017-11-27 14:02:23 -05:00
Davanum Srinivas
90a531ceb5 Set bs-version to auto for tests 2017-11-27 13:48:27 -05:00
Davanum Srinivas
c31bfe67a4 LBaaS v1 is no longer supported 2017-11-27 13:44:46 -05:00
Avesh Agarwal
b571001999 Implement resource limit priority function. This function checks if the input pod's
resource limits are satisfied by the input node's allocatable resources or not.
If yes, the node is assigned a score of 1, otherwise the node's score is not changed.
2017-11-27 12:53:47 -05:00
Kubernetes Submit Queue
e551886c5e Merge pull request #56359 from stewart-yu/fixExceptedError
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>.

Modify the judgment conditions

**What this PR does / why we need it**:
It should be a mistakes.

**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 #56347 

**Special notes for your reviewer**:

**Release note**:

```release-note
NONE
```
2017-11-27 09:22:09 -08:00
Shyam Jeedigunta
02a7e92bcf Requeue failed updates for retry in CIDR allocator 2017-11-27 12:57:29 +01:00
xiangpengzhao
cf699b6519 Use full gopath for externalTypes of missing API groups. 2017-11-26 00:12:45 +08:00
Maciej Szulik
f8ea46e8fe Use full gopath for externalTypes 2017-11-26 00:12:32 +08:00
stewart-yu
d8ebc4f88f fix bad return Error 2017-11-25 08:50:01 +08:00
Tomas Nozicka
ba6ed642a1 Wait for controllerrevision informer to sync on statefulset controller startup 2017-11-24 22:02:28 +01:00
Marek Grabowski
94b6a1fb70 Fix panic in GCE loadbalancer library 2017-11-24 13:27:57 +00:00
Wojciech Tyczynski
38597a26a8 Update NetworkUnavailable condition in ipam controller 2017-11-24 12:46:22 +01:00
Shyam Jeedigunta
19e56eb42d Use PATCHs instead of PUTs in CIDR allocator 2017-11-23 21:04:28 +01:00
Kubernetes Submit Queue
028c4c9399 Merge pull request #56240 from shyamjvs/improve-cidr-allocator
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>.

Migrate CIDR allocators to shared node informer

Ref https://github.com/kubernetes/kubernetes/issues/52292

/cc @wojtek-t @bowei
2017-11-23 08:48:06 -08:00
Hemant Kumar
ac2c68ad8f AWS: Implement fix for detaching volume from stopped instances
Clean up detach disk functions and remove duplication
2017-11-23 11:02:09 -05:00
Kubernetes Submit Queue
f0e337cd56 Merge pull request #53179 from wanghaoran1988/psp_flexvolume
Automatic merge from submit-queue (batch tested with PRs 55824, 53179). 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 Pod Security Policy to manage access to the Flexvolumes

**What this PR does / why we need it**:
For proposal: a1b9495e1b/contributors/design-proposals/auth/flex-volumes-drivers-psp.md (https://github.com/kubernetes/community/issues/723)
**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
Pod Security Policy can now manage access to specific FlexVolume drivers
```
2017-11-23 06:00:33 -08:00
jsafrane
4ad4ee3153 Added PVC Protection Controller
This controller removes protection finalizer from PVCs that are being
deleted and are not referenced by any pod.
2017-11-23 11:46:34 +01:00
pospispa
a06901a868 Admission Controller PVC Finalizer Plugin
This admission plugin puts finalizer to every created PVC. The finalizer is
removed by PVCProtectionController when the PVC is not referenced by any
pods and thus the PVC can be deleted.
2017-11-23 11:46:28 +01:00
pospispa
4d6d9817b0 PVC Being Deleted Checks in kubelet
Kubelet must not start pods that use PVCs that are being
deleted.
2017-11-23 11:41:30 +01:00
pospispa
a258269cf8 Added PVC Finalizing Controller feature switch. 2017-11-23 11:41:30 +01:00
Kubernetes Submit Queue
76e2bed9b1 Merge pull request #55810 from abhi/cri_stats
Automatic merge from submit-queue (batch tested with PRs 54843, 55810). 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>.

Integrating cadvisor stats to CRI Pod stats collection

For kubernetes-incubator/cri-containerd#341



**What this PR does / why we need it**:
This PR contains changes to update kubelet to consume metrics from both CRI and cadvisor for Pod stats. The corresponding cadvisor changes are here https://github.com/google/cadvisor/pull/1795.
Today CRI stats provides only core metrics such as CPU Usage, Memory Usage, Disk Usage. However the stats exposed by kubelet which is a superset of core metrics and Network, User Defined Metrics are not provided by CRI stats. Hence kubelet stats are extracted from 2 sources , CRI stats and cadvisor stats. 
After the change the summary stats for a CRI based runtime (containerd) is as show here:
https://gist.github.com/abhi/d7351861df6430eb4bc5d711d274ec35

This PR also contains test case change to reflect the stats for cri stats provider. 

**Which issue(s) this PR fixes** :
Fixes # https://github.com/kubernetes-incubator/cri-containerd/issues/341

**Special notes for your reviewer**:
https://docs.google.com/presentation/d/1Os3nyMRBlFuiBLCjPgeaPv6jXylrZW5jiDXJejlA3Wg/edit#slide=id.g27cb4cb6d8_0_0

Godep will be updated to main tree after https://github.com/google/cadvisor/pull/1795 PR is merged.

**Release note**:

```release-note
Kubelet can provide full summary api support except container log stats for CRI container runtime now.
```
cc @Random-Liu
2017-11-23 01:13:18 -08:00
Kubernetes Submit Queue
fb70e82ec6 Merge pull request #54843 from linyouchong/linyouchong-20171031
Automatic merge from submit-queue (batch tested with PRs 54843, 55810). 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 testSupport in downwardapi_test

add some code to test plugin.CanSupport function
/sig storage
2017-11-23 01:13:16 -08:00
Haoran Wang
e297a819ef generated 2017-11-23 16:23:53 +08:00
Kubernetes Submit Queue
2db7c41cfa Merge pull request #56252 from deads2k/rbac-04-migrate
Automatic merge from submit-queue (batch tested with PRs 56249, 56118, 56255, 56252, 56256). 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>.

clear resourceversion for migrated cluster roles

Fixes #56248

Need to clear the resource version.  Alternatively, we could clear it in storage when we clear and stomp other fields.

Works locally for me.
2017-11-23 00:23:26 -08:00
Kubernetes Submit Queue
d42af03d8a Merge pull request #56118 from gnufied/implement-ebs-resize
Automatic merge from submit-queue (batch tested with PRs 56249, 56118, 56255, 56252, 56256). 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>.

Implement ebs resize

Implement EBS disk resizing. 

xref - kubernetes/features#284


```release-note
Add support for resizing EBS disks
```
2017-11-23 00:23:21 -08:00
Haoran Wang
98faf6b39c support flexvlome in psp 2017-11-23 16:06:51 +08:00
abhi
e19f213027 Modifying cri stats test cases
This commit container modification to cri stats test
to verify CPU, Memory, Network stats collected by
cadvisor.

Signed-off-by: abhi <abhi@docker.com>
2017-11-22 22:48:43 -08:00
abhi
84b0e82db4 Integrating cadvisor stats to CRI Pod stats collection
This commit addresses the issue described here
https://github.com/kubernetes-incubator/cri-containerd/issues/341
The changes include using cadvisor stats in addition to CRI stats
for CRI runtimes. As described in the issue above , the CRI stats
currently doesnt provide all the necessary stats for the kubelet.
This commit addreses the need to extract stats from cadvisor which
is not available as CRI stats.

Signed-off-by: abhi <abhi@docker.com>
2017-11-22 22:48:37 -08:00
Kubernetes Submit Queue
1fdc688638 Merge pull request #55450 from feiskyer/seccomp-path
Automatic merge from submit-queue (batch tested with PRs 55952, 49112, 55450, 56178, 56151). 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 the wrong localhost seccomp path of CRI

**What this PR does / why we need it**:

Fix the wrong seccomp path comment.

**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 #55359

**Special notes for your reviewer**:

**Release note**:

```release-note
Fix CRI localhost seccomp path in format localhost//profileRoot/profileName.
```
2017-11-22 21:48:45 -08:00
Kubernetes Submit Queue
60c2090191 Merge pull request #49112 from gmarek/eventAPI
Automatic merge from submit-queue (batch tested with PRs 55952, 49112, 55450, 56178, 56151). 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>.

New API group for Events.

Fix kubernetes/features#383

cc @shyamjvs 

```release-note
Add events.k8s.io api group with v1beta1 API containing redesigned Event type.
```
2017-11-22 21:48:42 -08:00