Commit Graph

42458 Commits

Author SHA1 Message Date
Takeaki Matsumoto
785fac6826 Make updateAllocatedDevices() as a public method and call it in
podresources api
2020-02-07 13:26:56 +09:00
Kubernetes Prow Robot
69e01784c1 Merge pull request #87838 from Ikuyadeu/path2filepath
Move 'path' package usage to 'path/filepath' (repeat the #87713)
2020-02-06 17:48:11 -08:00
Kubernetes Prow Robot
9c1e124b15 Merge pull request #87823 from tallclair/test-tokens
Ensure testing credentials are labeled as such
2020-02-06 17:47:29 -08:00
Kubernetes Prow Robot
ef4e2c9f04 Merge pull request #87818 from tedyu/node-info-err
Check getNodeInfoError against nil
2020-02-06 17:47:13 -08:00
Kubernetes Prow Robot
3dc3c7c653 Merge pull request #87788 from ahg-g/ahg-filter
Reduce overhead of error message formatting and allocation for NodeResource filter
2020-02-06 17:46:50 -08:00
Kubernetes Prow Robot
6ac8f2c70c Merge pull request #87758 from klueska/upstream-cleanup-topology-manager
Cleanup TopologyManager and update policy.Merge()
2020-02-06 17:46:27 -08:00
Kubernetes Prow Robot
04e28b3674 Merge pull request #87098 from davidz627/feature/migratedRollback
Add annotation updating for migration for PVs and PVCs
2020-02-06 17:45:32 -08:00
Alena Prokharchyk
d634ed3850 Removed unnecessary not nil check in node registration process 2020-02-06 15:49:15 -08:00
Aldo Culquicondor
1071eb89f3 Remove deprecated fields from kubescheduler.config.k8s.io/v1alpha2
Signed-off-by: Aldo Culquicondor <acondor@google.com>
2020-02-06 16:22:30 -05:00
Aldo Culquicondor
e9e11de4e1 Remove Error log for nil StartTime
Signed-off-by: Aldo Culquicondor <acondor@google.com>
2020-02-06 14:09:44 -05:00
Ted Yu
871739e4df Check getNodeInfoError against nil 2020-02-06 10:57:43 -08:00
marosset
999fdfaddf Calling hcsshim instead of docker api to get stats for windows to greatly reduce latency 2020-02-06 17:59:10 +00:00
Alexander Zimmermann
a1c837022c Fixed Golint errors in pkg/controller/podautoscaler 2020-02-06 17:16:38 +01:00
Wei Huang
815206685c PodTopologySpread excludes terminatingPods when making scheduling decision 2020-02-06 07:38:50 -08:00
Alexander Zimmermann
22bd92a390 Changed comments to match with interfaces method description 2020-02-06 11:48:57 +01:00
Alexander Zimmermann
026ba54961 Fixed Golint errors in pkg/registry/core/pod 2020-02-06 11:40:37 +01:00
Kubernetes Prow Robot
a67238eed6 Merge pull request #87007 from wojtek-t/cleanup_manual_conversions
Cleanup manual conversions
2020-02-06 02:27:33 -08:00
Clayton Coleman
aed4d639a5 kubelet: Debug pod status output diff is wrong
The types were different so the diff output is not useful, both
should be pointers:

```
Feb 05 19:44:40 ci-ln-6k7l4-w-c-w9wbb.c.openshift-gce-devel-ci.internal hyperkube[2737]: I0205 19:44:40.222259    2737 status_manager.go:642] Pod status is inconsistent with cached status for pod "prometheus-k8s-1_openshift-monitoring(0e9137b8-3bd2-4353-b7f5-672749106dc1)", a reconciliation should be triggered:
Feb 05 19:44:40 ci-ln-6k7l4-w-c-w9wbb.c.openshift-gce-devel-ci.internal hyperkube[2737]:    interface{}(
Feb 05 19:44:40 ci-ln-6k7l4-w-c-w9wbb.c.openshift-gce-devel-ci.internal hyperkube[2737]: -         s"&PodStatus{Phase:Running,Conditions:[]PodCondition{PodCondition{Type:Initialized,Status:True,LastProbeTime:0001-01-01 00:00:00 +0000 UTC,LastTransitionTime:2020-02-05 19:13:30 +0000 UTC,Reason:,Message:,},PodCondit>
Feb 05 19:44:40 ci-ln-6k7l4-w-c-w9wbb.c.openshift-gce-devel-ci.internal hyperkube[2737]: +         v1.PodStatus{
Feb 05 19:44:40 ci-ln-6k7l4-w-c-w9wbb.c.openshift-gce-devel-ci.internal hyperkube[2737]: +                 Phase: "Running",
Feb 05 19:44:40 ci-ln-6k7l4-w-c-w9wbb.c.openshift-gce-devel-ci.internal hyperkube[2737]: +                 Conditions: []v1.PodCondition{
```
2020-02-05 14:52:46 -05:00
ksubrmnn
378fcc809a Check for node IP 2020-02-05 11:44:26 -08:00
Kubernetes Prow Robot
d90dd93855 Merge pull request #82111 from xieyanker/xieyanker-patch-2
remove stateCheckPeriod
2020-02-05 04:17:55 -08:00
Yuki Ueda
d1bc1875aa Move 'path' package usage to 'path/filepath' 2020-02-05 14:29:27 +09:00
Monis Khan
9b23f22472 Make oidc authenticator audience agnostic
This change removes the audience logic from the oidc authenticator
and collapses it onto the same logic used by other audience unaware
authenticators.

oidc is audience unaware in the sense that it does not know or
understand the API server's audience.  As before, the authenticator
will continue to check that the token audience matches the
configured client ID.

The reasoning for this simplification is:

1. The previous code tries to make the client ID on the oidc token
a valid audience.  But by not returning any audience, the token is
not valid when used via token review on a server that is configured
to honor audiences (the token works against the Kube API because the
audience check is skipped).

2. It is unclear what functionality would be gained by allowing
token review to check the client ID as a valid audience.  It could
serve as a proxy to know that the token was honored by the oidc
authenticator, but that does not seem like a valid use case.

3. It has never been possible to use the client ID as an audience
with token review as it would have always failed the audience
intersection check.  Thus this change is backwards compatible.

It is strange that the oidc authenticator would be considered
audience unaware when oidc tokens have an audience claim, but from
the perspective of the Kube API (and for backwards compatibility),
these tokens are only valid for the API server's audience.

This change seems to be the least magical and most consistent way to
honor backwards compatibility and to allow oidc tokens to be used
via token review when audience support in enabled.

Signed-off-by: Monis Khan <mok@vmware.com>
2020-02-04 13:24:49 -08:00
Tim Allclair
9d3670f358 Ensure testing credentials are labeled as such 2020-02-04 10:36:05 -08:00
Quan Tian
ea693833c8 Validate Except of IPBlock for NetworkPolicy spec
This patch enhances the validation of Except field that the values will
be rejected if they are not strictly within the CIDR range.
2020-02-05 01:35:44 +08:00
Abdullah Gharaibeh
0a476eb7d4 reduce overhead of error message formatting and allocation for scheudler NodeResource filter 2020-02-04 11:02:29 -05:00
wojtekt
1fc80c57ee Autogenerated 2020-02-04 16:06:36 +01:00
wojtekt
881dde8bee Remove unnecessary manual conversions 2020-02-04 16:05:25 +01:00
Kubernetes Prow Robot
76c89645c5 Merge pull request #87713 from kkmsft/filepath-fixes-windows
Move 'path' package usage to 'path/filepath'.
2020-02-03 17:07:25 -08:00
Eric Weber
adcdb2fd72 Enable FC mount options 2020-02-03 16:13:58 -06:00
Aldo Culquicondor
60c972098c Add defaults to pod affinity args
And remove pod affinity parameter from algorithm provider registry

Signed-off-by: Aldo Culquicondor <acondor@google.com>
2020-02-03 14:58:28 -05:00
Kubernetes Prow Robot
845b232321 Merge pull request #87451 from jnaulty/jnaulty/fix-aws-ebs-createvolume-signature
Fix CreateVolume signature in comment
2020-02-03 10:51:21 -08:00
Kevin Klues
d5addb4090 Cleanup logging and creation logic of TopologyManager in prep for beta 2020-02-03 17:13:29 +00:00
Kevin Klues
bc686ea27b Update TopologyManager.GetTopologyHints() to take pointers
Previously, this function was taking full Pod and Container objects
unnecessarily. This commit updates this so that they will take pointers
instead.
2020-02-03 17:13:28 +00:00
Kevin Klues
adaa58b6cb Update TopologyManager.Policy.Merge() to return a simple bool
Previously, the verious Merge() policies of the TopologyManager all
eturned their own lifecycle.PodAdmitResult result. However, for
consistency in any failed admits, this is better handled in the
top-level Topology manager, with each policy only returning a boolean
about whether or not they would like to admit the pod or not. This
commit changes the semantics to match this logic.
2020-02-03 17:13:28 +00:00
Kevin Klues
95a3ac447f Fix bug in TopologManager RemoveContainer()
Previously, we unconditionally removed *all* topology hints from a pod
whenever just one container was being removed. This commit makes it so
we only remove the hints for the single container being removed, and
then conditionally remove the pod from the podTopologyHints[podUID] when
no containers left in it.
2020-02-03 17:13:14 +00:00
Kubernetes Prow Robot
5caeec8023 Merge pull request #87710 from alena1108/jan30proxy
Ineffassign fixes for pkg/proxy
2020-02-02 19:09:20 -08:00
Kubernetes Prow Robot
7e5bfe4417 Merge pull request #85472 from dcbw/kubelet-network-approvers
kubelet/network: add sig-network-approvers to OWNERS
2020-02-01 12:55:19 -08:00
Kubernetes Prow Robot
35dfa2c4df Merge pull request #87565 from SomtochiAma/more-unit-tests-scheduler
More unit tests for scheduler
2020-02-01 01:11:20 -08:00
Kubernetes Prow Robot
536c2c8918 Merge pull request #87706 from liggitt/fix-statefulset-conversion
Fix statefulset conversion
2020-01-31 20:08:16 -08:00
Somtochi Onyekwere
0e00e0af11 Adds more unit test on Bind extension for the scheduler 2020-01-31 22:37:22 +01:00
wojtekt
b11b7d354d WatchBasedManager stops watching immutable objects 2020-01-31 20:53:21 +01:00
Mike Spreitzer
ff8256de3f Added server-side print column about FlowSchema referential integrity 2020-01-31 13:33:07 -05:00
Alena Prokharchyk
331cd94ccc Ineffassign fixes for pkg/volume
Along resolves errors whitelisted in hack/.staticcheck_failures
2020-01-31 10:14:45 -08:00
Antonio Ojea
4844b382dc kube-proxy: validate dual-stack cidrs
kube-proxy was not validating correctly the clusterCIDRs, if
dual-stack it MAY have 1 or more clusterCIDRs. If it has 2 cidrs and
at least one of each IP family.

It also fixes a bug where validation was not taking into account
the feature gates global state.
2020-01-31 16:00:17 +01:00
Kubernetes Prow Robot
7f0ea141d2 Merge pull request #87692 from everpeace/fix-scheduler-queue-metrics
Fix some scheduler metrics(pending_pods and schedule_attempts_total) are not recorded.
2020-01-31 05:30:19 -08:00
Jordan Liggitt
82107ff8ab Restore statefulset conversion that populates apiVersion/kind in volume templates 2020-01-31 07:47:35 -05:00
Shingo Omura
c9c4be66d3 Fix pending_pods, schedule_attempts_total was not recorded
because metric initializations are too early. This causes actual metric
instance become no-op.

modification made in thie commit to make sure actual metric instance won't be no-op metrics:

- re-initialize scheduler/metrics.PodSchedule{Successes, Failure, Errors} after metric creation
- scheduler/metrics.Register() should be called before initializing SchedulingQueue,
2020-01-31 17:03:41 +09:00
Kubernetes Prow Robot
1baceba376 Merge pull request #87394 from mattjmcnaughton/mattjmcnaughton/delete-sysctl-runtime-admit-handler
Delete the sysctl runtime admit handler
2020-01-30 21:20:45 -08:00
Kubernetes Prow Robot
fa4bc10ef1 Merge pull request #86793 from prameshj/gce-finalizer
Attach a new finalizer in GCE ILB creation.
2020-01-30 21:20:32 -08:00
Kubernetes Prow Robot
95a1d5e699 Merge pull request #82820 from odinuge/hugetlb-node-multiple-size
Add support for pre-allocated hugepages with 2+ sizes
2020-01-30 21:20:18 -08:00