Commit Graph

36702 Commits

Author SHA1 Message Date
Kubernetes Prow Robot
ac6c77c440 Merge pull request #84121 from zouyee/renamefit
rename FilterPlugin NodeResources
2019-10-20 20:45:37 -07:00
Kubernetes Prow Robot
019b662ff5 Merge pull request #84017 from ahg-g/ahg-csi
Remove CSINode from scheduler cache.
2019-10-20 03:17:37 -07:00
zouyee
04340eaa34 rename FilterPlugin NodeResources
Signed-off-by: Zou Nengren <zouyee1989@gmail.com>
2019-10-20 12:51:16 +08:00
Kubernetes Prow Robot
e1685b5b59 Merge pull request #84074 from zouyee/proirity
LeastRequestedPriority/MostRequestedPriority/BalancedResourceAllocation as Score plugins
2019-10-19 17:21:37 -07:00
Kubernetes Prow Robot
9c25b16d88 Merge pull request #84099 from draveness/feature/remove-lister-wrapper
feat(scheduler): replace several algorithm listers with client listers
2019-10-19 08:05:37 -07:00
zouyee
bd4167d149 remove unused meta and rename lablance_allocated
Signed-off-by: Zou Nengren <zouyee1989@gmail.com>
2019-10-19 22:43:15 +08:00
Kubernetes Prow Robot
7e53c9d808 Merge pull request #83756 from hex108/permit
Refactor scheduler's framework permit API
2019-10-19 06:47:37 -07:00
zouyee
408c9da2a6 LeastRequestedPriority/MostRequestedPriority/BalancedResourceAllocation as Score plugins
Signed-off-by: Zou Nengren <zouyee1989@gmail.com>
2019-10-19 20:49:05 +08:00
draveness
ce33fcc311 feat: remove FakePDBLister 2019-10-19 19:10:42 +08:00
draveness
00a12c787c feat: implement node unschedulable as a filter plugin 2019-10-19 17:29:25 +08:00
draveness
e1f86e3460 feat(scheduler): replace several algorithm listers with client listers 2019-10-19 17:26:32 +08:00
Jun Gong
38b7668bb3 Refactor scheduler's framework permit API 2019-10-19 16:22:39 +08:00
Kubernetes Prow Robot
2b79d284fd Merge pull request #84066 from Huang-Wei/eps-migration
Migrate EvenPodsSpread Predicate to Filter plugin
2019-10-18 21:54:12 -07:00
Abdullah Gharaibeh
a772722660 Remove CSINode from scheduler cache. 2019-10-19 03:52:22 +00:00
Kubernetes Prow Robot
aab740ffc2 Merge pull request #82703 from draveness/feature/graduate-taint-nodes-by-condition-to-ga
feat: update taint nodes by condition to GA
2019-10-18 20:01:37 -07:00
draveness
1163a1d51e feat: update taint nodes by condition to GA 2019-10-19 09:17:41 +08:00
Kubernetes Prow Robot
98fcf2e6c7 Merge pull request #84034 from mrkm4ntr/use-informer-factory
Use frameworkHandle to get listers
2019-10-18 15:21:57 -07:00
Kubernetes Prow Robot
70f68062ad Merge pull request #84014 from ahg-g/ahg-tree
Make node tree order part of the snapshot
2019-10-18 15:21:37 -07:00
Wei Huang
64ff958e69 migrate EvenPodsSpread Predicate to Filter plugin 2019-10-18 14:18:02 -07:00
Kubernetes Prow Robot
de9a7d863d Merge pull request #83934 from wccsama/wcc-service-dev
Convert error messages to use event recorder
2019-10-18 12:37:46 -07:00
Abdullah Gharaibeh
63d7733e98 create an ordered list of nodes instead of iterating over the tree 2019-10-18 12:51:46 -04:00
Shintaro Murakami
bf256bcb00 Use frameworkHandle to get listers 2019-10-19 01:29:05 +09:00
Kubernetes Prow Robot
27c679baca Merge pull request #83982 from lichuqiang/frame_mig1
[migration phase 1] MatchInterPodAffinity as filter plugin
2019-10-18 09:23:58 -07:00
Kubernetes Prow Robot
422256110e Merge pull request #84073 from draveness/feature/cleanup-framework-plugins
feat: several cleanups in the scheduling package
2019-10-18 04:43:57 -07:00
lichuqiang
671f7690fe [migration phase 1] MatchInterPodAffinity as filter plugin 2019-10-18 16:26:34 +08:00
wccsama
18cf49e3df Convert error messages to use event recorder
remove mix protocol validation
remove check nil
2019-10-18 13:30:00 +08:00
Kubernetes Prow Robot
d1a79f136b Merge pull request #84054 from ahg-g/ahg-gp
GeneralPredicate as framework plugin config
2019-10-17 21:19:58 -07:00
draveness
39af760930 feat: several cleanups in the scheduling package
+ Remove unused variable in tests.
+ Use more common statement for interface conformance check.
+ Fix several comments in the framework plugins.
2019-10-18 11:14:05 +08:00
Kubernetes Prow Robot
9bf2ba7369 Merge pull request #84015 from ahg-g/ahg-filters
cleanup unnecessary func parameters in genericScheduler methods
2019-10-17 19:50:56 -07:00
Kubernetes Prow Robot
91050062f9 Merge pull request #83894 from notpad/feature/migration_nodevolumelimit
[migration phase 1] CSIMaxVolumeLimitChecker as filter plugin
2019-10-17 19:50:41 -07:00
Kubernetes Prow Robot
e129a6bc3f Merge pull request #80004 from Miciah/prefer-to-delete-doubled-up-pods-of-a-replicaset
Prefer to delete doubled-up pods of a ReplicaSet
2019-10-17 15:09:58 -07:00
Sean Sullivan
37fca51d74 Relocate tableprinter tests (#84027)
* Moves TestPrintUnstructuredObject to tableprinter_test.go

* Move TestUnknownTypePrinting to correct location of tableprinter_test.go

* Removes NewTablePrinter from TestCustomTypePrinting
2019-10-17 09:52:53 -07:00
Miciah Masters
980b6406b2 Prefer to delete doubled-up pods of a ReplicaSet
When scaling down a ReplicaSet, delete doubled up replicas first, where a
"doubled up replica" is defined as one that is on the same node as an
active replica belonging to a related ReplicaSet.  ReplicaSets are
considered "related" if they have a common controller (typically a
Deployment).

The intention of this change is to make a rolling update of a Deployment
scale down the old ReplicaSet as it scales up the new ReplicaSet by
deleting pods from the old ReplicaSet that are colocated with ready pods of
the new ReplicaSet.  This change in the behavior of rolling updates can be
combined with pod affinity rules to preserve the locality of a Deployment's
pods over rollout.

A specific scenario that benefits from this change is when a Deployment's
pods are exposed by a Service that has type "LoadBalancer" and external
traffic policy "Local".  In this scenario, the load balancer uses health
checks to determine whether it should forward traffic for the Service to a
particular node.  If the node has no local endpoints for the Service, the
health check will fail for that node.  Eventually, the load balancer will
stop forwarding traffic to that node.  In the meantime, the service proxy
drops traffic for that Service.  Thus, in order to reduce risk of dropping
traffic during a rolling update, it is desirable preserve node locality of
endpoints.

* pkg/controller/controller_utils.go (ActivePodsWithRanks): New type to
sort pods using a given ranking.
* pkg/controller/controller_utils_test.go (TestSortingActivePodsWithRanks):
New test for ActivePodsWithRanks.
* pkg/controller/replicaset/replica_set.go
(getReplicaSetsWithSameController): New method.  Given a ReplicaSet, return
all ReplicaSets that have the same owner.
(manageReplicas): Call getIndirectlyRelatedPods, and pass its result to
getPodsToDelete.
(getIndirectlyRelatedPods): New method.  Given a ReplicaSet, return all
pods that are owned by any ReplicaSet with the same owner.
(getPodsToDelete): Add an argument for related pods.  Use related pods and
the new getPodsRankedByRelatedPodsOnSameNode function to take into account
whether a pod is doubled up when sorting pods for deletion.
(getPodsRankedByRelatedPodsOnSameNode): New function.  Return an
ActivePodsWithRanks value that wraps the given slice of pods and computes
ranks where each pod's rank is equal to the number of active related pods
that are colocated on the same node.
* pkg/controller/replicaset/replica_set_test.go (newReplicaSet): Set
OwnerReferences on the ReplicaSet.
(newPod): Set a unique UID on the pod.
(byName): New type to sort pods by name.
(TestGetReplicaSetsWithSameController): New test for
getReplicaSetsWithSameController.
(TestRelatedPodsLookup): New test for getIndirectlyRelatedPods.
(TestGetPodsToDelete): Augment the "various pod phases and conditions, diff
= len(pods)" test case to ensure that scale-down still selects doubled-up
pods if there are not enough other pods to scale down.  Add a "various pod
phases and conditions, diff = len(pods), relatedPods empty" test case to
verify that getPodsToDelete works even if related pods could not be
determined.  Add a "ready and colocated with another ready pod vs not
colocated, diff < len(pods)" test case to verify that a doubled-up pod gets
preferred for deletion.  Augment the "various pod phases and conditions,
diff < len(pods)" test case to ensure that not-ready pods are preferred
over ready but doubled-up pods.
* pkg/controller/replicaset/BUILD: Regenerate.
* test/e2e/apps/deployment.go
(testRollingUpdateDeploymentWithLocalTrafficLoadBalancer): New end-to-end
test.  Create a deployment with a rolling update strategy and affinity
rules and a load balancer with "Local" external traffic policy, and verify
that set of nodes with local endponts for the service remains unchanged
during rollouts.
(setAffinity): New helper, used by
testRollingUpdateDeploymentWithLocalTrafficLoadBalancer.
* test/e2e/framework/service/jig.go (GetEndpointNodes): Factor building the
set of node names out...
(GetEndpointNodeNames): ...into this new method.
2019-10-17 11:52:32 -04:00
Miciah Masters
865c3c5670 TestGetPodsToDelete: Use field names in test cases
* pkg/controller/replicaset/replica_set_test.go (TestGetPodsToDelete): Use
explicit field names in declarations of test cases.
2019-10-17 11:50:09 -04:00
Kubernetes Prow Robot
fef819254a Merge pull request #83998 from draveness/feature/node-affinity-score-plugin
feat(scheduler): implement node affinity as score plugin
2019-10-17 08:24:38 -07:00
Abdullah Gharaibeh
517116921b GeneralPredicate as plugin config 2019-10-17 11:13:57 -04:00
Abdullah Gharaibeh
1293facf7b cleanup unnecessary func parameters in genericScheduler methods 2019-10-17 10:31:14 -04:00
notpad
f06925b0ee [migration phase 1] CSIMaxVolumeLimitChecker as filter plugin 2019-10-17 22:04:02 +08:00
Kubernetes Prow Robot
fa99d4068b Merge pull request #83466 from wongma7/block-reconstruct
Fix AWS block volume reconstruction to be like file
2019-10-17 03:18:51 -07:00
draveness
3d74da4d53 feat(scheduler): implement node affinity as score plugin 2019-10-17 17:02:10 +08:00
Kubernetes Prow Robot
bdc3f96838 Merge pull request #83989 from wojtek-t/remove_coordination_v1beta1
Swtich nodelifecyclecontroller to coordination/v1
2019-10-17 01:47:29 -07:00
Kubernetes Prow Robot
3719fcdffa Merge pull request #82072 from draveness/feature/use-context-instead-of-channel
feat(scheduler): use context in scheduler package
2019-10-17 01:46:39 -07:00
Kubernetes Prow Robot
6a5f0e6eda Merge pull request #81348 from yastij/code-org-service-controller
move service helpers to k8s.io/cloud-provider
2019-10-17 00:20:38 -07:00
Kubernetes Prow Robot
bc25814fb9 Merge pull request #84028 from ahg-g/ahg-hasfilter
make sure filters are executed when no predicates configured
2019-10-16 22:25:52 -07:00
Hongwei Yu
7574f95f7d optimize comments on exported constants (#83875)
* optimize comments on exported constants

* remove package in hack/.golint_failures are passing golint
2019-10-16 22:25:30 -07:00
Kubernetes Prow Robot
c63ec82490 Merge pull request #75190 from deads2k/kubelet-timeout
set backup value to ensure connections against kubelets eventually close
2019-10-16 22:24:38 -07:00
Abdullah Gharaibeh
17a6a7914c make sure filters are executed when no predicates configured 2019-10-16 21:27:57 -04:00
draveness
47a6c5b693 feat(scheduler): use context in the scheduler package
+ Use context instead of stopCh
+ Add context to the scheduling framework interface
2019-10-17 09:03:55 +08:00
Kubernetes Prow Robot
f7091992c0 Merge pull request #83577 from liu-cong/queue-metrics
Add incoming pod metrics to scheduler queue.
2019-10-16 17:34:39 -07:00
Matthew Wong
82786ff720 Fix AWS block volume reconstruction to be like file 2019-10-16 14:27:57 -07:00