github/kubernetes
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
97,757 Commits 1 Branch 31 Tags 1,019 MiB
c88ef90b1e
Commit Graph

484 Commits

Author SHA1 Message Date
lixiaobing1
e12fab6779 optimize the logic of noderesources comparision 2020-11-24 08:31:28 +08:00
rootlh
42c00bc523 fix bug: concurrent map writes error 2020-11-22 01:40:51 +08:00
lixiaobing1
8b351ac9d0 fix typos in type functionShapePoint 2020-11-19 20:22:09 +08:00
Tim Hockin
4068402459 Change trivial topology labels 
In these cases the actual label key is incidental.
 2020-11-12 11:21:37 -08:00
Kubernetes Prow Robot
3e43d5b92a
Merge pull request #96292 from wangyx1992/cleanup-scheduler-log-capatilization 
cleanup: fix log capitalization in scheduler
 2020-11-12 11:20:45 -08:00
Yixiang2019
842cc6b4e2 cleanup: fix log capitalization in scheduler 
Signed-off-by: Yixiang2019 <wang.yixiang@zte.com.cn>
 2020-11-12 20:10:26 +08:00
Aldo Culquicondor
3ce145787a Add AddedAffinity to the NodeAffinityArgs 
And use it in Filter and Score.

Change-Id: I173d8f2d5578762e9873181d5b44ea30b6dbbbc2
 2020-11-09 15:19:15 -05:00
Kubernetes Prow Robot
bd95fb101d
Merge pull request #94814 from adtac/preemption2 
optimise defaultpreemption: enumerate fewer candidates
 2020-11-05 21:18:50 -08:00
Tim Hockin
819ff9b087
Use topology labels instead of old beta names (#96033) 
* Rename const for topology.../zone

* Rename const for topology.../region

* Rename const for failure-domain.../zone

* Rename const for failure-domain.../region

* Restore old names for compat
 2020-11-05 20:26:50 -08:00
Kubernetes Prow Robot
2d6cd683bd
Merge pull request #95541 from cofyc/fix95538 
volume binding: report UnschedulableAndUnresolvable status instead of an error when bound PVs not found
 2020-11-05 15:16:51 -08:00
Yecheng Fu
0961891a7a report UnschedulableAndUnresolvable status instead of an error when PVCs can't find bound 
persistent volumes

This is an user error. We should't report an error.
 2020-11-05 10:28:40 +08:00
Adhityaa Chandrasekar
a3d94b53ca optimise defaultpreemption: enumerate fewer candidates 
Signed-off-by: Adhityaa Chandrasekar <adtac@google.com>
 2020-11-04 22:39:58 +00:00
Kubernetes Prow Robot
9866d4303a
Merge pull request #96151 from warmchang/volumelimittype 
Remove maximum volume limit comment which is easily outdated
 2020-11-04 07:02:05 -08:00
William Zhang
a7919225e2 Remove maximum volume limit comment which is easily outdated. 2020-11-04 21:39:22 +08:00
Aldo Culquicondor
d572249d30 Add runtime representation of []v1.PreferredSchedulingTerm 
to be used for repeatedly scoring nodes.

Change-Id: Ib1a0866979ce6cf75d1d9668c4bf8f6fb57298b2
 2020-11-03 10:21:31 -05:00
Kubernetes Prow Robot
43edb40632
Merge pull request #96092 from Huang-Wei/honor-false-LocalStorageCapacityIsolation 
Honor disabled LocalStorageCapacityIsolation in scheduling
 2020-11-02 17:12:15 -08:00
Wei Huang
fb782eee57
Honor disabled LocalStorageCapacityIsolation in scheduling 2020-11-02 13:21:08 -08:00
Kubernetes Prow Robot
17dcccbe07
Merge pull request #96109 from ingvagabund/move-IsScalarResourceName-under-scheduler 
Move pkg/apis/core/v1.IsScalarResourceName under pkg/scheduler/util
 2020-11-02 10:24:27 -08:00
Mike Dame
b7ba77c25b Move MatchNodeSelectorTerms to k8s.io/component-helpers 2020-10-29 13:39:13 -04:00
Kubernetes Prow Robot
2729b8e375
Merge pull request #95871 from damemi/move-unchanged-scheduler-helpers 
Change function signature for MatchNodeSelectorTerms
 2020-10-29 04:18:59 -07:00
Mike Dame
d9f334683e Change function signature for MatchNodeSelectorTerms 
As part of externalizing this function to the k8s.io/component-helpers repo,
this commit simplifies the function signature and makes its 2 helpers private
(nodeSelectorRequirementsAsSelector and nodeSelectorRequirementsAsFieldSelector).
 2020-10-27 12:52:06 -04:00
Jan Chaloupka
6822a0ec20 Move pkg/apis/core/v1.IsScalarResourceName under pkg/scheduler/util 
IsScalarResourceName is imported only inside pkg/scheduler packages.
 2020-10-27 13:40:32 +01:00
Kubernetes Prow Robot
5ed903dbfd
Merge pull request #95809 from alculquicondor/rebench-spread 
Optimize NormalizeScore for PodTopologySpread
 2020-10-26 11:23:02 -07:00
Kubernetes Prow Robot
1fcd02cc2e
Merge pull request #95812 from Huang-Wei/fix-pts-node-miss-label 
Fix a bug that Pods with topologySpreadConstraints get scheduled to nodes without required labels
 2020-10-23 09:55:57 -07:00
Aldo Culquicondor
e721576d55 Optimize NormalizeScore for PodTopologySpread 
Change-Id: I928174a220c8c0f4b69180fdfb2f075278a482be
 2020-10-23 11:15:15 -04:00
Aldo Culquicondor
ab7a38220f Add 5k nodes benchmark for pod topology spreading 
Change-Id: I1a5ed73ba03570671c0a92277678c0f3f786385f
 2020-10-23 11:15:15 -04:00
Wei Huang
f7a93ea915
Fix a bug that Pods with topologySpreadConstraints get scheduled to nodes without required labels 2020-10-22 15:56:16 -07:00
Aldo Culquicondor
15d2ab1615 Fix labels for spreading benchmark 
Nodes where being created with beta zone label, whereas the service was using stable.

Change-Id: I08ba1b6c083217759524236bde3888e527b4a44f
 2020-10-21 15:51:44 -04:00
Aldo Culquicondor
87c8349bdc Graduate DefaultPodTopologySpread to beta 
And set to enabled by default

Change-Id: Ie4cc4758c52492924cb0663450f2747908cb5882
 2020-10-19 16:58:09 -04:00
Kubernetes Prow Robot
fbe806ef9e
Merge pull request #95448 from alculquicondor/policy-spread 
Map SelectorSpreadPriority to PodTopologySpread plugin
 2020-10-15 12:55:25 -07:00
Aldo Culquicondor
95dd99905f Map SelectorSpreadPriority to PodTopologySpread plugin 
when DefaultPodTopologySpread feature is enabled

If SelectorSpreadPriority is in use, PodTopologySpread gets inevitably enabled.
When only EvenPodsSpreadPriority is in use, PodTopologySpread is configured without system defaults.

Change-Id: I2389a585cd8ad0bd35b0d2acae1665cd46908b3e
 2020-10-14 13:45:26 -04:00
Kubernetes Prow Robot
6268e6f990
Merge pull request #95069 from farah/split-scheduler 
Move scheduler interface to pkg/scheduler/framework
 2020-10-13 14:23:54 -07:00
Kubernetes Prow Robot
e3d7d067ab
Merge pull request #95066 from SataQiu/wrap-errors-2020092502 
Wrap errors for NodeLabel, NodePorts, NodePreferAvoidPods and NodeResourcesBalancedAllocation plugins
 2020-10-13 09:46:27 -07:00
Kubernetes Prow Robot
dc98ed96f6
Merge pull request #95085 from arghya88/wrap-error-taint-toleration-plugin 
Wrap errors in taint-toleration, service-affinity plugin
 2020-10-13 08:16:26 -07:00
Ali
09b2e8f638 Move scheduler interface to pkg/scheduler/framework 2020-10-13 13:13:27 +11:00
Kubernetes Prow Robot
4bbf4111e2
Merge pull request #92507 from ingvagabund/staging-library-go 
Initialize k8s.io/component-helpers staging repository
 2020-10-12 10:30:48 -07:00
Jan Chaloupka
274c536da3 Removing GetPodPriority from pkg/api and importing PodPriority from k8s.io/component-helpers 2020-10-11 21:40:11 +02:00
SataQiu
a0efc33158 wrap errors from NodeLabel, NodePorts, NodePreferAvoidPods and NodeResourcesBalancedAllocation plugins 2020-10-10 11:00:40 +08:00
Aldo Culquicondor
c8a0b9e8e7 Add defaultingType to PodTopologySpreadArgs 
Change-Id: Ibf6a4fdb39a31fe9deed68de7e7cb24a9bf9d06a
 2020-10-09 17:12:12 -04:00
Kubernetes Prow Robot
281412b5e9
Merge pull request #95191 from alculquicondor/policy-default-spread 
Move predicates and priorities configuration creation to Policy mapping
 2020-10-05 17:04:33 -07:00
Kubernetes Prow Robot
2f15024e7d
Merge pull request #95157 from adtac/schedplugins 
scheduler framework: document how to make plugins configurable
 2020-10-05 17:04:20 -07:00
Adhityaa Chandrasekar
3e7b2c0707 scheduler framework: document how to make plugins configurable 
Signed-off-by: Adhityaa Chandrasekar <adtac@google.com>
 2020-10-05 21:29:11 +00:00
Aldo Culquicondor
a0b72dbef9 Move predicates and priorities configuration creation to Policy mapping 
For separation of concerns. Also make the producers receive the
existing configuration so they can take decisions based on them.

Change-Id: Id5db5588aabb90e7b8f087bd3f3c967e86dc4acc
 2020-10-01 15:04:24 -04:00
Kubernetes Prow Robot
5b33feb943
Merge pull request #95146 from draveness/patch-5 
fix: use ">" instead of ">=" in resource allocation
 2020-09-30 12:55:08 -07:00
Draven
79f05c65a0 fix: use ">" instead of ">=" in resource allocation 2020-09-30 11:06:17 +08:00
Arghya Sadhu
8154dc95b5 wrap errors in selectorspread and podtoplogyspread plugin 2020-09-28 21:28:54 +05:30
Arghya Sadhu
ad415b9f54 wrap errors in service affinity plugin 2020-09-27 11:28:31 +05:30
Arghya Sadhu
ff3c751afc wrap errors in taint-toleration plugin 2020-09-26 21:16:41 +05:30
SataQiu
8c51c9955c wrap errors from DefaultPreemption, ImageLocality and NodeAffinity plugins 2020-09-25 10:38:27 +08:00
Arghya Sadhu
078b355da3 deprecate scheduler metrics BindingLatency and SchedulingAlgorithmPreemptionEvaluationDuration 2020-09-23 17:15:13 +05:30
Arghya Sadhu
c62f0dd165 removing deprecated scheduler metrics 2020-09-22 21:04:15 +05:30
Kubernetes Prow Robot
965137a992
Merge pull request #94692 from alculquicondor/wrap_errors_min 
Wrap errors from VolumeBinding and DefaultBinder plugins
 2020-09-15 18:27:34 -07:00
Wei Huang
185ba08fcd
Move podPassesBasicChecks() to VolumeBinding plugin 2020-09-11 13:54:02 -07:00
Aldo Culquicondor
7fb40fc03c Wrap errors on VolumeBinding plugin 
Signed-off-by: Aldo Culquicondor <acondor@google.com>
Change-Id: I23053528ac6857124fddd7f9fa26e122202ff4bd
Signed-off-by: Aldo Culquicondor <acondor@google.com>
 2020-09-10 16:22:16 -04:00
Aldo Culquicondor
94985e28ac Wrap errors on DefaultBinder plugin 
Signed-off-by: Aldo Culquicondor <acondor@google.com>
Change-Id: I2e3c8aa2c1a2a5102e9110b6cff91d66a79a90f1
 2020-09-10 16:22:16 -04:00
Wei Huang
4f7ae54f3e
fixup: add podLister as a member field of DefaultPreemption 2020-09-08 18:27:23 -07:00
Wei Huang
52bf6ba8ba
Preemption plugin to fetch pod from informer cache 2020-09-08 18:27:23 -07:00
Kubernetes Prow Robot
44ecd80cf1
Merge pull request #91557 from chendave/cleanup 
cleanup: remove useless methods
 2020-09-08 06:27:44 -07:00
Kubernetes Prow Robot
d239cdfbc0
Merge pull request #94059 from ahg-g/ahg-anti-affinity 
Track pods with required anti-affinity
 2020-09-08 04:54:12 -07:00
Kubernetes Prow Robot
b837699f74
Merge pull request #94125 from soulxu/only_includes_all_nodes_for_preferred 
Only process all nodes when incoming pod has no preferred affinity
 2020-09-07 14:01:58 -07:00
Kubernetes Prow Robot
217d89a59f
Merge pull request #93843 from soulxu/fast_path_podaffinity 
Fast return when no any matched anti-affinity terms
 2020-09-04 03:31:55 -07:00
Kubernetes Prow Robot
72350a3cf7
Merge pull request #93083 from Huang-Wei/dupe-import-alias 
Remove duplicate path imports
 2020-08-28 04:34:53 -07:00
Kubernetes Prow Robot
a4103dfeaf
Merge pull request #92819 from chendave/cleanup_machine 
Change the node name from "machine" to "node"
 2020-08-27 19:06:57 -07:00
Kubernetes Prow Robot
9062c43b76
Merge pull request #94211 from soulxu/cleanup_candidates 
Initialize candidate directly instead of iterating the array of candidates
 2020-08-27 16:08:06 -07:00
Kubernetes Prow Robot
a5160414e0
Merge pull request #93706 from SimpCosm/fix/scheduler-plugin-comment 
Fix an error in NodeUnschedulable plugin comment
 2020-08-27 16:06:48 -07:00
He Jie Xu
0e8bd4c550 Initialize candidate directly instead of iterating the array of candidates 
Using existing victimsMap to get the victims, then it is easy to build candidate
directly.
 2020-08-27 23:29:36 +08:00
He Jie Xu
ccd8eb3b1b Only process all nodes when incoming pod has no preferred affinity 
Currently, in interpodaffinty plugin, it only processes all nodes when the incoming
pod with affinity. Actually, it only cares about all nodes when the incoming pod
with preferred affinity. Then it will reduces the number of nodes need to be
processed.
 2020-08-25 16:13:17 +08:00
Kubernetes Prow Robot
1c93be24ee
Merge pull request #93629 from cofyc/93009 
fix flaky TestVolumeBinding unit test
 2020-08-21 08:33:40 -07:00
Abdullah Gharaibeh
a8873e1a43 Track pods with required anti-affinity 
This is a performance optimization that reduces the overhead of inter-pod affinity PreFilter calculaitons. Basically
eliminates that overhead when no pods in the cluster use required pod anti-affinity. This offered 20% improvement on 5k clusters for preferred anti-affinity benchmarks.
 2020-08-21 10:09:21 -04:00
He Jie Xu
75ccb90407 Fast return when no any matched anti-affinity terms 
When check the incoming pod's anti-affinity rules, there is change to
return early when there is no any matched anti-affinity terms in the
whole cluster.
 2020-08-10 14:53:10 +08:00
houmin
868dd41a96 Fix an error in NodeUnschedulable plugin comment 2020-08-10 11:20:23 +08:00
Mike Dame
012245c5b9 Add LabelSelector validation in Pod Affinity/AntiAffinity Filter and Score plugins 
The lack of this validation on incoming pods causes unpredictable cluster outcomes
when later calculating affinity results against existing pods (see #92714). This fix
quickly addresses the main source where these problems should be caught.

It is unfortunately difficult to add this validation directly to the API server due
to the fact that it may break migrations with existing pods that fail this check. This
is a compromise to address the current issue.
 2020-08-07 12:17:40 -04:00
Yecheng Fu
96d0408a89 fix TestVolumeBinding unit test 2020-08-03 07:06:06 +08:00
Abdullah Gharaibeh
5e81a2de98 Optimize VolumeRestrictions scheduler plugin 2020-07-22 23:00:01 -04:00
Wei Huang
bc04d73330
remove duplicate path import 2020-07-14 16:34:09 -07:00
Wei Huang
4e8ccf0187
Refactor and expose common preemption functions 2020-07-11 23:17:21 -07:00
Kubernetes Prow Robot
d06ff65943
Merge pull request #92876 from Huang-Wei/pdbLister 
Add pdbLister as a member field of struct DefaultPreemption
 2020-07-11 20:57:42 -07:00
Kubernetes Prow Robot
016c2f64de
Merge pull request #92840 from adtac/listers 
selectorspread: access listers in plugin instantiation
 2020-07-11 20:56:23 -07:00
Kubernetes Prow Robot
36b4c2942b
Merge pull request #92815 from Huang-Wei/bypass-prefilter-svcaffinity 
Bypass PreFilter in ServiceAfffinity if AffinityLabels arg is not present
 2020-07-10 15:43:11 -07:00
Kubernetes Prow Robot
0cb7e320a5
Merge pull request #92784 from pohly/generic-ephemeral-inline-volumes 
generic ephemeral inline volumes
 2020-07-10 15:41:46 -07:00
Dave Chen
a1b2a7765d Change the node name from "machine" to "node" 
Latest change on master rename the node name from "machine" to "node"
but haven't update all the affected code, which causes some of testcases
invalid.

Signed-off-by: Dave Chen <dave.chen@arm.com>
 2020-07-10 10:17:58 +08:00
Patrick Ohly
ff3e5e06a7 GenericEphemeralVolume: initial implementation 
The implementation consists of
- identifying all places where VolumeSource.PersistentVolumeClaim has
  a special meaning and then ensuring that the same code path is taken
  for an ephemeral volume, with the ownership check
- adding a controller that produces the PVCs for each embedded
  VolumeSource.EphemeralVolume
- relaxing the PVC protection controller such that it removes
  the finalizer already before the pod is deleted (only
  if the GenericEphemeralVolume feature is enabled): this is
  needed to break a cycle where foreground deletion of the pod
  blocks on removing the PVC, which waits for deletion of the pod

The controller was derived from the endpointslices controller.
 2020-07-09 23:29:24 +02:00
Kubernetes Prow Robot
3a5e7ea986
Merge pull request #92752 from chendave/skip_preemption 
Cut off the cost to run filter plugins when no victim pods are found
 2020-07-09 09:10:10 -07:00
Kubernetes Prow Robot
70e09f2c24
Merge pull request #88842 from angao/fit-arg 
add args for NodeResourcesFit plugin
 2020-07-09 05:04:10 -07:00
Kubernetes Prow Robot
55d77ade67
Merge pull request #92489 from alculquicondor/sig-storage-ownership 
Add SIG storage owner aliases
 2020-07-09 00:05:20 -07:00
Kubernetes Prow Robot
94a08e159a
Merge pull request #92387 from pohly/csi-storage-capacity 
CSI storage capacity check
 2020-07-09 00:04:59 -07:00
Wei Huang
9d377eb655
Add pdbLister as a member field of struct DefaultPreemption 2020-07-07 12:25:53 -07:00
Adhityaa Chandrasekar
832a53acdb selectorspread: access listers in plugin instantiation 2020-07-07 14:45:28 +00:00
Aldo Culquicondor
27ec356d76 Add SIG storage owner aliases 
And give ownership to pkg/scheduler/framework/plugins/volumebinding

Signed-off-by: Aldo Culquicondor <acondor@google.com>
Change-Id: I4bd89b1745a2be0e458601056ab905bdd6692195
 2020-07-07 10:26:16 -04:00
Dave Chen
028af0970f Cut off the cost to run filter plugins when no victim pods are found 
If no potential victims could be found, there is no need to evaluate the node
again, since its state didn't change.

It's safe to return and thus prevent scheduling from running the filter plugins
again.

NOTE:
A node that is filtered out by filter plugins could pass the filter plugins if
there is a change on that node, i.e. pods termination on that node.

Previously, this could be either caught by the normal `schedule` or `preempt` (pods
are terminated when the preemption logic tries to find the nodes and re-evaluate
the filter plugins.)

Actually, this shouldn't be taken care by the preemption, consider the routine
of `schedule` is always running when the interval is "zero", let `schedule`
take care of it will release `preempt` from something irrelevant with the `preemption`.

Due to above reason, couple of testcase as well as the logic of checking the existence
of victim pods are removed as it will never happen after the change.

Signed-off-by: Dave Chen <dave.chen@arm.com>
 2020-07-07 09:55:34 +08:00
Patrick Ohly
0efbbe8555 CSIStorageCapacity: check for sufficient storage in volume binder 
This uses the information provided by a CSI driver deployment for
checking whether a node has access to enough storage to create the
currently unbound volumes, if the CSI driver opts into that checking
with CSIDriver.Spec.VolumeCapacity != false.

This resolves a TODO from commit 95b530366a.
 2020-07-06 19:20:10 +02:00
Wei Huang
07583bf95b
Bypass PreFilter in ServiceAfffinity if AffinityLabels arg is not present 2020-07-05 23:37:04 -07:00
Kubernetes Prow Robot
86096addb1
Merge pull request #92689 from chendave/fix_testcase 
Fix the nits found in the testcases of `PodTopologySpread`
 2020-07-03 20:31:26 -07:00
Kubernetes Prow Robot
19883b50f8
Merge pull request #92604 from soulxu/fix_preemption_with_nominated_node 
The Pod is eligible to preempt when previous nominanted node is UnschedulableAndUnresolvable
 2020-07-03 05:03:01 -07:00
Dave Chen
3e65fe4378 Change the exception to avoid the cost of preemption 
node's labels doesn't contain the required topologyKeys in `Constraints`
cannot be resolved by preempting the pods on that pods.

One use case that could easily reproduce the issue is,
- set `alwaysCheckAllPredicates` to true.
- one node contains all the required topologyKeys but is failed in predicates
  such as 'taint'.
- another node doesn't hold all the required topologyKeys, and thus return `Unschedulable`
  status code.
- scheduler will try to preempt the pods on the above node with lower priorities.

Signed-off-by: Dave Chen <dave.chen@arm.com>
 2020-07-03 10:17:31 +08:00
He Jie Xu
b3741f344e The Pod is eligible to preempt when previous nominanted node is UnschedulableAndUnresolvable 
If the Pod's previous nominated node is UnschedulableAndUnresolvable from previous
filtering, it should be considered for preemption again.
 2020-07-03 08:57:45 +08:00
Dave Chen
41fd19760e Fix the nits found in the testcases of PodTopologySpread 
Signed-off-by: Dave Chen <dave.chen@arm.com>
 2020-07-02 12:37:46 +08:00
Wei Huang
7362fccdd7
Polish unit tests of defaultpreemptio plugin 2020-06-30 14:05:48 -07:00
Kubernetes Prow Robot
784b0738b5
Merge pull request #92578 from zhouya0/fix_preemt_comment 
Fix scheduler preemt function comment
 2020-06-29 18:35:27 -07:00
zhouya0
59f9a7d81e Fix preemt function comment 2020-06-28 18:29:55 +08:00