github
/
k3s
mirror of https://github.com/k3s-io/k3s
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Merge pull request #37384 from derekwaynecarr/fix-cross-build 

Fix cross-build for memcg notification
pull/6/head
Dawn Chen 2016-11-23 13:59:18 -08:00 committed by GitHub
parent 5190ace8f0 1ec69f658c
commit e968173d6b
4 changed files with 44 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
pkg/kubelet/eviction/threshold_notifier.go → pkg/kubelet/eviction/threshold_notifier_linux.go
View File

@ -28,11 +28,6 @@ import (
"github.com/golang/glog"
)
// ThresholdNotifier notifies the user when an attribute crosses a threshold value
type ThresholdNotifier interface {
Start(stopCh <-chan struct{})
}
type memcgThresholdNotifier struct {
watchfd int
controlfd int

27
pkg/kubelet/eviction/threshold_notifier_unsupported.go Normal file
View File

@ -0,0 +1,27 @@
// +build !linux
/*
Copyright 2016 The Kubernetes Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package eviction
import "fmt"
// NewMemCGThresholdNotifier sends notifications when a cgroup threshold
// is crossed (in either direction) for a given cgroup attribute
func NewMemCGThresholdNotifier(path, attribute, threshold, description string, handler thresholdNotifierHandlerFunc) (ThresholdNotifier, error) {
return nil, fmt.Errorf("threshold notification not supported")
}

5
pkg/kubelet/eviction/types.go
View File

@ -164,3 +164,8 @@ type nodeReclaimFuncs []nodeReclaimFunc
// thresholdNotifierHandlerFunc is a function that takes action in response to a crossed threshold
type thresholdNotifierHandlerFunc func(thresholdDescription string)
// ThresholdNotifier notifies the user when an attribute crosses a threshold value
type ThresholdNotifier interface {
Start(stopCh <-chan struct{})
}

20
test/test_owners.csv
View File

@ -44,6 +44,7 @@ CronJob should not schedule jobs when suspended,soltysh,1
CronJob should not schedule new jobs when ForbidConcurrent,soltysh,1
CronJob should replace jobs when ReplaceConcurrent,soltysh,1
CronJob should schedule multiple jobs concurrently,soltysh,1
DNS config map should be able to change configuration,derekwaynecarr,0
DNS horizontal autoscaling kube-dns-autoscaler should scale kube-dns pods in both nonfaulty and faulty scenarios,MrHohn,0
DNS horizontal autoscaling kube-dns-autoscaler should scale kube-dns pods when cluster size changed,MrHohn,0
DNS should provide DNS for ExternalName services,rmmh,1
@ -175,21 +176,18 @@ Federation secrets Secret objects should be created and deleted successfully,pmo
Federation secrets Secret objects should be deleted from underlying clusters when OrphanDependents is false,nikhiljindal,0
Federation secrets Secret objects should not be deleted from underlying clusters when OrphanDependents is nil,nikhiljindal,0
Federation secrets Secret objects should not be deleted from underlying clusters when OrphanDependents is true,nikhiljindal,0
GCP Volumes GlusterFS should be mountable,rkouj,0
GCP Volumes NFSv4 should be mountable for NFSv4,rkouj,0
GCP Volumes GlusterFS should be mountable,nikhiljindal,0
GCP Volumes NFSv4 should be mountable for NFSv4,nikhiljindal,0
GKE local SSD should write and read from node local SSD,fabioy,0
GKE node pools should create a cluster with multiple node pools,fabioy,1
Garbage Collection Test: * Should eventually garbage collect containers when we exceed the number of dead containers per container,Random-Liu,0
Garbage collector should delete pods created by rc when not orphaning,justinsb,1
Garbage collector should orphan pods created by rc if delete options say so,fabioy,1
Garbage collector should orphan pods created by rc if deleteOptions.OrphanDependents is nil,zmerlynn,1
"Generated release_1_5 clientset should create pods, delete pods, watch pods",ghodss,1
"Generated release_1_5 clientset should create v2alpha1 cronJobs, delete cronJobs, watch cronJobs",soltysh,1
HA-master pods survive addition/removal different zones,rkouj,0
HA-master pods survive addition/removal same zone,rkouj,0
HA-master pods survive addition/removal different zones,nikhiljindal,0
HA-master pods survive addition/removal same zone,nikhiljindal,0
HA-master survive addition/removal replicas different zones,derekwaynecarr,0
HA-master survive addition/removal replicas same zone,derekwaynecarr,0
Hazelcast should create and scale hazelcast,mikedanese,1
Horizontal pod autoscaling (scale resource: CPU) Deployment Should scale from 1 pod to 3 pods and from 3 to 5,jszczepkowski,0
Horizontal pod autoscaling (scale resource: CPU) Deployment Should scale from 5 pods to 3 pods and from 3 to 1,jszczepkowski,0
@ -262,6 +260,7 @@ Kubelet Container Manager Validate OOM score adjustments once the node is setup
Kubelet Container Manager Validate OOM score adjustments once the node is setup guaranteed container's oom-score-adj should be -998,kargakis,1
Kubelet Container Manager Validate OOM score adjustments once the node is setup pod infra containers oom-score-adj should be -998 and best effort container's should be 1000,timothysc,1
Kubelet Eviction Manager hard eviction test pod using the most disk space gets evicted when the node disk usage is above the eviction hard threshold should evict the pod using the most disk space,karlkfi,1
Kubelet Volume Manager Volume Manager On terminatation of pod with memory backed volume should remove the volume from the node,derekwaynecarr,0
Kubelet experimental resource usage tracking resource tracking for * pods per node,yujuhong,0
Kubelet regular resource usage tracking resource tracking for * pods per node,yujuhong,0
Kubelet when scheduling a busybox command in a pod it should print the output to logs,ixdy,1
@ -336,7 +335,6 @@ PersistentVolumes with multiple PVs and PVCs all in same ns should create 2 PVs
PersistentVolumes with multiple PVs and PVCs all in same ns should create 3 PVs and 3 PVCs: test write access,caesarxuchao,1
PersistentVolumes with multiple PVs and PVCs all in same ns should create 4 PVs and 2 PVCs: test write access,caesarxuchao,1
Pet Store should scale to persist a nominal number ( * ) of transactions in * seconds,xiang90,1
Pet set recreate should recreate evicted statefulset,roberthbailey,1
"Pod Disks Should schedule a pod w/ a RW PD, gracefully remove it, then schedule it on another host",alex-mohr,1
"Pod Disks Should schedule a pod w/ a readonly PD on two hosts, then remove both gracefully.",ghodss,1
"Pod Disks should schedule a pod w/ a RW PD shared between multiple containers, write to PD, delete pod, verify contents, and repeat in rapid succession",saad-ali,0
@ -458,8 +456,13 @@ Services should work after restarting kube-proxy,bprashanth,0
SimpleMount should be able to mount an emptydir on a container,karlkfi,1
"Spark should start spark master, driver and workers",girishkalele,1
"Staging client repo client should create pods, delete pods, watch pods",jbeda,1
Stateful Set recreate should recreate evicted statefulset,derekwaynecarr,0
StatefulSet Basic StatefulSet functionality Scaling down before scale up is finished should wait until current pod will be running and ready before it will be removed,derekwaynecarr,0
StatefulSet Basic StatefulSet functionality Scaling should happen in predictable order and halt if any pet is unhealthy,derekwaynecarr,0
StatefulSet Basic StatefulSet functionality should allow template updates,derekwaynecarr,0
StatefulSet Basic StatefulSet functionality should handle healthy pet restarts during scale,girishkalele,1
StatefulSet Basic StatefulSet functionality should provide basic identity,bprashanth,1
StatefulSet Deploy clustered applications should creating a working CockroachDB cluster,derekwaynecarr,0
StatefulSet Deploy clustered applications should creating a working mysql cluster,yujuhong,1
StatefulSet Deploy clustered applications should creating a working redis cluster,yifan-gu,1
StatefulSet Deploy clustered applications should creating a working zookeeper cluster,pmorie,1
@ -636,6 +639,8 @@ k8s.io/kubernetes/pkg/credentialprovider,justinsb,1
k8s.io/kubernetes/pkg/credentialprovider/aws,zmerlynn,1
k8s.io/kubernetes/pkg/credentialprovider/gcp,mml,1
k8s.io/kubernetes/pkg/dns,jdef,1
k8s.io/kubernetes/pkg/dns/config,derekwaynecarr,0
k8s.io/kubernetes/pkg/dns/federation,derekwaynecarr,0
k8s.io/kubernetes/pkg/dns/treecache,bowei,0
k8s.io/kubernetes/pkg/fieldpath,childsb,1
k8s.io/kubernetes/pkg/fields,jsafrane,1
@ -956,4 +961,3 @@ k8s.io/kubernetes/test/integration/storageclasses,andyzheng0831,1
k8s.io/kubernetes/test/integration/thirdparty,davidopp,1
k8s.io/kubernetes/test/list,maisem,1
kubelet Clean up pods on node kubelet should be able to delete * pods per node in *.,yujuhong,0
Garbage Collection Test: * Should eventually garbage collect containers when we exceed the number of dead containers per container,Random-Liu,0

1 name owner auto-assigned
44 CronJob should not schedule new jobs when ForbidConcurrent soltysh 1
45 CronJob should replace jobs when ReplaceConcurrent soltysh 1
46 CronJob should schedule multiple jobs concurrently soltysh 1
47 DNS config map should be able to change configuration derekwaynecarr 0
48 DNS horizontal autoscaling kube-dns-autoscaler should scale kube-dns pods in both nonfaulty and faulty scenarios MrHohn 0
49 DNS horizontal autoscaling kube-dns-autoscaler should scale kube-dns pods when cluster size changed MrHohn 0
50 DNS should provide DNS for ExternalName services rmmh 1
176 Federation secrets Secret objects should be deleted from underlying clusters when OrphanDependents is false nikhiljindal 0
177 Federation secrets Secret objects should not be deleted from underlying clusters when OrphanDependents is nil nikhiljindal 0
178 Federation secrets Secret objects should not be deleted from underlying clusters when OrphanDependents is true nikhiljindal 0
GCP Volumes GlusterFS should be mountable rkouj 0
GCP Volumes NFSv4 should be mountable for NFSv4 rkouj 0
179 GCP Volumes GlusterFS should be mountable nikhiljindal 0
180 GCP Volumes NFSv4 should be mountable for NFSv4 nikhiljindal 0
181 GKE local SSD should write and read from node local SSD fabioy 0
182 GKE node pools should create a cluster with multiple node pools fabioy 1
183 Garbage Collection Test: * Should eventually garbage collect containers when we exceed the number of dead containers per container Random-Liu 0
184 Garbage collector should delete pods created by rc when not orphaning justinsb 1
185 Garbage collector should orphan pods created by rc if delete options say so fabioy 1
186 Garbage collector should orphan pods created by rc if deleteOptions.OrphanDependents is nil zmerlynn 1
187 Generated release_1_5 clientset should create pods, delete pods, watch pods ghodss 1
188 Generated release_1_5 clientset should create v2alpha1 cronJobs, delete cronJobs, watch cronJobs soltysh 1
189 HA-master pods survive addition/removal different zones HA-master survive addition/removal replicas different zones rkouj derekwaynecarr 0
190 HA-master pods survive addition/removal same zone HA-master survive addition/removal replicas same zone rkouj derekwaynecarr 0
HA-master pods survive addition/removal different zones nikhiljindal 0
HA-master pods survive addition/removal same zone nikhiljindal 0
191 Hazelcast should create and scale hazelcast mikedanese 1
192 Horizontal pod autoscaling (scale resource: CPU) Deployment Should scale from 1 pod to 3 pods and from 3 to 5 jszczepkowski 0
193 Horizontal pod autoscaling (scale resource: CPU) Deployment Should scale from 5 pods to 3 pods and from 3 to 1 jszczepkowski 0
260 Kubelet Container Manager Validate OOM score adjustments once the node is setup guaranteed container's oom-score-adj should be -998 kargakis 1
261 Kubelet Container Manager Validate OOM score adjustments once the node is setup pod infra containers oom-score-adj should be -998 and best effort container's should be 1000 timothysc 1
262 Kubelet Eviction Manager hard eviction test pod using the most disk space gets evicted when the node disk usage is above the eviction hard threshold should evict the pod using the most disk space karlkfi 1
263 Kubelet Volume Manager Volume Manager On terminatation of pod with memory backed volume should remove the volume from the node derekwaynecarr 0
264 Kubelet experimental resource usage tracking resource tracking for * pods per node yujuhong 0
265 Kubelet regular resource usage tracking resource tracking for * pods per node yujuhong 0
266 Kubelet when scheduling a busybox command in a pod it should print the output to logs ixdy 1
335 PersistentVolumes with multiple PVs and PVCs all in same ns should create 3 PVs and 3 PVCs: test write access caesarxuchao 1
336 PersistentVolumes with multiple PVs and PVCs all in same ns should create 4 PVs and 2 PVCs: test write access caesarxuchao 1
337 Pet Store should scale to persist a nominal number ( * ) of transactions in * seconds xiang90 1
Pet set recreate should recreate evicted statefulset roberthbailey 1
338 Pod Disks Should schedule a pod w/ a RW PD, gracefully remove it, then schedule it on another host alex-mohr 1
339 Pod Disks Should schedule a pod w/ a readonly PD on two hosts, then remove both gracefully. ghodss 1
340 Pod Disks should schedule a pod w/ a RW PD shared between multiple containers, write to PD, delete pod, verify contents, and repeat in rapid succession saad-ali 0
456 SimpleMount should be able to mount an emptydir on a container karlkfi 1
457 Spark should start spark master, driver and workers girishkalele 1
458 Staging client repo client should create pods, delete pods, watch pods jbeda 1
459 Stateful Set recreate should recreate evicted statefulset derekwaynecarr 0
460 StatefulSet Basic StatefulSet functionality Scaling down before scale up is finished should wait until current pod will be running and ready before it will be removed derekwaynecarr 0
461 StatefulSet Basic StatefulSet functionality Scaling should happen in predictable order and halt if any pet is unhealthy derekwaynecarr 0
462 StatefulSet Basic StatefulSet functionality should allow template updates derekwaynecarr 0
463 StatefulSet Basic StatefulSet functionality should handle healthy pet restarts during scale girishkalele 1
464 StatefulSet Basic StatefulSet functionality should provide basic identity bprashanth 1
465 StatefulSet Deploy clustered applications should creating a working CockroachDB cluster derekwaynecarr 0
466 StatefulSet Deploy clustered applications should creating a working mysql cluster yujuhong 1
467 StatefulSet Deploy clustered applications should creating a working redis cluster yifan-gu 1
468 StatefulSet Deploy clustered applications should creating a working zookeeper cluster pmorie 1
639 k8s.io/kubernetes/pkg/credentialprovider/aws zmerlynn 1
640 k8s.io/kubernetes/pkg/credentialprovider/gcp mml 1
641 k8s.io/kubernetes/pkg/dns jdef 1
642 k8s.io/kubernetes/pkg/dns/config derekwaynecarr 0
643 k8s.io/kubernetes/pkg/dns/federation derekwaynecarr 0
644 k8s.io/kubernetes/pkg/dns/treecache bowei 0
645 k8s.io/kubernetes/pkg/fieldpath childsb 1
646 k8s.io/kubernetes/pkg/fields jsafrane 1
961 k8s.io/kubernetes/test/integration/thirdparty davidopp 1
962 k8s.io/kubernetes/test/list maisem 1
963 kubelet Clean up pods on node kubelet should be able to delete * pods per node in *. yujuhong 0
Garbage Collection Test: * Should eventually garbage collect containers when we exceed the number of dead containers per container Random-Liu 0