-
Notifications
You must be signed in to change notification settings - Fork 40.5k
/
Copy pathkubelet.go
2618 lines (2297 loc) · 105 KB
/
kubelet.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
Copyright 2015 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 kubelet
import (
"context"
"crypto/tls"
"fmt"
"math"
"net"
"net/http"
"os"
"path"
"path/filepath"
sysruntime "runtime"
"sort"
"sync"
"sync/atomic"
"time"
"github.com/opencontainers/selinux/go-selinux"
"k8s.io/client-go/informers"
cadvisorapi "github.com/google/cadvisor/info/v1"
libcontaineruserns "github.com/opencontainers/runc/libcontainer/userns"
"go.opentelemetry.io/otel/trace"
"k8s.io/mount-utils"
"k8s.io/utils/integer"
netutils "k8s.io/utils/net"
v1 "k8s.io/api/core/v1"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/apimachinery/pkg/fields"
"k8s.io/apimachinery/pkg/labels"
"k8s.io/apimachinery/pkg/types"
utilruntime "k8s.io/apimachinery/pkg/util/runtime"
"k8s.io/apimachinery/pkg/util/sets"
"k8s.io/apimachinery/pkg/util/wait"
utilfeature "k8s.io/apiserver/pkg/util/feature"
clientset "k8s.io/client-go/kubernetes"
v1core "k8s.io/client-go/kubernetes/typed/core/v1"
corelisters "k8s.io/client-go/listers/core/v1"
"k8s.io/client-go/tools/cache"
"k8s.io/client-go/tools/record"
"k8s.io/client-go/util/certificate"
"k8s.io/client-go/util/flowcontrol"
cloudprovider "k8s.io/cloud-provider"
"k8s.io/component-helpers/apimachinery/lease"
internalapi "k8s.io/cri-api/pkg/apis"
runtimeapi "k8s.io/cri-api/pkg/apis/runtime/v1"
"k8s.io/klog/v2"
pluginwatcherapi "k8s.io/kubelet/pkg/apis/pluginregistration/v1"
statsapi "k8s.io/kubelet/pkg/apis/stats/v1alpha1"
"k8s.io/kubernetes/pkg/features"
kubeletconfiginternal "k8s.io/kubernetes/pkg/kubelet/apis/config"
"k8s.io/kubernetes/pkg/kubelet/apis/podresources"
"k8s.io/kubernetes/pkg/kubelet/cadvisor"
kubeletcertificate "k8s.io/kubernetes/pkg/kubelet/certificate"
"k8s.io/kubernetes/pkg/kubelet/cloudresource"
"k8s.io/kubernetes/pkg/kubelet/cm"
draplugin "k8s.io/kubernetes/pkg/kubelet/cm/dra/plugin"
"k8s.io/kubernetes/pkg/kubelet/config"
"k8s.io/kubernetes/pkg/kubelet/configmap"
kubecontainer "k8s.io/kubernetes/pkg/kubelet/container"
"k8s.io/kubernetes/pkg/kubelet/cri/remote"
"k8s.io/kubernetes/pkg/kubelet/events"
"k8s.io/kubernetes/pkg/kubelet/eviction"
"k8s.io/kubernetes/pkg/kubelet/images"
"k8s.io/kubernetes/pkg/kubelet/kuberuntime"
"k8s.io/kubernetes/pkg/kubelet/lifecycle"
"k8s.io/kubernetes/pkg/kubelet/logs"
"k8s.io/kubernetes/pkg/kubelet/metrics"
"k8s.io/kubernetes/pkg/kubelet/metrics/collectors"
"k8s.io/kubernetes/pkg/kubelet/network/dns"
"k8s.io/kubernetes/pkg/kubelet/nodeshutdown"
oomwatcher "k8s.io/kubernetes/pkg/kubelet/oom"
"k8s.io/kubernetes/pkg/kubelet/pleg"
"k8s.io/kubernetes/pkg/kubelet/pluginmanager"
plugincache "k8s.io/kubernetes/pkg/kubelet/pluginmanager/cache"
kubepod "k8s.io/kubernetes/pkg/kubelet/pod"
"k8s.io/kubernetes/pkg/kubelet/preemption"
"k8s.io/kubernetes/pkg/kubelet/prober"
proberesults "k8s.io/kubernetes/pkg/kubelet/prober/results"
"k8s.io/kubernetes/pkg/kubelet/runtimeclass"
"k8s.io/kubernetes/pkg/kubelet/secret"
"k8s.io/kubernetes/pkg/kubelet/server"
servermetrics "k8s.io/kubernetes/pkg/kubelet/server/metrics"
serverstats "k8s.io/kubernetes/pkg/kubelet/server/stats"
"k8s.io/kubernetes/pkg/kubelet/stats"
"k8s.io/kubernetes/pkg/kubelet/status"
"k8s.io/kubernetes/pkg/kubelet/sysctl"
"k8s.io/kubernetes/pkg/kubelet/token"
kubetypes "k8s.io/kubernetes/pkg/kubelet/types"
"k8s.io/kubernetes/pkg/kubelet/util"
"k8s.io/kubernetes/pkg/kubelet/util/manager"
"k8s.io/kubernetes/pkg/kubelet/util/queue"
"k8s.io/kubernetes/pkg/kubelet/util/sliceutils"
"k8s.io/kubernetes/pkg/kubelet/volumemanager"
httpprobe "k8s.io/kubernetes/pkg/probe/http"
"k8s.io/kubernetes/pkg/security/apparmor"
"k8s.io/kubernetes/pkg/util/oom"
"k8s.io/kubernetes/pkg/volume"
"k8s.io/kubernetes/pkg/volume/csi"
"k8s.io/kubernetes/pkg/volume/util/hostutil"
"k8s.io/kubernetes/pkg/volume/util/subpath"
"k8s.io/kubernetes/pkg/volume/util/volumepathhandler"
"k8s.io/utils/clock"
)
const (
// Max amount of time to wait for the container runtime to come up.
maxWaitForContainerRuntime = 30 * time.Second
// nodeStatusUpdateRetry specifies how many times kubelet retries when posting node status failed.
nodeStatusUpdateRetry = 5
// nodeReadyGracePeriod is the period to allow for before fast status update is
// terminated and container runtime not being ready is logged without verbosity guard.
nodeReadyGracePeriod = 120 * time.Second
// DefaultContainerLogsDir is the location of container logs.
DefaultContainerLogsDir = "/var/log/containers"
// MaxContainerBackOff is the max backoff period, exported for the e2e test
MaxContainerBackOff = 300 * time.Second
// Period for performing global cleanup tasks.
housekeepingPeriod = time.Second * 2
// Duration at which housekeeping failed to satisfy the invariant that
// housekeeping should be fast to avoid blocking pod config (while
// housekeeping is running no new pods are started or deleted).
housekeepingWarningDuration = time.Second * 15
// Period for performing eviction monitoring.
// ensure this is kept in sync with internal cadvisor housekeeping.
evictionMonitoringPeriod = time.Second * 10
// The path in containers' filesystems where the hosts file is mounted.
linuxEtcHostsPath = "/etc/hosts"
windowsEtcHostsPath = "C:\\Windows\\System32\\drivers\\etc\\hosts"
// Capacity of the channel for receiving pod lifecycle events. This number
// is a bit arbitrary and may be adjusted in the future.
plegChannelCapacity = 1000
// Generic PLEG relies on relisting for discovering container events.
// A longer period means that kubelet will take longer to detect container
// changes and to update pod status. On the other hand, a shorter period
// will cause more frequent relisting (e.g., container runtime operations),
// leading to higher cpu usage.
// Note that even though we set the period to 1s, the relisting itself can
// take more than 1s to finish if the container runtime responds slowly
// and/or when there are many container changes in one cycle.
genericPlegRelistPeriod = time.Second * 1
genericPlegRelistThreshold = time.Minute * 3
// Generic PLEG relist period and threshold when used with Evented PLEG.
eventedPlegRelistPeriod = time.Second * 300
eventedPlegRelistThreshold = time.Minute * 10
eventedPlegMaxStreamRetries = 5
// backOffPeriod is the period to back off when pod syncing results in an
// error. It is also used as the base period for the exponential backoff
// container restarts and image pulls.
backOffPeriod = time.Second * 10
// ContainerGCPeriod is the period for performing container garbage collection.
ContainerGCPeriod = time.Minute
// ImageGCPeriod is the period for performing image garbage collection.
ImageGCPeriod = 5 * time.Minute
// Minimum number of dead containers to keep in a pod
minDeadContainerInPod = 1
// nodeLeaseRenewIntervalFraction is the fraction of lease duration to renew the lease
nodeLeaseRenewIntervalFraction = 0.25
)
var (
// ContainerLogsDir can be overwrited for testing usage
ContainerLogsDir = DefaultContainerLogsDir
etcHostsPath = getContainerEtcHostsPath()
)
func getContainerEtcHostsPath() string {
if sysruntime.GOOS == "windows" {
return windowsEtcHostsPath
}
return linuxEtcHostsPath
}
// SyncHandler is an interface implemented by Kubelet, for testability
type SyncHandler interface {
HandlePodAdditions(pods []*v1.Pod)
HandlePodUpdates(pods []*v1.Pod)
HandlePodRemoves(pods []*v1.Pod)
HandlePodReconcile(pods []*v1.Pod)
HandlePodSyncs(pods []*v1.Pod)
HandlePodCleanups(ctx context.Context) error
}
// Option is a functional option type for Kubelet
type Option func(*Kubelet)
// Bootstrap is a bootstrapping interface for kubelet, targets the initialization protocol
type Bootstrap interface {
GetConfiguration() kubeletconfiginternal.KubeletConfiguration
BirthCry()
StartGarbageCollection()
ListenAndServe(kubeCfg *kubeletconfiginternal.KubeletConfiguration, tlsOptions *server.TLSOptions, auth server.AuthInterface, tp trace.TracerProvider)
ListenAndServeReadOnly(address net.IP, port uint)
ListenAndServePodResources()
Run(<-chan kubetypes.PodUpdate)
RunOnce(<-chan kubetypes.PodUpdate) ([]RunPodResult, error)
}
// Dependencies is a bin for things we might consider "injected dependencies" -- objects constructed
// at runtime that are necessary for running the Kubelet. This is a temporary solution for grouping
// these objects while we figure out a more comprehensive dependency injection story for the Kubelet.
type Dependencies struct {
Options []Option
// Injected Dependencies
Auth server.AuthInterface
CAdvisorInterface cadvisor.Interface
Cloud cloudprovider.Interface
ContainerManager cm.ContainerManager
EventClient v1core.EventsGetter
HeartbeatClient clientset.Interface
OnHeartbeatFailure func()
KubeClient clientset.Interface
Mounter mount.Interface
HostUtil hostutil.HostUtils
OOMAdjuster *oom.OOMAdjuster
OSInterface kubecontainer.OSInterface
PodConfig *config.PodConfig
ProbeManager prober.Manager
Recorder record.EventRecorder
Subpather subpath.Interface
TracerProvider trace.TracerProvider
VolumePlugins []volume.VolumePlugin
DynamicPluginProber volume.DynamicPluginProber
TLSOptions *server.TLSOptions
RemoteRuntimeService internalapi.RuntimeService
RemoteImageService internalapi.ImageManagerService
PodStartupLatencyTracker util.PodStartupLatencyTracker
// remove it after cadvisor.UsingLegacyCadvisorStats dropped.
useLegacyCadvisorStats bool
}
// makePodSourceConfig creates a config.PodConfig from the given
// KubeletConfiguration or returns an error.
func makePodSourceConfig(kubeCfg *kubeletconfiginternal.KubeletConfiguration, kubeDeps *Dependencies, nodeName types.NodeName, nodeHasSynced func() bool) (*config.PodConfig, error) {
manifestURLHeader := make(http.Header)
if len(kubeCfg.StaticPodURLHeader) > 0 {
for k, v := range kubeCfg.StaticPodURLHeader {
for i := range v {
manifestURLHeader.Add(k, v[i])
}
}
}
// source of all configuration
cfg := config.NewPodConfig(config.PodConfigNotificationIncremental, kubeDeps.Recorder, kubeDeps.PodStartupLatencyTracker)
// TODO: it needs to be replaced by a proper context in the future
ctx := context.TODO()
// define file config source
if kubeCfg.StaticPodPath != "" {
klog.InfoS("Adding static pod path", "path", kubeCfg.StaticPodPath)
config.NewSourceFile(kubeCfg.StaticPodPath, nodeName, kubeCfg.FileCheckFrequency.Duration, cfg.Channel(ctx, kubetypes.FileSource))
}
// define url config source
if kubeCfg.StaticPodURL != "" {
klog.InfoS("Adding pod URL with HTTP header", "URL", kubeCfg.StaticPodURL, "header", manifestURLHeader)
config.NewSourceURL(kubeCfg.StaticPodURL, manifestURLHeader, nodeName, kubeCfg.HTTPCheckFrequency.Duration, cfg.Channel(ctx, kubetypes.HTTPSource))
}
if kubeDeps.KubeClient != nil {
klog.InfoS("Adding apiserver pod source")
config.NewSourceApiserver(kubeDeps.KubeClient, nodeName, nodeHasSynced, cfg.Channel(ctx, kubetypes.ApiserverSource))
}
return cfg, nil
}
// PreInitRuntimeService will init runtime service before RunKubelet.
func PreInitRuntimeService(kubeCfg *kubeletconfiginternal.KubeletConfiguration,
kubeDeps *Dependencies,
remoteRuntimeEndpoint string,
remoteImageEndpoint string) error {
// remoteImageEndpoint is same as remoteRuntimeEndpoint if not explicitly specified
if remoteRuntimeEndpoint != "" && remoteImageEndpoint == "" {
remoteImageEndpoint = remoteRuntimeEndpoint
}
var err error
if kubeDeps.RemoteRuntimeService, err = remote.NewRemoteRuntimeService(remoteRuntimeEndpoint, kubeCfg.RuntimeRequestTimeout.Duration, kubeDeps.TracerProvider); err != nil {
return err
}
if kubeDeps.RemoteImageService, err = remote.NewRemoteImageService(remoteImageEndpoint, kubeCfg.RuntimeRequestTimeout.Duration, kubeDeps.TracerProvider); err != nil {
return err
}
kubeDeps.useLegacyCadvisorStats = cadvisor.UsingLegacyCadvisorStats(remoteRuntimeEndpoint)
return nil
}
// NewMainKubelet instantiates a new Kubelet object along with all the required internal modules.
// No initialization of Kubelet and its modules should happen here.
func NewMainKubelet(kubeCfg *kubeletconfiginternal.KubeletConfiguration,
kubeDeps *Dependencies,
crOptions *config.ContainerRuntimeOptions,
hostname string,
hostnameOverridden bool,
nodeName types.NodeName,
nodeIPs []net.IP,
providerID string,
cloudProvider string,
certDirectory string,
rootDirectory string,
imageCredentialProviderConfigFile string,
imageCredentialProviderBinDir string,
registerNode bool,
registerWithTaints []v1.Taint,
allowedUnsafeSysctls []string,
experimentalMounterPath string,
kernelMemcgNotification bool,
experimentalNodeAllocatableIgnoreEvictionThreshold bool,
minimumGCAge metav1.Duration,
maxPerPodContainerCount int32,
maxContainerCount int32,
masterServiceNamespace string,
registerSchedulable bool,
keepTerminatedPodVolumes bool,
nodeLabels map[string]string,
nodeStatusMaxImages int32,
seccompDefault bool,
) (*Kubelet, error) {
ctx := context.Background()
logger := klog.TODO()
if rootDirectory == "" {
return nil, fmt.Errorf("invalid root directory %q", rootDirectory)
}
if kubeCfg.SyncFrequency.Duration <= 0 {
return nil, fmt.Errorf("invalid sync frequency %d", kubeCfg.SyncFrequency.Duration)
}
if kubeCfg.MakeIPTablesUtilChains {
if kubeCfg.IPTablesMasqueradeBit > 31 || kubeCfg.IPTablesMasqueradeBit < 0 {
return nil, fmt.Errorf("iptables-masquerade-bit is not valid. Must be within [0, 31]")
}
if kubeCfg.IPTablesDropBit > 31 || kubeCfg.IPTablesDropBit < 0 {
return nil, fmt.Errorf("iptables-drop-bit is not valid. Must be within [0, 31]")
}
if kubeCfg.IPTablesDropBit == kubeCfg.IPTablesMasqueradeBit {
return nil, fmt.Errorf("iptables-masquerade-bit and iptables-drop-bit must be different")
}
}
if utilfeature.DefaultFeatureGate.Enabled(features.DisableCloudProviders) && cloudprovider.IsDeprecatedInternal(cloudProvider) {
cloudprovider.DisableWarningForProvider(cloudProvider)
return nil, fmt.Errorf("cloud provider %q was specified, but built-in cloud providers are disabled. Please set --cloud-provider=external and migrate to an external cloud provider", cloudProvider)
}
var nodeHasSynced cache.InformerSynced
var nodeLister corelisters.NodeLister
// If kubeClient == nil, we are running in standalone mode (i.e. no API servers)
// If not nil, we are running as part of a cluster and should sync w/API
if kubeDeps.KubeClient != nil {
kubeInformers := informers.NewSharedInformerFactoryWithOptions(kubeDeps.KubeClient, 0, informers.WithTweakListOptions(func(options *metav1.ListOptions) {
options.FieldSelector = fields.Set{metav1.ObjectNameField: string(nodeName)}.String()
}))
nodeLister = kubeInformers.Core().V1().Nodes().Lister()
nodeHasSynced = func() bool {
return kubeInformers.Core().V1().Nodes().Informer().HasSynced()
}
kubeInformers.Start(wait.NeverStop)
klog.InfoS("Attempting to sync node with API server")
} else {
// we don't have a client to sync!
nodeIndexer := cache.NewIndexer(cache.MetaNamespaceKeyFunc, cache.Indexers{})
nodeLister = corelisters.NewNodeLister(nodeIndexer)
nodeHasSynced = func() bool { return true }
klog.InfoS("Kubelet is running in standalone mode, will skip API server sync")
}
if kubeDeps.PodConfig == nil {
var err error
kubeDeps.PodConfig, err = makePodSourceConfig(kubeCfg, kubeDeps, nodeName, nodeHasSynced)
if err != nil {
return nil, err
}
}
containerGCPolicy := kubecontainer.GCPolicy{
MinAge: minimumGCAge.Duration,
MaxPerPodContainer: int(maxPerPodContainerCount),
MaxContainers: int(maxContainerCount),
}
daemonEndpoints := &v1.NodeDaemonEndpoints{
KubeletEndpoint: v1.DaemonEndpoint{Port: kubeCfg.Port},
}
imageGCPolicy := images.ImageGCPolicy{
MinAge: kubeCfg.ImageMinimumGCAge.Duration,
HighThresholdPercent: int(kubeCfg.ImageGCHighThresholdPercent),
LowThresholdPercent: int(kubeCfg.ImageGCLowThresholdPercent),
}
enforceNodeAllocatable := kubeCfg.EnforceNodeAllocatable
if experimentalNodeAllocatableIgnoreEvictionThreshold {
// Do not provide kubeCfg.EnforceNodeAllocatable to eviction threshold parsing if we are not enforcing Evictions
enforceNodeAllocatable = []string{}
}
thresholds, err := eviction.ParseThresholdConfig(enforceNodeAllocatable, kubeCfg.EvictionHard, kubeCfg.EvictionSoft, kubeCfg.EvictionSoftGracePeriod, kubeCfg.EvictionMinimumReclaim)
if err != nil {
return nil, err
}
evictionConfig := eviction.Config{
PressureTransitionPeriod: kubeCfg.EvictionPressureTransitionPeriod.Duration,
MaxPodGracePeriodSeconds: int64(kubeCfg.EvictionMaxPodGracePeriod),
Thresholds: thresholds,
KernelMemcgNotification: kernelMemcgNotification,
PodCgroupRoot: kubeDeps.ContainerManager.GetPodCgroupRoot(),
}
var serviceLister corelisters.ServiceLister
var serviceHasSynced cache.InformerSynced
if kubeDeps.KubeClient != nil {
kubeInformers := informers.NewSharedInformerFactory(kubeDeps.KubeClient, 0)
serviceLister = kubeInformers.Core().V1().Services().Lister()
serviceHasSynced = kubeInformers.Core().V1().Services().Informer().HasSynced
kubeInformers.Start(wait.NeverStop)
} else {
serviceIndexer := cache.NewIndexer(cache.MetaNamespaceKeyFunc, cache.Indexers{cache.NamespaceIndex: cache.MetaNamespaceIndexFunc})
serviceLister = corelisters.NewServiceLister(serviceIndexer)
serviceHasSynced = func() bool { return true }
}
// construct a node reference used for events
nodeRef := &v1.ObjectReference{
Kind: "Node",
Name: string(nodeName),
UID: types.UID(nodeName),
Namespace: "",
}
oomWatcher, err := oomwatcher.NewWatcher(kubeDeps.Recorder)
if err != nil {
if libcontaineruserns.RunningInUserNS() {
if utilfeature.DefaultFeatureGate.Enabled(features.KubeletInUserNamespace) {
// oomwatcher.NewWatcher returns "open /dev/kmsg: operation not permitted" error,
// when running in a user namespace with sysctl value `kernel.dmesg_restrict=1`.
klog.V(2).InfoS("Failed to create an oomWatcher (running in UserNS, ignoring)", "err", err)
oomWatcher = nil
} else {
klog.ErrorS(err, "Failed to create an oomWatcher (running in UserNS, Hint: enable KubeletInUserNamespace feature flag to ignore the error)")
return nil, err
}
} else {
return nil, err
}
}
clusterDNS := make([]net.IP, 0, len(kubeCfg.ClusterDNS))
for _, ipEntry := range kubeCfg.ClusterDNS {
ip := netutils.ParseIPSloppy(ipEntry)
if ip == nil {
klog.InfoS("Invalid clusterDNS IP", "IP", ipEntry)
} else {
clusterDNS = append(clusterDNS, ip)
}
}
// A TLS transport is needed to make HTTPS-based container lifecycle requests,
// but we do not have the information necessary to do TLS verification.
//
// This client must not be modified to include credentials, because it is
// critical that credentials not leak from the client to arbitrary hosts.
insecureContainerLifecycleHTTPClient := &http.Client{}
if utilfeature.DefaultFeatureGate.Enabled(features.ConsistentHTTPGetHandlers) {
insecureTLSTransport := &http.Transport{
TLSClientConfig: &tls.Config{InsecureSkipVerify: true},
}
insecureContainerLifecycleHTTPClient.Transport = insecureTLSTransport
insecureContainerLifecycleHTTPClient.CheckRedirect = httpprobe.RedirectChecker(false)
}
klet := &Kubelet{
hostname: hostname,
hostnameOverridden: hostnameOverridden,
nodeName: nodeName,
kubeClient: kubeDeps.KubeClient,
heartbeatClient: kubeDeps.HeartbeatClient,
onRepeatedHeartbeatFailure: kubeDeps.OnHeartbeatFailure,
rootDirectory: rootDirectory,
resyncInterval: kubeCfg.SyncFrequency.Duration,
sourcesReady: config.NewSourcesReady(kubeDeps.PodConfig.SeenAllSources),
registerNode: registerNode,
registerWithTaints: registerWithTaints,
registerSchedulable: registerSchedulable,
dnsConfigurer: dns.NewConfigurer(kubeDeps.Recorder, nodeRef, nodeIPs, clusterDNS, kubeCfg.ClusterDomain, kubeCfg.ResolverConfig),
serviceLister: serviceLister,
serviceHasSynced: serviceHasSynced,
nodeLister: nodeLister,
nodeHasSynced: nodeHasSynced,
masterServiceNamespace: masterServiceNamespace,
streamingConnectionIdleTimeout: kubeCfg.StreamingConnectionIdleTimeout.Duration,
recorder: kubeDeps.Recorder,
cadvisor: kubeDeps.CAdvisorInterface,
cloud: kubeDeps.Cloud,
externalCloudProvider: cloudprovider.IsExternal(cloudProvider),
providerID: providerID,
nodeRef: nodeRef,
nodeLabels: nodeLabels,
nodeStatusUpdateFrequency: kubeCfg.NodeStatusUpdateFrequency.Duration,
nodeStatusReportFrequency: kubeCfg.NodeStatusReportFrequency.Duration,
os: kubeDeps.OSInterface,
oomWatcher: oomWatcher,
cgroupsPerQOS: kubeCfg.CgroupsPerQOS,
cgroupRoot: kubeCfg.CgroupRoot,
mounter: kubeDeps.Mounter,
hostutil: kubeDeps.HostUtil,
subpather: kubeDeps.Subpather,
maxPods: int(kubeCfg.MaxPods),
podsPerCore: int(kubeCfg.PodsPerCore),
syncLoopMonitor: atomic.Value{},
daemonEndpoints: daemonEndpoints,
containerManager: kubeDeps.ContainerManager,
nodeIPs: nodeIPs,
nodeIPValidator: validateNodeIP,
clock: clock.RealClock{},
enableControllerAttachDetach: kubeCfg.EnableControllerAttachDetach,
makeIPTablesUtilChains: kubeCfg.MakeIPTablesUtilChains,
iptablesMasqueradeBit: int(kubeCfg.IPTablesMasqueradeBit),
iptablesDropBit: int(kubeCfg.IPTablesDropBit),
experimentalHostUserNamespaceDefaulting: utilfeature.DefaultFeatureGate.Enabled(features.ExperimentalHostUserNamespaceDefaultingGate),
keepTerminatedPodVolumes: keepTerminatedPodVolumes,
nodeStatusMaxImages: nodeStatusMaxImages,
}
if klet.cloud != nil {
klet.cloudResourceSyncManager = cloudresource.NewSyncManager(klet.cloud, nodeName, klet.nodeStatusUpdateFrequency)
}
var secretManager secret.Manager
var configMapManager configmap.Manager
if klet.kubeClient != nil {
switch kubeCfg.ConfigMapAndSecretChangeDetectionStrategy {
case kubeletconfiginternal.WatchChangeDetectionStrategy:
secretManager = secret.NewWatchingSecretManager(klet.kubeClient, klet.resyncInterval)
configMapManager = configmap.NewWatchingConfigMapManager(klet.kubeClient, klet.resyncInterval)
case kubeletconfiginternal.TTLCacheChangeDetectionStrategy:
secretManager = secret.NewCachingSecretManager(
klet.kubeClient, manager.GetObjectTTLFromNodeFunc(klet.GetNode))
configMapManager = configmap.NewCachingConfigMapManager(
klet.kubeClient, manager.GetObjectTTLFromNodeFunc(klet.GetNode))
case kubeletconfiginternal.GetChangeDetectionStrategy:
secretManager = secret.NewSimpleSecretManager(klet.kubeClient)
configMapManager = configmap.NewSimpleConfigMapManager(klet.kubeClient)
default:
return nil, fmt.Errorf("unknown configmap and secret manager mode: %v", kubeCfg.ConfigMapAndSecretChangeDetectionStrategy)
}
klet.secretManager = secretManager
klet.configMapManager = configMapManager
}
if klet.experimentalHostUserNamespaceDefaulting {
klog.InfoS("Experimental host user namespace defaulting is enabled")
}
machineInfo, err := klet.cadvisor.MachineInfo()
if err != nil {
return nil, err
}
// Avoid collector collects it as a timestamped metric
// See PR #95210 and #97006 for more details.
machineInfo.Timestamp = time.Time{}
klet.setCachedMachineInfo(machineInfo)
imageBackOff := flowcontrol.NewBackOff(backOffPeriod, MaxContainerBackOff)
klet.livenessManager = proberesults.NewManager()
klet.readinessManager = proberesults.NewManager()
klet.startupManager = proberesults.NewManager()
klet.podCache = kubecontainer.NewCache()
// podManager is also responsible for keeping secretManager and configMapManager contents up-to-date.
mirrorPodClient := kubepod.NewBasicMirrorClient(klet.kubeClient, string(nodeName), nodeLister)
klet.podManager = kubepod.NewBasicPodManager(mirrorPodClient, secretManager, configMapManager)
klet.statusManager = status.NewManager(klet.kubeClient, klet.podManager, klet, kubeDeps.PodStartupLatencyTracker)
klet.resourceAnalyzer = serverstats.NewResourceAnalyzer(klet, kubeCfg.VolumeStatsAggPeriod.Duration, kubeDeps.Recorder)
klet.runtimeService = kubeDeps.RemoteRuntimeService
if kubeDeps.KubeClient != nil {
klet.runtimeClassManager = runtimeclass.NewManager(kubeDeps.KubeClient)
}
// setup containerLogManager for CRI container runtime
containerLogManager, err := logs.NewContainerLogManager(
klet.runtimeService,
kubeDeps.OSInterface,
kubeCfg.ContainerLogMaxSize,
int(kubeCfg.ContainerLogMaxFiles),
)
if err != nil {
return nil, fmt.Errorf("failed to initialize container log manager: %v", err)
}
klet.containerLogManager = containerLogManager
klet.reasonCache = NewReasonCache()
klet.workQueue = queue.NewBasicWorkQueue(klet.clock)
klet.podWorkers = newPodWorkers(
klet.syncPod,
klet.syncTerminatingPod,
klet.syncTerminatedPod,
kubeDeps.Recorder,
klet.workQueue,
klet.resyncInterval,
backOffPeriod,
klet.podCache,
)
runtime, err := kuberuntime.NewKubeGenericRuntimeManager(
kubecontainer.FilterEventRecorder(kubeDeps.Recorder),
klet.livenessManager,
klet.readinessManager,
klet.startupManager,
rootDirectory,
machineInfo,
klet.podWorkers,
kubeDeps.OSInterface,
klet,
insecureContainerLifecycleHTTPClient,
imageBackOff,
kubeCfg.SerializeImagePulls,
float32(kubeCfg.RegistryPullQPS),
int(kubeCfg.RegistryBurst),
imageCredentialProviderConfigFile,
imageCredentialProviderBinDir,
kubeCfg.CPUCFSQuota,
kubeCfg.CPUCFSQuotaPeriod,
kubeDeps.RemoteRuntimeService,
kubeDeps.RemoteImageService,
kubeDeps.ContainerManager.InternalContainerLifecycle(),
klet.containerLogManager,
klet.runtimeClassManager,
seccompDefault,
kubeCfg.MemorySwap.SwapBehavior,
kubeDeps.ContainerManager.GetNodeAllocatableAbsolute,
*kubeCfg.MemoryThrottlingFactor,
kubeDeps.PodStartupLatencyTracker,
)
if err != nil {
return nil, err
}
klet.containerRuntime = runtime
klet.streamingRuntime = runtime
klet.runner = runtime
runtimeCache, err := kubecontainer.NewRuntimeCache(klet.containerRuntime)
if err != nil {
return nil, err
}
klet.runtimeCache = runtimeCache
// common provider to get host file system usage associated with a pod managed by kubelet
hostStatsProvider := stats.NewHostStatsProvider(kubecontainer.RealOS{}, func(podUID types.UID) string {
return getEtcHostsPath(klet.getPodDir(podUID))
})
if kubeDeps.useLegacyCadvisorStats {
klet.StatsProvider = stats.NewCadvisorStatsProvider(
klet.cadvisor,
klet.resourceAnalyzer,
klet.podManager,
klet.runtimeCache,
klet.containerRuntime,
klet.statusManager,
hostStatsProvider)
} else {
klet.StatsProvider = stats.NewCRIStatsProvider(
klet.cadvisor,
klet.resourceAnalyzer,
klet.podManager,
klet.runtimeCache,
kubeDeps.RemoteRuntimeService,
kubeDeps.RemoteImageService,
hostStatsProvider,
utilfeature.DefaultFeatureGate.Enabled(features.PodAndContainerStatsFromCRI))
}
eventChannel := make(chan *pleg.PodLifecycleEvent, plegChannelCapacity)
if utilfeature.DefaultFeatureGate.Enabled(features.EventedPLEG) {
// adjust Generic PLEG relisting period and threshold to higher value when Evented PLEG is turned on
genericRelistDuration := &pleg.RelistDuration{
RelistPeriod: eventedPlegRelistPeriod,
RelistThreshold: eventedPlegRelistThreshold,
}
klet.pleg = pleg.NewGenericPLEG(klet.containerRuntime, eventChannel, genericRelistDuration, klet.podCache, clock.RealClock{})
// In case Evented PLEG has to fall back on Generic PLEG due to an error,
// Evented PLEG should be able to reset the Generic PLEG relisting duration
// to the default value.
eventedRelistDuration := &pleg.RelistDuration{
RelistPeriod: genericPlegRelistPeriod,
RelistThreshold: genericPlegRelistThreshold,
}
klet.eventedPleg = pleg.NewEventedPLEG(klet.containerRuntime, klet.runtimeService, eventChannel,
klet.podCache, klet.pleg, eventedPlegMaxStreamRetries, eventedRelistDuration, clock.RealClock{})
} else {
genericRelistDuration := &pleg.RelistDuration{
RelistPeriod: genericPlegRelistPeriod,
RelistThreshold: genericPlegRelistThreshold,
}
klet.pleg = pleg.NewGenericPLEG(klet.containerRuntime, eventChannel, genericRelistDuration, klet.podCache, clock.RealClock{})
}
klet.runtimeState = newRuntimeState(maxWaitForContainerRuntime)
klet.runtimeState.addHealthCheck("PLEG", klet.pleg.Healthy)
if utilfeature.DefaultFeatureGate.Enabled(features.EventedPLEG) {
klet.runtimeState.addHealthCheck("EventedPLEG", klet.eventedPleg.Healthy)
}
if _, err := klet.updatePodCIDR(ctx, kubeCfg.PodCIDR); err != nil {
klog.ErrorS(err, "Pod CIDR update failed")
}
// setup containerGC
containerGC, err := kubecontainer.NewContainerGC(klet.containerRuntime, containerGCPolicy, klet.sourcesReady)
if err != nil {
return nil, err
}
klet.containerGC = containerGC
klet.containerDeletor = newPodContainerDeletor(klet.containerRuntime, integer.IntMax(containerGCPolicy.MaxPerPodContainer, minDeadContainerInPod))
// setup imageManager
imageManager, err := images.NewImageGCManager(klet.containerRuntime, klet.StatsProvider, kubeDeps.Recorder, nodeRef, imageGCPolicy, crOptions.PodSandboxImage)
if err != nil {
return nil, fmt.Errorf("failed to initialize image manager: %v", err)
}
klet.imageManager = imageManager
if kubeCfg.ServerTLSBootstrap && kubeDeps.TLSOptions != nil && utilfeature.DefaultFeatureGate.Enabled(features.RotateKubeletServerCertificate) {
klet.serverCertificateManager, err = kubeletcertificate.NewKubeletServerCertificateManager(klet.kubeClient, kubeCfg, klet.nodeName, klet.getLastObservedNodeAddresses, certDirectory)
if err != nil {
return nil, fmt.Errorf("failed to initialize certificate manager: %v", err)
}
kubeDeps.TLSOptions.Config.GetCertificate = func(*tls.ClientHelloInfo) (*tls.Certificate, error) {
cert := klet.serverCertificateManager.Current()
if cert == nil {
return nil, fmt.Errorf("no serving certificate available for the kubelet")
}
return cert, nil
}
}
if kubeDeps.ProbeManager != nil {
klet.probeManager = kubeDeps.ProbeManager
} else {
klet.probeManager = prober.NewManager(
klet.statusManager,
klet.livenessManager,
klet.readinessManager,
klet.startupManager,
klet.runner,
kubeDeps.Recorder)
}
tokenManager := token.NewManager(kubeDeps.KubeClient)
// NewInitializedVolumePluginMgr initializes some storageErrors on the Kubelet runtimeState (in csi_plugin.go init)
// which affects node ready status. This function must be called before Kubelet is initialized so that the Node
// ReadyState is accurate with the storage state.
klet.volumePluginMgr, err =
NewInitializedVolumePluginMgr(klet, secretManager, configMapManager, tokenManager, kubeDeps.VolumePlugins, kubeDeps.DynamicPluginProber)
if err != nil {
return nil, err
}
klet.pluginManager = pluginmanager.NewPluginManager(
klet.getPluginsRegistrationDir(), /* sockDir */
kubeDeps.Recorder,
)
// If the experimentalMounterPathFlag is set, we do not want to
// check node capabilities since the mount path is not the default
if len(experimentalMounterPath) != 0 {
// Replace the nameserver in containerized-mounter's rootfs/etc/resolv.conf with kubelet.ClusterDNS
// so that service name could be resolved
klet.dnsConfigurer.SetupDNSinContainerizedMounter(experimentalMounterPath)
}
// setup volumeManager
klet.volumeManager = volumemanager.NewVolumeManager(
kubeCfg.EnableControllerAttachDetach,
nodeName,
klet.podManager,
klet.podWorkers,
klet.kubeClient,
klet.volumePluginMgr,
klet.containerRuntime,
kubeDeps.Mounter,
kubeDeps.HostUtil,
klet.getPodsDir(),
kubeDeps.Recorder,
keepTerminatedPodVolumes,
volumepathhandler.NewBlockVolumePathHandler())
klet.backOff = flowcontrol.NewBackOff(backOffPeriod, MaxContainerBackOff)
// setup eviction manager
evictionManager, evictionAdmitHandler := eviction.NewManager(klet.resourceAnalyzer, evictionConfig,
killPodNow(klet.podWorkers, kubeDeps.Recorder), klet.podManager.GetMirrorPodByPod, klet.imageManager, klet.containerGC, kubeDeps.Recorder, nodeRef, klet.clock, kubeCfg.LocalStorageCapacityIsolation)
klet.evictionManager = evictionManager
klet.admitHandlers.AddPodAdmitHandler(evictionAdmitHandler)
// Safe, allowed sysctls can always be used as unsafe sysctls in the spec.
// Hence, we concatenate those two lists.
safeAndUnsafeSysctls := append(sysctl.SafeSysctlAllowlist(), allowedUnsafeSysctls...)
sysctlsAllowlist, err := sysctl.NewAllowlist(safeAndUnsafeSysctls)
if err != nil {
return nil, err
}
klet.admitHandlers.AddPodAdmitHandler(sysctlsAllowlist)
// enable active deadline handler
activeDeadlineHandler, err := newActiveDeadlineHandler(klet.statusManager, kubeDeps.Recorder, klet.clock)
if err != nil {
return nil, err
}
klet.AddPodSyncLoopHandler(activeDeadlineHandler)
klet.AddPodSyncHandler(activeDeadlineHandler)
klet.admitHandlers.AddPodAdmitHandler(klet.containerManager.GetAllocateResourcesPodAdmitHandler())
criticalPodAdmissionHandler := preemption.NewCriticalPodAdmissionHandler(klet.GetActivePods, killPodNow(klet.podWorkers, kubeDeps.Recorder), kubeDeps.Recorder)
klet.admitHandlers.AddPodAdmitHandler(lifecycle.NewPredicateAdmitHandler(klet.getNodeAnyWay, criticalPodAdmissionHandler, klet.containerManager.UpdatePluginResources))
// apply functional Option's
for _, opt := range kubeDeps.Options {
opt(klet)
}
if sysruntime.GOOS == "linux" {
// AppArmor is a Linux kernel security module and it does not support other operating systems.
klet.appArmorValidator = apparmor.NewValidator()
klet.softAdmitHandlers.AddPodAdmitHandler(lifecycle.NewAppArmorAdmitHandler(klet.appArmorValidator))
}
leaseDuration := time.Duration(kubeCfg.NodeLeaseDurationSeconds) * time.Second
renewInterval := time.Duration(float64(leaseDuration) * nodeLeaseRenewIntervalFraction)
klet.nodeLeaseController = lease.NewController(
klet.clock,
klet.heartbeatClient,
string(klet.nodeName),
kubeCfg.NodeLeaseDurationSeconds,
klet.onRepeatedHeartbeatFailure,
renewInterval,
string(klet.nodeName),
v1.NamespaceNodeLease,
util.SetNodeOwnerFunc(klet.heartbeatClient, string(klet.nodeName)))
// setup node shutdown manager
shutdownManager, shutdownAdmitHandler := nodeshutdown.NewManager(&nodeshutdown.Config{
Logger: logger,
ProbeManager: klet.probeManager,
Recorder: kubeDeps.Recorder,
NodeRef: nodeRef,
GetPodsFunc: klet.GetActivePods,
KillPodFunc: killPodNow(klet.podWorkers, kubeDeps.Recorder),
SyncNodeStatusFunc: klet.syncNodeStatus,
ShutdownGracePeriodRequested: kubeCfg.ShutdownGracePeriod.Duration,
ShutdownGracePeriodCriticalPods: kubeCfg.ShutdownGracePeriodCriticalPods.Duration,
ShutdownGracePeriodByPodPriority: kubeCfg.ShutdownGracePeriodByPodPriority,
StateDirectory: rootDirectory,
})
klet.shutdownManager = shutdownManager
klet.usernsManager, err = MakeUserNsManager(klet)
if err != nil {
return nil, err
}
klet.admitHandlers.AddPodAdmitHandler(shutdownAdmitHandler)
// Finally, put the most recent version of the config on the Kubelet, so
// people can see how it was configured.
klet.kubeletConfiguration = *kubeCfg
// Generating the status funcs should be the last thing we do,
// since this relies on the rest of the Kubelet having been constructed.
klet.setNodeStatusFuncs = klet.defaultNodeStatusFuncs()
return klet, nil
}
type serviceLister interface {
List(labels.Selector) ([]*v1.Service, error)
}
// Kubelet is the main kubelet implementation.
type Kubelet struct {
kubeletConfiguration kubeletconfiginternal.KubeletConfiguration
// hostname is the hostname the kubelet detected or was given via flag/config
hostname string
// hostnameOverridden indicates the hostname was overridden via flag/config
hostnameOverridden bool
nodeName types.NodeName
runtimeCache kubecontainer.RuntimeCache
kubeClient clientset.Interface
heartbeatClient clientset.Interface
rootDirectory string
lastObservedNodeAddressesMux sync.RWMutex
lastObservedNodeAddresses []v1.NodeAddress
// onRepeatedHeartbeatFailure is called when a heartbeat operation fails more than once. optional.
onRepeatedHeartbeatFailure func()
// podWorkers handle syncing Pods in response to events.
podWorkers PodWorkers
// resyncInterval is the interval between periodic full reconciliations of
// pods on this node.
resyncInterval time.Duration
// sourcesReady records the sources seen by the kubelet, it is thread-safe.
sourcesReady config.SourcesReady
// podManager is a facade that abstracts away the various sources of pods
// this Kubelet services.
podManager kubepod.Manager
// Needed to observe and respond to situations that could impact node stability
evictionManager eviction.Manager
// Optional, defaults to /logs/ from /var/log
logServer http.Handler
// Optional, defaults to simple Docker implementation
runner kubecontainer.CommandRunner
// cAdvisor used for container information.
cadvisor cadvisor.Interface
// Set to true to have the node register itself with the apiserver.
registerNode bool
// List of taints to add to a node object when the kubelet registers itself.
registerWithTaints []v1.Taint
// Set to true to have the node register itself as schedulable.
registerSchedulable bool
// for internal book keeping; access only from within registerWithApiserver
registrationCompleted bool
// dnsConfigurer is used for setting up DNS resolver configuration when launching pods.
dnsConfigurer *dns.Configurer
// masterServiceNamespace is the namespace that the master service is exposed in.
masterServiceNamespace string
// serviceLister knows how to list services
serviceLister serviceLister
// serviceHasSynced indicates whether services have been sync'd at least once.
// Check this before trusting a response from the lister.
serviceHasSynced cache.InformerSynced
// nodeLister knows how to list nodes
nodeLister corelisters.NodeLister
// nodeHasSynced indicates whether nodes have been sync'd at least once.
// Check this before trusting a response from the node lister.
nodeHasSynced cache.InformerSynced
// a list of node labels to register
nodeLabels map[string]string
// Last timestamp when runtime responded on ping.
// Mutex is used to protect this value.
runtimeState *runtimeState
// Volume plugins.