-
Notifications
You must be signed in to change notification settings - Fork 124
/
machine.go
895 lines (782 loc) · 32.2 KB
/
machine.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
/*
Copyright 2017 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 controller
import (
"errors"
"fmt"
"net"
"regexp"
"strings"
"sync"
"time"
"github.com/golang/glog"
"github.com/heptiolabs/healthcheck"
machineclientset "github.com/kubermatic/machine-controller/pkg/client/clientset/versioned"
machinescheme "github.com/kubermatic/machine-controller/pkg/client/clientset/versioned/scheme"
machinelistersv1alpha1 "github.com/kubermatic/machine-controller/pkg/client/listers/machines/v1alpha1"
"github.com/kubermatic/machine-controller/pkg/cloudprovider"
"github.com/kubermatic/machine-controller/pkg/cloudprovider/cloud"
cloudprovidererrors "github.com/kubermatic/machine-controller/pkg/cloudprovider/errors"
"github.com/kubermatic/machine-controller/pkg/cloudprovider/instance"
"github.com/kubermatic/machine-controller/pkg/containerruntime"
"github.com/kubermatic/machine-controller/pkg/containerruntime/docker"
machinev1alpha1 "github.com/kubermatic/machine-controller/pkg/machines/v1alpha1"
"github.com/kubermatic/machine-controller/pkg/providerconfig"
"github.com/kubermatic/machine-controller/pkg/userdata"
"github.com/prometheus/client_golang/prometheus"
corev1 "k8s.io/api/core/v1"
"k8s.io/apimachinery/pkg/api/equality"
kerrors "k8s.io/apimachinery/pkg/api/errors"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/apimachinery/pkg/labels"
utilruntime "k8s.io/apimachinery/pkg/util/runtime"
"k8s.io/apimachinery/pkg/util/sets"
"k8s.io/apimachinery/pkg/util/wait"
"k8s.io/client-go/kubernetes"
"k8s.io/client-go/kubernetes/scheme"
typedcorev1 "k8s.io/client-go/kubernetes/typed/core/v1"
listerscorev1 "k8s.io/client-go/listers/core/v1"
"k8s.io/client-go/tools/cache"
clientcmdapi "k8s.io/client-go/tools/clientcmd/api"
"k8s.io/client-go/tools/record"
"k8s.io/client-go/tools/reference"
"k8s.io/client-go/util/retry"
"k8s.io/client-go/util/workqueue"
)
const (
finalizerDeleteInstance = "machine-delete-finalizer"
metricsUpdatePeriod = 10 * time.Second
deletionRetryWaitPeriod = 10 * time.Second
machineKind = "Machine"
controllerNameAnnotationKey = "machine.k8s.io/controller"
latestKubernetesVersion = "1.9.6"
)
// Controller is the controller implementation for machine resources
type Controller struct {
kubeClient kubernetes.Interface
machineClient machineclientset.Interface
nodesLister listerscorev1.NodeLister
machinesLister machinelistersv1alpha1.MachineLister
secretSystemNsLister listerscorev1.SecretLister
workqueue workqueue.RateLimitingInterface
recorder record.EventRecorder
clusterDNSIPs []net.IP
metrics *MetricsCollection
kubeconfigProvider KubeconfigProvider
validationCache map[string]bool
validationCacheMutex sync.Mutex
name string
}
type KubeconfigProvider interface {
GetKubeconfig() (*clientcmdapi.Config, error)
}
// MetricsCollection is a struct of all metrics used in
// this controller.
type MetricsCollection struct {
Workers prometheus.Gauge
Errors prometheus.Counter
}
// NewMachineController returns a new machine controller
func NewMachineController(
kubeClient kubernetes.Interface,
machineClient machineclientset.Interface,
nodeInformer cache.SharedIndexInformer,
nodeLister listerscorev1.NodeLister,
machineInformer cache.SharedIndexInformer,
machineLister machinelistersv1alpha1.MachineLister,
secretSystemNsLister listerscorev1.SecretLister,
clusterDNSIPs []net.IP,
metrics *MetricsCollection,
prometheusRegistry prometheus.Registerer,
kubeconfigProvider KubeconfigProvider,
name string) *Controller {
machinescheme.AddToScheme(scheme.Scheme)
eventBroadcaster := record.NewBroadcaster()
eventBroadcaster.StartLogging(glog.V(4).Infof)
eventBroadcaster.StartRecordingToSink(&typedcorev1.EventSinkImpl{Interface: kubeClient.CoreV1().Events("")})
if prometheusRegistry != nil {
prometheusRegistry.MustRegister(metrics.Errors)
prometheusRegistry.MustRegister(metrics.Workers)
}
controller := &Controller{
kubeClient: kubeClient,
nodesLister: nodeLister,
machineClient: machineClient,
machinesLister: machineLister,
secretSystemNsLister: secretSystemNsLister,
workqueue: workqueue.NewNamedRateLimitingQueue(workqueue.NewItemExponentialFailureRateLimiter(1*time.Second, 5*time.Minute), "Machines"),
recorder: eventBroadcaster.NewRecorder(scheme.Scheme, corev1.EventSource{Component: "machine-controller"}),
clusterDNSIPs: clusterDNSIPs,
metrics: metrics,
kubeconfigProvider: kubeconfigProvider,
validationCache: map[string]bool{},
name: name,
}
machineInformer.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: controller.enqueueMachine,
UpdateFunc: func(old, new interface{}) {
controller.enqueueMachine(new)
},
})
nodeInformer.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: controller.handleObject,
UpdateFunc: func(old, new interface{}) {
newNode := new.(*corev1.Node)
oldNode := old.(*corev1.Node)
if newNode.ResourceVersion == oldNode.ResourceVersion {
return
}
controller.handleObject(new)
},
DeleteFunc: controller.handleObject,
})
utilruntime.ErrorHandlers = append(utilruntime.ErrorHandlers, func(err error) {
controller.metrics.Errors.Add(1)
})
return controller
}
// Run starts the control loop
func (c *Controller) Run(threadiness int, stopCh <-chan struct{}) error {
defer utilruntime.HandleCrash()
defer c.workqueue.ShutDown()
for i := 0; i < threadiness; i++ {
go wait.Until(c.runWorker, time.Second, stopCh)
}
c.metrics.Workers.Set(float64(threadiness))
<-stopCh
return nil
}
func (c *Controller) runWorker() {
for c.processNextWorkItem() {
}
}
func (c *Controller) processNextWorkItem() bool {
key, quit := c.workqueue.Get()
if quit {
return false
}
defer c.workqueue.Done(key)
glog.V(6).Infof("Processing machine: %s", key)
err := c.syncHandler(key.(string))
if err == nil {
c.workqueue.Forget(key)
return true
}
utilruntime.HandleError(fmt.Errorf("%v failed with: %v", key, err))
c.workqueue.AddRateLimited(key)
return true
}
func (c *Controller) nodeIsReady(node *corev1.Node) bool {
for _, condition := range node.Status.Conditions {
if condition.Type == corev1.NodeReady {
if condition.Status == corev1.ConditionTrue {
return true
}
}
}
return false
}
func (c *Controller) getNodeByNodeRef(nodeRef *corev1.ObjectReference) (*corev1.Node, error) {
listerNode, err := c.nodesLister.Get(nodeRef.Name)
if err != nil {
return nil, err
}
return listerNode.DeepCopy(), nil
}
func (c *Controller) updateMachine(name string, modify func(*machinev1alpha1.Machine)) (*machinev1alpha1.Machine, error) {
var updatedMachine *machinev1alpha1.Machine
err := retry.RetryOnConflict(retry.DefaultBackoff, func() error {
var retryErr error
//Get latest version from API
currentMachine, err := c.machineClient.Machine().Machines().Get(name, metav1.GetOptions{})
if err != nil {
return err
}
// Apply modifications
modify(currentMachine)
// Update the machine
updatedMachine, retryErr = c.machineClient.MachineV1alpha1().Machines().Update(currentMachine)
return retryErr
})
return updatedMachine, err
}
func (c *Controller) clearMachineErrorIfSet(machine *machinev1alpha1.Machine) (*machinev1alpha1.Machine, error) {
if machine.Status.ErrorMessage != nil || machine.Status.ErrorReason != nil {
return c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Status.ErrorMessage = nil
m.Status.ErrorReason = nil
})
}
return machine, nil
}
// updateMachine updates machine's ErrorMessage and ErrorReason regardless if they were set or not
// this essentially overwrites previous values
func (c *Controller) updateMachineError(machine *machinev1alpha1.Machine, reason machinev1alpha1.MachineStatusError, message string) (*machinev1alpha1.Machine, error) {
return c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Status.ErrorMessage = &message
m.Status.ErrorReason = &reason
})
}
// updateMachineErrorIfTerminalError is a convenience method that will update machine's Status if the given err is terminal
// and at the same time terminal error will be returned to the caller
// otherwise it will return formatted error according to errMsg
func (c *Controller) updateMachineErrorIfTerminalError(machine *machinev1alpha1.Machine, stReason machinev1alpha1.MachineStatusError, stMessage string, err error, errMsg string) error {
c.recorder.Eventf(machine, corev1.EventTypeWarning, string(stReason), stMessage)
if ok, _, _ := cloudprovidererrors.IsTerminalError(err); ok {
if _, errNested := c.updateMachineError(machine, stReason, stMessage); errNested != nil {
return fmt.Errorf("failed to update machine error after due to %v, terminal error = %v", errNested, stMessage)
}
return err
}
return fmt.Errorf("%s, due to %v", errMsg, err)
}
func (c *Controller) getProviderInstance(prov cloud.Provider, machine *machinev1alpha1.Machine) (instance.Instance, error) {
return prov.Get(machine)
}
func (c *Controller) deleteProviderInstance(prov cloud.Provider, machine *machinev1alpha1.Machine) error {
return prov.Delete(machine, c.updateMachine)
}
func (c *Controller) createProviderInstance(prov cloud.Provider, machine *machinev1alpha1.Machine, userdata string) (instance.Instance, error) {
// Ensure finalizer is there
machine, err := c.ensureDeleteFinalizerExists(machine)
if err != nil {
return nil, err
}
return prov.Create(machine, c.updateMachine, userdata)
}
func (c *Controller) validateMachine(prov cloud.Provider, machine *machinev1alpha1.Machine) error {
err := prov.Validate(machine.Spec)
if err != nil {
c.recorder.Eventf(machine, corev1.EventTypeWarning, "ValidationFailed", "Validation failed: %v", err)
return err
}
c.recorder.Event(machine, corev1.EventTypeNormal, "ValidationSucceeded", "Validation succeeded")
return nil
}
func (c *Controller) syncHandler(key string) error {
listerMachine, err := c.machinesLister.Get(key)
if err != nil {
if kerrors.IsNotFound(err) {
glog.V(2).Infof("machine '%s' in work queue no longer exists", key)
return nil
}
return err
}
machine := listerMachine.DeepCopy()
// step 1: check if the machine can be processed by this controller.
// set the annotation "machine.k8s.io/controller": my-controller
// and the flag --name=my-controller to make only this controller process a node
machineControllerName := machine.Annotations[controllerNameAnnotationKey]
if machineControllerName != c.name {
glog.V(6).Infof("skipping machine '%s' as it is not meant for this controller", machine.Name)
if machineControllerName == "" && c.name != "" {
glog.V(6).Infof("this controller is configured to only process machines with the annotation %s:%s", controllerNameAnnotationKey, c.name)
return nil
}
if machineControllerName != "" && c.name == "" {
glog.V(6).Infof("this controller is configured to process all machines which have no controller specified via annotation %s. The machine has %s:%s", controllerNameAnnotationKey, controllerNameAnnotationKey, machineControllerName)
return nil
}
glog.V(6).Infof("this controller is configured to process machines which the annotation %s:%s. The machine has %s:%s", controllerNameAnnotationKey, c.name, controllerNameAnnotationKey, machineControllerName)
return nil
}
providerConfig, err := providerconfig.GetConfig(machine.Spec.ProviderConfig)
if err != nil {
return fmt.Errorf("failed to get provider config: %v", err)
}
skg := providerconfig.NewConfigVarResolver(c.kubeClient)
prov, err := cloudprovider.ForProvider(providerConfig.CloudProvider, skg)
if err != nil {
return fmt.Errorf("failed to get cloud provider %q: %v", providerConfig.CloudProvider, err)
}
// step 2: check if a user requested to delete the machine
if machine.DeletionTimestamp != nil {
if err := c.deleteMachineAndProviderInstance(prov, machine); err != nil {
return err
}
// As the deletion got triggered but the instance might not been gone yet, we need to recheck in a few seconds.
c.workqueue.AddAfter(machine.Name, deletionRetryWaitPeriod)
return nil
}
// step 3: essentially creates an instance for the given machine
userdataProvider, err := userdata.ForOS(providerConfig.OperatingSystem)
if err != nil {
return fmt.Errorf("failed to userdata provider for '%s': %v", providerConfig.OperatingSystem, err)
}
// We use a new variable here to be able to put the Event on the machine even thought
// c.defaultContainerRuntime returns a nil pointer for the machine in case of an error
machineWithDefaultedContainerRuntime, err := c.defaultContainerRuntime(machine, userdataProvider)
if err != nil {
errorMessage := fmt.Sprintf("failed to default the container runtime version: %v", err)
c.recorder.Event(machine, corev1.EventTypeWarning, "ContainerRuntimeDefaultingFailed", errorMessage)
return errors.New(errorMessage)
}
machine = machineWithDefaultedContainerRuntime
// case 3.2: creates an instance if there is no node associated with the given machine
if machine.Status.NodeRef == nil {
return c.ensureInstanceExistsForMachine(prov, machine, userdataProvider, providerConfig)
}
node, err := c.getNodeByNodeRef(machine.Status.NodeRef)
if err != nil {
//In case we cannot find a node for the NodeRef we must remove the NodeRef & recreate an instance on the next sync
if kerrors.IsNotFound(err) {
glog.V(4).Infof("found invalid NodeRef on machine %s. Deleting reference...", machine.Name)
_, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Status.NodeRef = nil
})
return err
}
return fmt.Errorf("failed to check if node for machine exists: '%s'", err)
}
if c.nodeIsReady(node) {
// If we have an ready node, we should clear the error in case one was set.
// Useful when there was a network outage & a cloud-provider api outage at the same time
if machine, err = c.clearMachineErrorIfSet(machine); err != nil {
return fmt.Errorf("failed to clear machine error: %v", err)
}
} else {
// Node is not ready anymore? Maybe it got deleted
return c.ensureInstanceExistsForMachine(prov, machine, userdataProvider, providerConfig)
}
// case 3.3: if the node exists make sure if it has labels and taints attached to it.
return c.ensureNodeLabelsAnnotationsAndTaints(node, machine)
}
func (c *Controller) cleanupMachineAfterDeletion(machine *machinev1alpha1.Machine) error {
var err error
glog.V(4).Infof("Removing finalizers from machine machine %s", machine.Name)
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
finalizers := sets.NewString(m.Finalizers...)
finalizers.Delete(finalizerDeleteInstance)
m.Finalizers = finalizers.List()
}); err != nil {
return fmt.Errorf("failed to update machine after removing the delete instance finalizer: %v", err)
}
glog.V(4).Infof("Removed delete finalizer from machine %s", machine.Name)
return nil
}
// deleteMachineAndProviderInstance makes sure that an instance has gone in a series of steps.
func (c *Controller) deleteMachineAndProviderInstance(prov cloud.Provider, machine *machinev1alpha1.Machine) error {
if err := c.deleteProviderInstance(prov, machine); err != nil {
message := fmt.Sprintf("%v. Please manually delete finalizers from the machine object.", err)
c.recorder.Eventf(machine, corev1.EventTypeWarning, "DeletionFailed", "Failed to delete machine: %v", err)
return c.updateMachineErrorIfTerminalError(machine, machinev1alpha1.DeleteMachineError, message, err, "failed to delete machine at cloudprovider")
}
return c.cleanupMachineAfterDeletion(machine)
}
func (c *Controller) ensureInstanceExistsForMachine(prov cloud.Provider, machine *machinev1alpha1.Machine, userdataProvider userdata.Provider, providerConfig *providerconfig.Config) error {
glog.V(6).Infof("Requesting instance for machine '%s' from cloudprovider because no associated node with status ready found...", machine.Name)
// case 1: validate the machine spec before getting the instance from cloud provider.
// even though this is a little bit premature and inefficient, it helps us detect invalid specification
defaultedMachineSpec, changed, err := prov.AddDefaults(machine.Spec)
if err != nil {
return c.updateMachineErrorIfTerminalError(machine, machinev1alpha1.InvalidConfigurationMachineError, err.Error(), err, "failed to add defaults to machine")
}
if changed {
glog.V(4).Infof("updating machine '%s' with defaults...", machine.Name)
c.recorder.Event(machine, corev1.EventTypeNormal, "Defaulted", "Updated machine with defaults")
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Spec = defaultedMachineSpec
}); err != nil {
return fmt.Errorf("failed to update machine '%s' after adding defaults: '%v'", machine.Name, err)
}
glog.V(4).Infof("Successfully updated machine '%s' with defaults!", machine.Name)
}
cacheKey := string(machine.UID) + machine.ResourceVersion
c.validationCacheMutex.Lock()
validationSuccess := c.validationCache[cacheKey]
c.validationCacheMutex.Unlock()
if !validationSuccess {
if err := c.validateMachine(prov, machine); err != nil {
if _, errNested := c.updateMachineError(machine, machinev1alpha1.InvalidConfigurationMachineError, err.Error()); errNested != nil {
return fmt.Errorf("failed to update machine error after failed validation: %v", errNested)
}
return fmt.Errorf("invalid provider config: %v", err)
}
c.validationCacheMutex.Lock()
c.validationCache[cacheKey] = true
c.validationCacheMutex.Unlock()
} else {
glog.V(6).Infof("Skipping validation as the machine was already successfully validated before")
}
providerInstance, err := prov.Get(machine)
// case 2: retrieving instance from provider was not successful
if err != nil {
//First invalidate the validation cache to make sure we run the validation on the next sync.
//This might happen in case the user invalidates his provider credentials...
c.validationCacheMutex.Lock()
c.validationCache[cacheKey] = false
c.validationCacheMutex.Unlock()
// case 2.1: instance was not found and we are going to create one
if err == cloudprovidererrors.ErrInstanceNotFound {
// remove an error message in case it was set
if machine, err = c.clearMachineErrorIfSet(machine); err != nil {
return fmt.Errorf("failed to update machine after removing the failed validation error: %v", err)
}
glog.V(4).Infof("Validated machine spec of %s", machine.Name)
kubeconfig, err := c.createBootstrapKubeconfig(machine.Name)
if err != nil {
c.recorder.Eventf(machine, corev1.EventTypeWarning, "CreateBootstrapKubeconfigFailed", "Creating bootstrap kubeconfig failed: %v", err)
return fmt.Errorf("failed to create bootstrap kubeconfig: %v", err)
}
userdata, err := userdataProvider.UserData(machine.Spec, kubeconfig, prov, c.clusterDNSIPs)
if err != nil {
c.recorder.Eventf(machine, corev1.EventTypeWarning, "UserdataRenderingFailed", "Userdata rendering failed: %v", err)
return fmt.Errorf("failed get userdata: %v", err)
}
// Create the instance
if providerInstance, err = c.createProviderInstance(prov, machine, userdata); err != nil {
c.recorder.Eventf(machine, corev1.EventTypeWarning, "CreateInstanceFailed", "Instance creation failed: %v", err)
message := fmt.Sprintf("%v. Unable to create a machine.", err)
return c.updateMachineErrorIfTerminalError(machine, machinev1alpha1.CreateMachineError, message, err, "failed to create machine at cloudprover")
}
c.recorder.Event(machine, corev1.EventTypeNormal, "Created", "Successfully created instance")
// remove error message in case it was set
if machine, err = c.clearMachineErrorIfSet(machine); err != nil {
return fmt.Errorf("failed to update machine after removing the create machine error: %v", err)
}
glog.V(4).Infof("Created machine %s at cloud provider", machine.Name)
return nil
}
// case 2.2: terminal error was returned and manual interaction is required to recover
if ok, _, message := cloudprovidererrors.IsTerminalError(err); ok {
message = fmt.Sprintf("%v. Unable to create a machine.", err)
return c.updateMachineErrorIfTerminalError(machine, machinev1alpha1.CreateMachineError, message, err, "failed to get instance from provider")
}
// case 2.3: transient error was returned, requeue the request and try again in the future
return fmt.Errorf("failed to get instance from provider: %v", err)
}
// Instance exists, so ensure finalizer does as well
machine, err = c.ensureDeleteFinalizerExists(machine)
if err != nil {
return err
}
// case 3: retrieving the instance from cloudprovider was successfull
eventMessage := fmt.Sprintf("Found instance at cloud provider, addresses: %v", providerInstance.Addresses())
c.recorder.Event(machine, corev1.EventTypeNormal, "InstanceFound", eventMessage)
return c.ensureNodeOwnerRefAndConfigSource(providerInstance, machine, providerConfig)
}
func (c *Controller) ensureNodeOwnerRefAndConfigSource(providerInstance instance.Instance, machine *machinev1alpha1.Machine, providerConfig *providerconfig.Config) error {
node, exists, err := c.getNode(providerInstance, providerConfig.CloudProvider)
if err != nil {
return fmt.Errorf("failed to get node for machine %s: %v", machine.Name, err)
}
if exists {
ownerRef := metav1.GetControllerOf(node)
if ownerRef == nil {
gv := machinev1alpha1.SchemeGroupVersion
node.OwnerReferences = []metav1.OwnerReference{*metav1.NewControllerRef(machine, gv.WithKind(machineKind))}
node, err = c.kubeClient.CoreV1().Nodes().Update(node)
if err != nil {
return fmt.Errorf("failed to update node %s after adding the owner ref: %v", node.Name, err)
}
glog.V(4).Infof("Added owner ref to node %s (machine=%s)", node.Name, machine.Name)
c.recorder.Eventf(machine, corev1.EventTypeNormal, "NodeMatched", "Successfully matched machine to node %s", node.Name)
}
if node.Spec.ConfigSource == nil && machine.Spec.ConfigSource != nil {
node.Spec.ConfigSource = machine.Spec.ConfigSource
node, err = c.kubeClient.CoreV1().Nodes().Update(node)
if err != nil {
return fmt.Errorf("failed to update node %s after setting the config source: %v", node.Name, err)
}
glog.V(4).Infof("Added config source to node %s (machine %s)", node.Name, machine.Name)
}
err = c.updateMachineStatus(machine, node)
if err != nil {
return fmt.Errorf("failed to update machine status: %v", err)
}
}
return nil
}
func (c *Controller) ensureNodeLabelsAnnotationsAndTaints(node *corev1.Node, machine *machinev1alpha1.Machine) error {
var labelsUpdated bool
for k, v := range machine.Spec.Labels {
if _, exists := node.Labels[k]; !exists {
labelsUpdated = true
node.Labels[k] = v
}
}
var annotationsUpdated bool
for k, v := range machine.Spec.Annotations {
if _, exists := node.Annotations[k]; !exists {
annotationsUpdated = true
node.Annotations[k] = v
}
}
taintExists := func(node *corev1.Node, taint corev1.Taint) bool {
for _, t := range node.Spec.Taints {
if t.MatchTaint(&taint) {
return true
}
}
return false
}
var taintsUpdated bool
for _, t := range machine.Spec.Taints {
if !taintExists(node, t) {
node.Spec.Taints = append(node.Spec.Taints, t)
taintsUpdated = true
}
}
if labelsUpdated || annotationsUpdated || taintsUpdated {
node, err := c.kubeClient.CoreV1().Nodes().Update(node)
if err != nil {
return fmt.Errorf("failed to update node %s after setting labels/annotations/taints: %v", node.Name, err)
}
c.recorder.Event(machine, corev1.EventTypeNormal, "LabelsAnnotationsTaintsUpdated", "Successfully updated labels/annotations/taints")
glog.V(4).Infof("Added labels/annotations/taints to node %s (machine %s)", node.Name, machine.Name)
}
return nil
}
func (c *Controller) updateMachineStatus(machine *machinev1alpha1.Machine, node *corev1.Node) error {
if node == nil {
return nil
}
var (
runtimeName, runtimeVersion string
err error
)
ref, err := reference.GetReference(scheme.Scheme, node)
if err != nil {
return fmt.Errorf("failed to get node reference for %s : %v", node.Name, err)
}
if !equality.Semantic.DeepEqual(machine.Status.NodeRef, ref) {
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Status.NodeRef = ref
}); err != nil {
return fmt.Errorf("failed to update machine: %v", err)
}
}
if machine.Status.Versions == nil {
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Status.Versions = &machinev1alpha1.MachineVersionInfo{}
}); err != nil {
return fmt.Errorf("failed to update machine: %v", err)
}
}
if node.Status.NodeInfo.ContainerRuntimeVersion != "" {
runtimeName, runtimeVersion, err = parseContainerRuntime(node.Status.NodeInfo.ContainerRuntimeVersion)
if err != nil {
glog.V(2).Infof("failed to parse container runtime from node %s: %v", node.Name, err)
runtimeName = "unknown"
runtimeVersion = "unknown"
}
if machine.Status.Versions.ContainerRuntime.Name != runtimeName || machine.Status.Versions.ContainerRuntime.Version != runtimeVersion {
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Status.Versions.ContainerRuntime.Name = runtimeName
m.Status.Versions.ContainerRuntime.Version = runtimeVersion
}); err != nil {
return fmt.Errorf("failed to update machine: %v", err)
}
}
}
if machine.Status.Versions.Kubelet != node.Status.NodeInfo.KubeletVersion {
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Status.Versions.Kubelet = node.Status.NodeInfo.KubeletVersion
}); err != nil {
return fmt.Errorf("failed to update machine: %v", err)
}
}
return nil
}
var (
containerRuntime = regexp.MustCompile(`(docker)://(.*)`)
)
func parseContainerRuntime(s string) (runtime, version string, err error) {
res := containerRuntime.FindStringSubmatch(s)
if len(res) == 3 {
return res[1], res[2], nil
}
return "", "", fmt.Errorf("invalid format. Expected 'runtime://version'")
}
func (c *Controller) getNode(instance instance.Instance, provider providerconfig.CloudProvider) (node *corev1.Node, exists bool, err error) {
if instance == nil {
return nil, false, fmt.Errorf("getNode called with nil provider instance!")
}
nodes, err := c.nodesLister.List(labels.Everything())
if err != nil {
return nil, false, err
}
// We trim leading slashes in raw ID, since we always want three slashes in full ID
providerID := fmt.Sprintf("%s:///%s", provider, strings.TrimLeft(instance.ID(), "/"))
for _, node := range nodes {
if provider == providerconfig.CloudProviderAzure {
// Azure IDs are case-insensitive
if strings.EqualFold(node.Spec.ProviderID, providerID) {
return node.DeepCopy(), true, nil
}
} else {
if node.Spec.ProviderID == providerID {
return node.DeepCopy(), true, nil
}
}
for _, nodeAddress := range node.Status.Addresses {
for _, instanceAddress := range instance.Addresses() {
if nodeAddress.Address == instanceAddress {
return node.DeepCopy(), true, nil
}
}
}
}
return nil, false, nil
}
func (c *Controller) defaultContainerRuntime(machine *machinev1alpha1.Machine, prov userdata.Provider) (*machinev1alpha1.Machine, error) {
var err error
if machine.Spec.Versions.Kubelet == "" {
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Spec.Versions.Kubelet = latestKubernetesVersion
}); err != nil {
return nil, err
}
}
if machine.Spec.Versions.ContainerRuntime.Name == "" {
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Spec.Versions.ContainerRuntime.Name = containerruntime.Docker
}); err != nil {
return nil, err
}
}
if machine.Spec.Versions.ContainerRuntime.Version == "" {
var (
defaultVersions []string
err error
)
switch machine.Spec.Versions.ContainerRuntime.Name {
case containerruntime.Docker:
defaultVersions, err = docker.GetOfficiallySupportedVersions(machine.Spec.Versions.Kubelet)
if err != nil {
return nil, fmt.Errorf("failed to get a officially supported docker version for the given kubelet version: %v", err)
}
default:
return nil, fmt.Errorf("invalid container runtime. Supported: '%s'", containerruntime.Docker)
}
var newVersion string
providerSupportedVersions := prov.SupportedContainerRuntimes()
for _, v := range defaultVersions {
for _, sv := range providerSupportedVersions {
if sv.Version == v {
// we should not return asap as we prefer the highest supported version
newVersion = sv.Version
}
}
}
if newVersion == "" {
return nil, fmt.Errorf("no supported versions available for '%s'", machine.Spec.Versions.ContainerRuntime.Name)
}
machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
m.Spec.Versions.ContainerRuntime.Version = newVersion
})
if err != nil {
return nil, err
}
return machine, err
}
return machine, nil
}
func (c *Controller) enqueueMachine(obj interface{}) {
var key string
var err error
if key, err = cache.MetaNamespaceKeyFunc(obj); err != nil {
utilruntime.HandleError(err)
return
}
c.workqueue.AddRateLimited(key)
}
func (c *Controller) handleObject(obj interface{}) {
var object metav1.Object
var ok bool
if object, ok = obj.(metav1.Object); !ok {
tombstone, ok := obj.(cache.DeletedFinalStateUnknown)
if !ok {
utilruntime.HandleError(fmt.Errorf("error decoding object, invalid type"))
return
}
object, ok = tombstone.Obj.(metav1.Object)
if !ok {
utilruntime.HandleError(fmt.Errorf("error decoding object tombstone, invalid type"))
return
}
glog.V(4).Infof("Recovered deleted object '%s' from tombstone", object.GetName())
}
ownerRef := metav1.GetControllerOf(object)
if ownerRef != nil {
if ownerRef.Kind != "Machine" {
return
}
machine, err := c.machinesLister.Get(ownerRef.Name)
if err != nil {
glog.V(4).Infof("ignoring orphaned object '%s' of machine '%s'", object.GetSelfLink(), ownerRef.Name)
return
}
glog.V(6).Infof("Processing node: %s (machine=%s)", object.GetName(), machine.Name)
c.enqueueMachine(machine)
return
}
if ownerRef == nil {
machines, err := c.machinesLister.List(labels.Everything())
if err != nil {
utilruntime.HandleError(fmt.Errorf("error listing machines: '%v'", err))
return
}
for _, machine := range machines {
// We get triggered by node{Add,Update}, so enqeue machines if they
// have no nodeRef yet to make matching happen ASAP
if machine.Status.NodeRef == nil {
c.enqueueMachine(machine)
}
}
}
}
func (c *Controller) ReadinessChecks() map[string]healthcheck.Check {
return map[string]healthcheck.Check{
"valid-info-kubeconfig": func() error {
cm, err := c.kubeconfigProvider.GetKubeconfig()
if err != nil {
err := fmt.Errorf("failed to get cluster-info configmap: %v", err)
glog.V(2).Info(err)
return err
}
if len(cm.Clusters) != 1 {
err := errors.New("invalid kubeconfig: no clusters found")
glog.V(2).Info(err)
return err
}
for name, c := range cm.Clusters {
if len(c.CertificateAuthorityData) == 0 {
err := fmt.Errorf("invalid kubeconfig: no certificate authority data was specified for kuberconfig.clusters.['%s']", name)
glog.V(2).Info(err)
return err
}
if len(c.Server) == 0 {
err := fmt.Errorf("invalid kubeconfig: no server was specified for kuberconfig.clusters.['%s']", name)
glog.V(2).Info(err)
return err
}
}
return nil
},
}
}
func (c *Controller) ensureDeleteFinalizerExists(machine *machinev1alpha1.Machine) (*machinev1alpha1.Machine, error) {
if !sets.NewString(machine.Finalizers...).Has(finalizerDeleteInstance) {
var err error
if machine, err = c.updateMachine(machine.Name, func(m *machinev1alpha1.Machine) {
finalizers := sets.NewString(m.Finalizers...)
finalizers.Insert(finalizerDeleteInstance)
m.Finalizers = finalizers.List()
}); err != nil {
return nil, fmt.Errorf("failed to update machine after adding the delete instance finalizer: %v", err)
}
glog.V(4).Infof("Added delete finalizer to machine %s", machine.Name)
}
return machine, nil
}