/
kubernetes.go
1340 lines (1209 loc) · 42.9 KB
/
kubernetes.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 2022 The KodeRover 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 jobcontroller
import (
"bytes"
"context"
"encoding/json"
"fmt"
"os"
"path"
"path/filepath"
"regexp"
"sort"
"strconv"
"strings"
"time"
"github.com/mozillazg/go-pinyin"
"github.com/pkg/errors"
"go.uber.org/zap"
"gopkg.in/yaml.v2"
batchv1 "k8s.io/api/batch/v1"
corev1 "k8s.io/api/core/v1"
"k8s.io/apimachinery/pkg/api/resource"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/apimachinery/pkg/fields"
"k8s.io/apimachinery/pkg/labels"
"k8s.io/client-go/informers"
"k8s.io/client-go/kubernetes"
"k8s.io/client-go/rest"
"sigs.k8s.io/controller-runtime/pkg/client"
crClient "sigs.k8s.io/controller-runtime/pkg/client"
"github.com/koderover/zadig/v2/pkg/microservice/aslan/config"
commonmodels "github.com/koderover/zadig/v2/pkg/microservice/aslan/core/common/repository/models"
commonrepo "github.com/koderover/zadig/v2/pkg/microservice/aslan/core/common/repository/mongodb"
"github.com/koderover/zadig/v2/pkg/microservice/aslan/core/common/service/kube"
"github.com/koderover/zadig/v2/pkg/microservice/aslan/core/multicluster/service"
"github.com/koderover/zadig/v2/pkg/microservice/warpdrive/core/service/types/task"
"github.com/koderover/zadig/v2/pkg/setting"
kubeclient "github.com/koderover/zadig/v2/pkg/shared/kube/client"
"github.com/koderover/zadig/v2/pkg/shared/kube/wrapper"
"github.com/koderover/zadig/v2/pkg/tool/kube/containerlog"
"github.com/koderover/zadig/v2/pkg/tool/kube/getter"
"github.com/koderover/zadig/v2/pkg/tool/kube/podexec"
"github.com/koderover/zadig/v2/pkg/tool/kube/updater"
"github.com/koderover/zadig/v2/pkg/tool/log"
s3tool "github.com/koderover/zadig/v2/pkg/tool/s3"
commontypes "github.com/koderover/zadig/v2/pkg/types"
"github.com/koderover/zadig/v2/pkg/types/job"
"github.com/koderover/zadig/v2/pkg/util"
)
const (
BusyBoxImage = "koderover.tencentcloudcr.com/koderover-public/busybox:latest"
ZadigContextDir = "/zadig/"
ZadigLogFile = ZadigContextDir + "zadig.log"
ZadigLifeCycleFile = ZadigContextDir + "lifecycle"
ExecutorResourceVolumeName = "executor-resource"
ExecutorVolumePath = "/executor"
JobExecutorFile = ExecutorVolumePath + "/jobexecutor"
defaultSecretEmail = "bot@koderover.com"
registrySecretSuffix = "-registry-secret"
workflowConfigMapRoleSA = "workflow-cm-sa"
defaultRetryCount = 3
defaultRetryInterval = time.Second * 3
// build job outputs key
IMAGEKEY = "IMAGE"
IMAGETAGKEY = "imageTag"
)
func GetK8sClients(hubServerAddr, clusterID string) (crClient.Client, kubernetes.Interface, *rest.Config, crClient.Reader, error) {
controllerRuntimeClient, err := kubeclient.GetKubeClient(hubServerAddr, clusterID)
if err != nil {
return nil, nil, nil, nil, fmt.Errorf("failed to get controller runtime client: %s", err)
}
clientset, err := kubeclient.GetKubeClientSet(hubServerAddr, clusterID)
if err != nil {
return nil, nil, nil, nil, fmt.Errorf("failed to get clientset: %s", err)
}
restConfig, err := kubeclient.GetRESTConfig(hubServerAddr, clusterID)
if err != nil {
return nil, nil, nil, nil, fmt.Errorf("failed to get rest config: %s", err)
}
kubeClientReader, err := kubeclient.GetKubeAPIReader(hubServerAddr, clusterID)
if err != nil {
return nil, nil, nil, nil, fmt.Errorf("failed to get api reader: %s", err)
}
return controllerRuntimeClient, clientset, restConfig, kubeClientReader, nil
}
type JobLabel struct {
JobName string
JobType string
}
func ensureDeleteConfigMap(namespace string, jobLabel *JobLabel, kubeClient crClient.Client) error {
ls := getJobLabels(jobLabel)
return updater.DeleteConfigMapsAndWait(namespace, labels.Set(ls).AsSelector(), kubeClient)
}
func ensureDeleteJob(namespace string, jobLabel *JobLabel, kubeClient crClient.Client) error {
ls := getJobLabels(jobLabel)
return updater.DeleteJobsAndWait(namespace, labels.Set(ls).AsSelector(), kubeClient)
}
// getJobLabels get labels k-v map from JobLabel struct
func getJobLabels(jobLabel *JobLabel) map[string]string {
retMap := map[string]string{
setting.JobLabelNameKey: strings.Replace(jobLabel.JobName, "_", "-", -1),
setting.JobLabelSTypeKey: strings.Replace(jobLabel.JobType, "_", "-", -1),
}
// no need to add labels with empty value to a job
for k, v := range retMap {
if len(v) == 0 {
delete(retMap, k)
}
}
return retMap
}
func GetJobContainerName(name string) string {
pyArgs := pinyin.NewArgs()
pyArgs.Fallback = func(r rune, a pinyin.Args) []string {
return []string{string(r)}
}
res := pinyin.Pinyin(name, pyArgs)
pinyins := make([]string, 0)
for _, py := range res {
pinyins = append(pinyins, strings.Join(py, ""))
}
resp := strings.Join(pinyins, "")
if len(resp) > 63 {
return resp[:63]
}
return resp
}
func createJobConfigMap(namespace, jobName string, jobLabel *JobLabel, jobCtx string, kubeClient crClient.Client) error {
cm := &corev1.ConfigMap{
ObjectMeta: metav1.ObjectMeta{
Name: jobName,
Namespace: namespace,
Labels: getJobLabels(jobLabel),
},
Data: map[string]string{
"job-config.xml": jobCtx,
},
}
return updater.CreateConfigMap(cm, kubeClient)
}
func getBaseImage(buildOS, imageFrom string) string {
// for built-in image, reaperImage and buildOs can generate a complete image
// reaperImage: koderover.tencentcloudcr.com/koderover-public/build-base:${BuildOS}-amd64
// buildOS: focal xenial bionic
jobImage := strings.ReplaceAll(config.ReaperImage(), "${BuildOS}", buildOS)
// for custom image, buildOS represents the exact custom image
if imageFrom == setting.ImageFromCustom {
jobImage = buildOS
}
return jobImage
}
func buildTolerations(clusterConfig *commonmodels.AdvancedConfig, strategyID string) []corev1.Toleration {
ret := make([]corev1.Toleration, 0)
if clusterConfig == nil || len(clusterConfig.ScheduleStrategy) == 0 {
return ret
}
var tolerations string
for _, strategy := range clusterConfig.ScheduleStrategy {
if strategyID != "" && strategy.StrategyID == strategyID {
tolerations = strategy.Tolerations
break
} else if strategyID == "" && strategy.Default {
tolerations = strategy.Tolerations
break
}
}
err := yaml.Unmarshal([]byte(tolerations), &ret)
if err != nil {
log.Errorf("failed to parse toleration config, err: %s", err)
return nil
}
return ret
}
func addNodeAffinity(clusterConfig *commonmodels.AdvancedConfig, strategyID string) *corev1.Affinity {
if clusterConfig == nil || len(clusterConfig.ScheduleStrategy) == 0 {
return nil
}
var strategy *commonmodels.ScheduleStrategy
for _, s := range clusterConfig.ScheduleStrategy {
if strategyID != "" && s.StrategyID == strategyID {
strategy = s
break
} else if strategyID == "" && s.Default {
strategy = s
break
}
}
if strategy == nil {
return nil
}
switch strategy.Strategy {
case setting.RequiredSchedule:
nodeSelectorTerms := make([]corev1.NodeSelectorTerm, 0)
for _, nodeLabel := range strategy.NodeLabels {
var matchExpressions []corev1.NodeSelectorRequirement
matchExpressions = append(matchExpressions, corev1.NodeSelectorRequirement{
Key: nodeLabel.Key,
Operator: nodeLabel.Operator,
Values: nodeLabel.Value,
})
nodeSelectorTerms = append(nodeSelectorTerms, corev1.NodeSelectorTerm{
MatchExpressions: matchExpressions,
})
}
affinity := &corev1.Affinity{
NodeAffinity: &corev1.NodeAffinity{
RequiredDuringSchedulingIgnoredDuringExecution: &corev1.NodeSelector{
NodeSelectorTerms: nodeSelectorTerms,
},
},
}
return affinity
case setting.PreferredSchedule:
preferredScheduleTerms := make([]corev1.PreferredSchedulingTerm, 0)
for _, nodeLabel := range strategy.NodeLabels {
var matchExpressions []corev1.NodeSelectorRequirement
matchExpressions = append(matchExpressions, corev1.NodeSelectorRequirement{
Key: nodeLabel.Key,
Operator: nodeLabel.Operator,
Values: nodeLabel.Value,
})
nodeSelectorTerm := corev1.NodeSelectorTerm{
MatchExpressions: matchExpressions,
}
preferredScheduleTerms = append(preferredScheduleTerms, corev1.PreferredSchedulingTerm{
Weight: 10,
Preference: nodeSelectorTerm,
})
}
affinity := &corev1.Affinity{
NodeAffinity: &corev1.NodeAffinity{
PreferredDuringSchedulingIgnoredDuringExecution: preferredScheduleTerms,
},
}
return affinity
default:
return nil
}
}
func buildPlainJob(jobName string, resReq setting.Request, resReqSpec setting.RequestSpec, jobTask *commonmodels.JobTask, jobTaskSpec *commonmodels.JobTaskPluginSpec, workflowCtx *commonmodels.WorkflowTaskCtx) (*batchv1.Job, error) {
collectJobOutput := `OLD_IFS=$IFS
export IFS=","
files='%s'
outputs='%s'
file_arr=($files)
output_arr=($outputs)
IFS="$OLD_IFS"
result="{"
for i in ${!file_arr[@]};
do
file_value=$(cat ${file_arr[$i]})
output_value=${output_arr[$i]}
result="$result\"$output_value\":\"$file_value\","
done
result=$(sed 's/,$/}/' <<< $result)
echo $result > %s
`
files := []string{}
outputs := []string{}
for _, output := range jobTask.Outputs {
outputFile := path.Join(job.JobOutputDir, output.Name)
files = append(files, outputFile)
outputs = append(outputs, output.Name)
}
collectJobOutputCommand := fmt.Sprintf(collectJobOutput, strings.Join(files, ","), strings.Join(outputs, ","), job.JobTerminationFile)
labels := getJobLabels(&JobLabel{
JobType: string(jobTask.JobType),
JobName: jobTask.K8sJobName,
})
ImagePullSecrets, err := getImagePullSecrets(jobTaskSpec.Properties.Registries)
if err != nil {
return nil, err
}
envs := []corev1.EnvVar{}
for _, env := range jobTaskSpec.Plugin.Envs {
envs = append(envs, corev1.EnvVar{Name: env.Name, Value: env.Value})
}
clusterID := jobTaskSpec.Properties.ClusterID
if clusterID == "" {
clusterID = setting.LocalClusterID
}
// fetch cluster to get nodeAffinity and tolerations
targetCluster, err := service.GetCluster(clusterID, log.SugaredLogger())
if err != nil {
return nil, fmt.Errorf("failed to find target cluster %s, err: %s", clusterID, err)
}
job := &batchv1.Job{
ObjectMeta: metav1.ObjectMeta{
Name: jobName,
Labels: labels,
},
Spec: batchv1.JobSpec{
Completions: int32Ptr(1),
Parallelism: int32Ptr(1),
BackoffLimit: int32Ptr(0),
// in case finished zombie job not cleaned up by zadig
TTLSecondsAfterFinished: int32Ptr(3600),
// in case zombie job never stop
ActiveDeadlineSeconds: int64Ptr(jobTaskSpec.Properties.Timeout*60 + 3600),
Template: corev1.PodTemplateSpec{
ObjectMeta: metav1.ObjectMeta{
Labels: labels,
},
Spec: corev1.PodSpec{
RestartPolicy: corev1.RestartPolicyNever,
ImagePullSecrets: ImagePullSecrets,
Containers: []corev1.Container{
{
ImagePullPolicy: corev1.PullAlways,
Name: GetJobContainerName(jobTask.Name),
Image: jobTaskSpec.Plugin.Image,
Args: jobTaskSpec.Plugin.Args,
Command: jobTaskSpec.Plugin.Cmds,
Lifecycle: &corev1.Lifecycle{
PreStop: &corev1.LifecycleHandler{
Exec: &corev1.ExecAction{
Command: []string{"/bin/sh", "-c", collectJobOutputCommand},
},
},
},
Env: envs,
VolumeMounts: []corev1.VolumeMount{
{
Name: "zadig-context",
MountPath: ZadigContextDir,
},
{
Name: "zadig-output",
MountPath: job.JobOutputDir,
},
},
Resources: getResourceRequirements(resReq, resReqSpec),
TerminationMessagePolicy: corev1.TerminationMessageReadFile,
TerminationMessagePath: job.JobTerminationFile,
},
},
Volumes: []corev1.Volume{
{
Name: "zadig-context",
VolumeSource: corev1.VolumeSource{
EmptyDir: &corev1.EmptyDirVolumeSource{},
},
},
{
Name: "zadig-output",
VolumeSource: corev1.VolumeSource{
EmptyDir: &corev1.EmptyDirVolumeSource{},
},
},
},
Tolerations: buildTolerations(targetCluster.AdvancedConfig, jobTaskSpec.Properties.StrategyID),
Affinity: addNodeAffinity(targetCluster.AdvancedConfig, jobTaskSpec.Properties.StrategyID),
},
},
},
}
setJobShareStorages(job, workflowCtx, jobTaskSpec.Properties.ShareStorageDetails, targetCluster)
ensureVolumeMounts(job)
return job, nil
}
func buildJob(jobType, jobImage, jobName, clusterID, currentNamespace string, resReq setting.Request, resReqSpec setting.RequestSpec, jobTask *commonmodels.JobTask, jobTaskSpec *commonmodels.JobTaskFreestyleSpec, workflowCtx *commonmodels.WorkflowTaskCtx, registries []*task.RegistryNamespace) (*batchv1.Job, error) {
var (
jobExecutorBootingScript string
jobExecutorBinaryFile = JobExecutorFile
)
if clusterID == "" {
clusterID = setting.LocalClusterID
}
// fetch cluster to get nodeAffinity and tolerations
targetCluster, err := service.GetCluster(clusterID, log.SugaredLogger())
if err != nil {
return nil, fmt.Errorf("failed to find target cluster %s, err: %s", clusterID, err)
}
jobExecutorBootingScript = fmt.Sprintf("mkdir %sdebug;", ZadigContextDir)
if jobTask.BreakpointBefore {
jobExecutorBootingScript += fmt.Sprintf("touch %sdebug/breakpoint_before;", ZadigContextDir)
}
if jobTask.BreakpointAfter {
jobExecutorBootingScript += fmt.Sprintf("touch %sdebug/breakpoint_after;", ZadigContextDir)
}
jobExecutorBootingScript += jobExecutorBinaryFile
labels := getJobLabels(&JobLabel{
JobType: jobType,
JobName: jobTask.K8sJobName,
})
ImagePullSecrets, err := getImagePullSecrets(jobTaskSpec.Properties.Registries)
if err != nil {
return nil, err
}
job := &batchv1.Job{
ObjectMeta: metav1.ObjectMeta{
Name: jobName,
Labels: labels,
},
Spec: batchv1.JobSpec{
Completions: int32Ptr(1),
Parallelism: int32Ptr(1),
BackoffLimit: int32Ptr(0),
// in case finished zombie job not cleaned up by zadig
TTLSecondsAfterFinished: int32Ptr(3600),
// in case zombie job never stop
ActiveDeadlineSeconds: int64Ptr(jobTaskSpec.Properties.Timeout*60 + 3600),
Template: corev1.PodTemplateSpec{
ObjectMeta: metav1.ObjectMeta{
Labels: labels,
},
Spec: corev1.PodSpec{
RestartPolicy: corev1.RestartPolicyNever,
ImagePullSecrets: ImagePullSecrets,
ServiceAccountName: workflowConfigMapRoleSA,
InitContainers: []corev1.Container{
{
ImagePullPolicy: corev1.PullIfNotPresent,
Name: "executor-resource-init",
Image: config.ExecutorImage(),
VolumeMounts: []corev1.VolumeMount{
{
Name: ExecutorResourceVolumeName,
MountPath: ExecutorVolumePath,
},
},
Command: []string{"/bin/sh", "-c", fmt.Sprintf("cp /app/* %s", ExecutorVolumePath)},
},
},
Containers: []corev1.Container{
{
ImagePullPolicy: corev1.PullAlways,
Name: GetJobContainerName(strings.ReplaceAll(jobTask.Name, "_", "-")),
Image: jobImage,
Command: []string{"/bin/sh", "-c"},
Args: []string{jobExecutorBootingScript},
Env: getEnvs(workflowCtx.ConfigMapMountDir, jobTaskSpec),
VolumeMounts: getVolumeMounts(workflowCtx.ConfigMapMountDir, jobTaskSpec.Properties.UseHostDockerDaemon),
Resources: getResourceRequirements(resReq, resReqSpec),
TerminationMessagePolicy: corev1.TerminationMessageReadFile,
TerminationMessagePath: job.JobTerminationFile,
},
},
Volumes: getVolumes(jobName, jobTaskSpec.Properties.UseHostDockerDaemon),
Tolerations: buildTolerations(targetCluster.AdvancedConfig, jobTaskSpec.Properties.StrategyID),
Affinity: addNodeAffinity(targetCluster.AdvancedConfig, jobTaskSpec.Properties.StrategyID),
},
},
},
}
setJobShareStorages(job, workflowCtx, jobTaskSpec.Properties.ShareStorageDetails, targetCluster)
if jobTaskSpec.Properties.CacheEnable && jobTaskSpec.Properties.Cache.MediumType == commontypes.NFSMedium {
volumeName := "build-cache"
job.Spec.Template.Spec.Volumes = append(job.Spec.Template.Spec.Volumes, corev1.Volume{
Name: volumeName,
VolumeSource: corev1.VolumeSource{
PersistentVolumeClaim: &corev1.PersistentVolumeClaimVolumeSource{
ClaimName: jobTaskSpec.Properties.Cache.NFSProperties.PVC,
},
},
})
mountPath := strings.ReplaceAll(jobTaskSpec.Properties.CacheUserDir, "$WORKSPACE", workflowCtx.Workspace)
if jobTaskSpec.Properties.CacheDirType == commontypes.WorkspaceCacheDir {
mountPath = workflowCtx.Workspace
}
job.Spec.Template.Spec.Containers[0].VolumeMounts = append(job.Spec.Template.Spec.Containers[0].VolumeMounts, corev1.VolumeMount{
Name: volumeName,
MountPath: mountPath,
SubPath: jobTaskSpec.Properties.Cache.NFSProperties.Subpath,
})
}
ensureVolumeMounts(job)
return job, nil
}
func BuildCleanJob(jobName, clusterID, workflowName string, taskID int64) (*batchv1.Job, error) {
workspace := "/workspace"
shareStorageDir := commontypes.GetShareStorageSubPathPrefix(workflowName, taskID)
image := strings.ReplaceAll(config.ReaperImage(), "${BuildOS}", "focal")
targetCluster, err := service.GetCluster(clusterID, log.SugaredLogger())
if err != nil {
return nil, fmt.Errorf("failed to find target cluster %s, err: %s", clusterID, err)
}
job := &batchv1.Job{
ObjectMeta: metav1.ObjectMeta{
Name: jobName,
},
Spec: batchv1.JobSpec{
Completions: int32Ptr(1),
Parallelism: int32Ptr(1),
BackoffLimit: int32Ptr(0),
// in case finished zombie job not cleaned up by zadig
TTLSecondsAfterFinished: int32Ptr(3600),
// in case zombie job never stop
ActiveDeadlineSeconds: int64Ptr(3600),
Template: corev1.PodTemplateSpec{
Spec: corev1.PodSpec{
RestartPolicy: corev1.RestartPolicyNever,
Containers: []corev1.Container{
{
ImagePullPolicy: corev1.PullAlways,
Name: jobName,
Image: image,
WorkingDir: workspace,
Command: []string{"/bin/sh", "-c"},
Args: []string{fmt.Sprintf("rm -rf %s", shareStorageDir)},
TerminationMessagePolicy: corev1.TerminationMessageReadFile,
TerminationMessagePath: job.JobTerminationFile,
},
},
Tolerations: buildTolerations(targetCluster.AdvancedConfig, ""),
Affinity: addNodeAffinity(targetCluster.AdvancedConfig, ""),
},
},
},
}
shareStorageName := "share-storage"
job.Spec.Template.Spec.Volumes = append(job.Spec.Template.Spec.Volumes, corev1.Volume{
Name: shareStorageName,
VolumeSource: corev1.VolumeSource{
PersistentVolumeClaim: &corev1.PersistentVolumeClaimVolumeSource{
ClaimName: targetCluster.ShareStorage.NFSProperties.PVC,
},
},
})
job.Spec.Template.Spec.Containers[0].VolumeMounts = append(job.Spec.Template.Spec.Containers[0].VolumeMounts, corev1.VolumeMount{
Name: shareStorageName,
MountPath: workspace,
})
return job, nil
}
func setJobShareStorages(job *batchv1.Job, workflowCtx *commonmodels.WorkflowTaskCtx, storageDetails []*commonmodels.StorageDetail, cluster *commonmodels.K8SCluster) {
if cluster == nil {
return
}
if cluster.ShareStorage.MediumType != commontypes.NFSMedium {
return
}
if cluster.ShareStorage.NFSProperties.PVC == "" {
return
}
if len(storageDetails) <= 0 {
return
}
// save cluster id so we can clean up share storage later
workflowCtx.ClusterIDAdd(cluster.ID.Hex())
volumeName := "share-storage"
job.Spec.Template.Spec.Volumes = append(job.Spec.Template.Spec.Volumes, corev1.Volume{
Name: volumeName,
VolumeSource: corev1.VolumeSource{
PersistentVolumeClaim: &corev1.PersistentVolumeClaimVolumeSource{
ClaimName: cluster.ShareStorage.NFSProperties.PVC,
},
},
})
for _, storageDetail := range storageDetails {
job.Spec.Template.Spec.Containers[0].VolumeMounts = append(job.Spec.Template.Spec.Containers[0].VolumeMounts, corev1.VolumeMount{
Name: volumeName,
MountPath: storageDetail.MountPath,
SubPath: storageDetail.SubPath,
})
}
}
func ensureVolumeMounts(job *batchv1.Job) {
for i := range job.Spec.Template.Spec.Containers {
mountPathMap := make(map[string]bool)
volumeMounts := []corev1.VolumeMount{}
for _, volumeMount := range job.Spec.Template.Spec.Containers[i].VolumeMounts {
if _, ok := mountPathMap[volumeMount.MountPath]; !ok {
volumeMounts = append(volumeMounts, volumeMount)
mountPathMap[volumeMount.MountPath] = true
}
}
job.Spec.Template.Spec.Containers[i].VolumeMounts = volumeMounts
}
}
func getImagePullSecrets(registries []*commonmodels.RegistryNamespace) ([]corev1.LocalObjectReference, error) {
ImagePullSecrets := []corev1.LocalObjectReference{
{
Name: setting.DefaultImagePullSecret,
},
}
for _, reg := range registries {
secretName, err := kube.GenRegistrySecretName(reg)
if err != nil {
return ImagePullSecrets, fmt.Errorf("failed to generate registry secret name: %s", err)
}
secret := corev1.LocalObjectReference{
Name: secretName,
}
ImagePullSecrets = append(ImagePullSecrets, secret)
}
return ImagePullSecrets, nil
}
func getEnvs(configMapMountDir string, jobTaskSpec *commonmodels.JobTaskFreestyleSpec) []corev1.EnvVar {
ret := make([]corev1.EnvVar, 0)
ret = append(ret, corev1.EnvVar{
Name: setting.JobConfigFile,
Value: path.Join(configMapMountDir, "job-config.xml"),
})
if !jobTaskSpec.Properties.UseHostDockerDaemon {
ret = append(ret, corev1.EnvVar{
Name: setting.DockerHost,
Value: jobTaskSpec.Properties.DockerHost,
})
}
return ret
}
func getVolumeMounts(configMapMountDir string, userHostDockerDaemon bool) []corev1.VolumeMount {
resp := make([]corev1.VolumeMount, 0)
resp = append(resp, corev1.VolumeMount{
Name: "job-config",
MountPath: configMapMountDir,
})
resp = append(resp, corev1.VolumeMount{
Name: "zadig-context",
MountPath: ZadigContextDir,
})
resp = append(resp, corev1.VolumeMount{
Name: ExecutorResourceVolumeName,
MountPath: ExecutorVolumePath,
})
if userHostDockerDaemon {
resp = append(resp, corev1.VolumeMount{
Name: "docker-sock",
MountPath: setting.DefaultDockSock,
})
}
return resp
}
func getVolumes(jobName string, userHostDockerDaemon bool) []corev1.Volume {
resp := make([]corev1.Volume, 0)
resp = append(resp, corev1.Volume{
Name: "job-config",
VolumeSource: corev1.VolumeSource{
ConfigMap: &corev1.ConfigMapVolumeSource{
LocalObjectReference: corev1.LocalObjectReference{
Name: jobName,
},
},
},
})
resp = append(resp, corev1.Volume{
Name: "zadig-context",
VolumeSource: corev1.VolumeSource{
EmptyDir: &corev1.EmptyDirVolumeSource{},
},
})
resp = append(resp, corev1.Volume{
Name: ExecutorResourceVolumeName,
VolumeSource: corev1.VolumeSource{
EmptyDir: &corev1.EmptyDirVolumeSource{},
},
})
if userHostDockerDaemon {
resp = append(resp, corev1.Volume{
Name: "docker-sock",
VolumeSource: corev1.VolumeSource{
HostPath: &corev1.HostPathVolumeSource{
Path: setting.DefaultDockSock,
Type: nil,
},
},
})
}
return resp
}
func getResourceRequirements(resReq setting.Request, resReqSpec setting.RequestSpec) corev1.ResourceRequirements {
switch resReq {
case setting.HighRequest:
return generateResourceRequirements(setting.HighRequest, setting.HighRequestSpec)
case setting.MediumRequest:
return generateResourceRequirements(setting.MediumRequest, setting.MediumRequestSpec)
case setting.LowRequest:
return generateResourceRequirements(setting.LowRequest, setting.LowRequestSpec)
case setting.MinRequest:
return generateResourceRequirements(setting.MinRequest, setting.MinRequestSpec)
case setting.DefineRequest:
return generateResourceRequirements(resReq, resReqSpec)
default:
return generateResourceRequirements(setting.DefaultRequest, setting.DefaultRequestSpec)
}
}
// generateResourceRequirements
// cpu Request:Limit=1:4
// memory default Request:Limit=1:4 ; if memoryLimit>= 8Gi,Request:Limit=1:8
func generateResourceRequirements(req setting.Request, reqSpec setting.RequestSpec) corev1.ResourceRequirements {
if req != setting.DefineRequest {
return corev1.ResourceRequirements{
Limits: corev1.ResourceList{
corev1.ResourceCPU: resource.MustParse(strconv.Itoa(reqSpec.CpuLimit) + setting.CpuUintM),
corev1.ResourceMemory: resource.MustParse(strconv.Itoa(reqSpec.MemoryLimit) + setting.MemoryUintMi),
},
Requests: corev1.ResourceList{
corev1.ResourceCPU: resource.MustParse(strconv.Itoa(reqSpec.CpuReq) + setting.CpuUintM),
corev1.ResourceMemory: resource.MustParse(strconv.Itoa(reqSpec.MemoryReq) + setting.MemoryUintMi),
},
}
}
limits := corev1.ResourceList{}
requests := corev1.ResourceList{}
if reqSpec.CpuLimit > 0 {
cpuReqInt := reqSpec.CpuLimit / 4
if cpuReqInt < 1 {
cpuReqInt = 1
}
limits[corev1.ResourceCPU] = resource.MustParse(strconv.Itoa(reqSpec.CpuLimit) + setting.CpuUintM)
requests[corev1.ResourceCPU] = resource.MustParse(strconv.Itoa(cpuReqInt) + setting.CpuUintM)
}
if reqSpec.MemoryLimit > 0 {
memoryReqInt := reqSpec.MemoryLimit / 4
if memoryReqInt >= 2*1024 {
memoryReqInt = memoryReqInt / 2
}
if memoryReqInt < 1 {
memoryReqInt = 1
}
limits[corev1.ResourceMemory] = resource.MustParse(strconv.Itoa(reqSpec.MemoryLimit) + setting.MemoryUintMi)
requests[corev1.ResourceMemory] = resource.MustParse(strconv.Itoa(memoryReqInt) + setting.MemoryUintMi)
}
// add gpu limit
if len(reqSpec.GpuLimit) > 0 {
reqSpec.GpuLimit = strings.ReplaceAll(reqSpec.GpuLimit, " ", "")
requestPair := strings.Split(reqSpec.GpuLimit, ":")
if len(requestPair) == 2 {
limits[corev1.ResourceName(requestPair[0])] = resource.MustParse(requestPair[1])
}
}
return corev1.ResourceRequirements{
Limits: limits,
Requests: requests,
}
}
func int32Ptr(i int32) *int32 { return &i }
func int64Ptr(i int64) *int64 { return &i }
func WaitPlainJobEnd(ctx context.Context, taskTimeout int, namespace, jobName string, kubeClient crClient.Client, apiServer crClient.Reader, xl *zap.SugaredLogger) config.Status {
timeout := time.After(time.Duration(taskTimeout) * time.Minute)
status, err := waitJobStart(ctx, namespace, jobName, kubeClient, apiServer, timeout, xl)
if err != nil {
xl.Errorf("wait job start error: %v", err)
}
if status != config.StatusRunning {
return status
}
return waitPlainJobEnd(ctx, taskTimeout, timeout, namespace, jobName, kubeClient, xl)
}
func waitPlainJobEnd(ctx context.Context, taskTimeout int, timeout <-chan time.Time, namespace, jobName string, kubeClient crClient.Client, xl *zap.SugaredLogger) config.Status {
// wait for the job to end.
xl.Infof("wait job to end: %s %s", namespace, jobName)
for {
select {
case <-ctx.Done():
return config.StatusCancelled
case <-timeout:
return config.StatusTimeout
default:
job, found, err := getter.GetJob(namespace, jobName, kubeClient)
if err != nil || !found {
xl.Errorf("failed to get pod with label job-name=%s %v", jobName, err)
return config.StatusFailed
}
if job.Status.Succeeded != 0 {
return config.StatusPassed
}
if job.Status.Failed != 0 {
return config.StatusFailed
}
}
time.Sleep(time.Second * 1)
}
}
func waitJobStart(ctx context.Context, namespace, jobName string, kubeClient crClient.Client, apiReader client.Reader, timeout <-chan time.Time, xl *zap.SugaredLogger) (config.Status, error) {
xl.Infof("wait job to start: %s/%s", namespace, jobName)
xl.Infof("Timeout of preparing Pod: %s.", 120*time.Second)
waitPodReadyTimeout := time.After(120 * time.Second)
var podReadyTimeout bool
for {
select {
case <-ctx.Done():
return config.StatusCancelled, nil
case <-timeout:
return config.StatusTimeout, fmt.Errorf("wait job ready timeout")
case <-waitPodReadyTimeout:
podReadyTimeout = true
default:
job, _, err := getter.GetJob(namespace, jobName, kubeClient)
if err != nil {
xl.Errorf("get job failed, namespace:%s, jobName:%s, err:%v", namespace, jobName, err)
}
if job != nil {
// Should ensure the status of pod is running
podList, err := getter.ListPods(namespace, labels.Set(getJobLabels(&JobLabel{
JobName: jobName,
})).AsSelector(), kubeClient)
if err != nil {
xl.Errorf("list pod failed, namespace:%s, jobName:%s, err:%v", namespace, jobName, err)
time.Sleep(time.Second)
continue
}
for _, pod := range podList {
if pod.Status.Phase == corev1.PodFailed {
msg := ""
for _, condition := range pod.Status.Conditions {
msg += fmt.Sprintf("type:%s, status:%s, reason:%s, message:%s\n", condition.Type, condition.Status, condition.Reason, condition.Message)
}
return config.StatusFailed, fmt.Errorf("waitJobStart: pod failed, jobName:%s, podName:%s\nconditions info: %s", jobName, pod.Name, msg)
}
if pod.Status.Phase != corev1.PodPending {
xl.Infof("waitJobStart: pod status %s namespace:%s, jobName:%s podList num %d", pod.Status.Phase, namespace, jobName, len(podList))
return config.StatusRunning, nil
}
// if pod is still pending afer 2 minutes, check pod events if is failed already
if !podReadyTimeout {
continue
}
if err := isPodFailed(pod.Name, namespace, apiReader, xl); err != nil {
return config.StatusFailed, err
}
}
}
}
time.Sleep(time.Second)
}
}
func isPodFailed(podName, namespace string, apiReader client.Reader, xl *zap.SugaredLogger) error {
selector := fields.Set{"involvedObject.name": podName, "involvedObject.kind": setting.Pod}.AsSelector()
events, err := getter.ListEvents(namespace, selector, apiReader)
if err != nil {
// list events error is not fatal
xl.Errorf("list events failed: %s", err)
return nil
}
var errMsg string
for _, event := range events {
if event.Type != "Warning" {
continue
}
// FailedScheduling means there is not enough resource to schedule the pod, so we should not fail the pod
if event.Reason == "FailedScheduling" {
continue
}
errMsg = errMsg + fmt.Sprintf("pod %s/%s event: %s\n", namespace, podName, event.Message)
}
if errMsg != "" {
return errors.New(errMsg)
}
return nil
}
func waitJobEndByCheckingConfigMap(ctx context.Context, taskTimeout <-chan time.Time, namespace, jobName string, checkFile bool, kubeClient crClient.Client, clientset kubernetes.Interface, restConfig *rest.Config, informer informers.SharedInformerFactory, jobTask *commonmodels.JobTask, ack func(), xl *zap.SugaredLogger) (status config.Status, errMsg string) {
xl.Infof("wait job to end: %s %s", namespace, jobName)
podLister := informer.Core().V1().Pods().Lister().Pods(namespace)
jobLister := informer.Batch().V1().Jobs().Lister().Jobs(namespace)
cmLister := informer.Core().V1().ConfigMaps().Lister().ConfigMaps(namespace)
for {
select {
case <-ctx.Done():
return config.StatusCancelled, ""
case <-taskTimeout:
return config.StatusTimeout, ""
default:
job, err := jobLister.Get(jobName)
if err != nil {
errMsg := fmt.Sprintf("failed to get job pod job-name=%s %v", jobName, err)
xl.Errorf(errMsg)
return config.StatusFailed, errMsg
}
// configMap name is the same as the k8s job name
cm, err := cmLister.Get(jobName)
if err != nil {
errMsg := fmt.Sprintf("failed to get job context configMap job-name=%s %v", jobName, err)
xl.Errorf(errMsg)
return config.StatusFailed, errMsg
}
// pod is still running
switch {
case job.Status.Active != 0:
pods, err := podLister.List(labels.Set{"job-name": jobName}.AsSelector())
if err != nil {
errMsg := fmt.Sprintf("failed to find pod with label job-name=%s %v", jobName, err)
xl.Errorf(errMsg)
return config.StatusFailed, errMsg
}
for _, pod := range pods {
ipod := wrapper.Pod(pod)
if ipod.Pending() {
continue
}
if ipod.Failed() {
return config.StatusFailed, ""
}
if !ipod.Finished() {
// check container whether is stuck in debug stage by checking stage file, if so, update job status to debug
switch cm.Data[commontypes.JobDebugStatusKey] {
case commontypes.JobDebugStatusBefore:
jobTask.Status = config.StatusDebugBefore
ack()
case commontypes.JobDebugStatusAfter:
jobTask.Status = config.StatusDebugAfter
ack()
case commontypes.JobDebugStatusNotIn:
if jobTask.Status == config.StatusDebugBefore || jobTask.Status == config.StatusDebugAfter {
jobTask.Status = config.StatusRunning
ack()
}