-
Notifications
You must be signed in to change notification settings - Fork 651
/
types.go
161 lines (131 loc) · 6.08 KB
/
types.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
// Copyright 2019 The Kubeflow 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 v1
import (
"k8s.io/api/core/v1"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
)
// +k8s:deepcopy-gen=true
// JobStatus represents the current observed state of the training job.
type JobStatus struct {
// An array of current observed job conditions.
Conditions []JobCondition `json:"conditions"`
// A map from ReplicaType (key) to ReplicaStatus (value), specifying the status of each replica.
ReplicaStatuses map[ReplicaType]*ReplicaStatus `json:"replicaStatuses"`
// Represents the time when the job was acknowledged by the job controller.
// It is not guaranteed to be set in happens-before order across separate operations.
// It is represented in RFC3339 form and is in UTC.
StartTime *metav1.Time `json:"startTime,omitempty"`
// Represents the time when the job was completed. It is not guaranteed to
// be set in happens-before order across separate operations.
// It is represented in RFC3339 form and is in UTC.
CompletionTime *metav1.Time `json:"completionTime,omitempty"`
// Represents the last time when the job was reconciled. It is not guaranteed to
// be set in happens-before order across separate operations.
// It is represented in RFC3339 form and is in UTC.
LastReconcileTime *metav1.Time `json:"lastReconcileTime,omitempty"`
}
// ReplicaType represents the type of the job replica. Each operator (e.g. TensorFlow, PyTorch)
// needs to define its own set of ReplicaTypes.
type ReplicaType string
// ReplicaStatus represents the current observed state of the replica.
type ReplicaStatus struct {
// The number of actively running pods.
Active int32 `json:"active,omitempty"`
// The number of pods which reached phase Succeeded.
Succeeded int32 `json:"succeeded,omitempty"`
// The number of pods which reached phase Failed.
Failed int32 `json:"failed,omitempty"`
}
// +k8s:deepcopy-gen=true
// ReplicaSpec is a description of the job replica.
type ReplicaSpec struct {
// The desired number of replicas of the given template.
// If unspecified, defaults to 1.
Replicas *int32 `json:"replicas,omitempty"`
// Describes the pod that will be created for this replica. Note that
// RestartPolicy in PodTemplateSpec will be overidden by RestartPolicy in ReplicaSpec.
Template v1.PodTemplateSpec `json:"template,omitempty"`
// Restart policy for all replicas within the job.
// One of Always, OnFailure, Never, or ExitCode.
// Defaults to Never.
RestartPolicy RestartPolicy `json:"restartPolicy,omitempty"`
}
// +k8s:deepcopy-gen=true
// JobCondition describes the state of the job at a certain point.
type JobCondition struct {
// Type of job condition.
Type JobConditionType `json:"type"`
// Status of the condition, one of True, False, or Unknown.
Status v1.ConditionStatus `json:"status"`
// The reason for the condition's last transition.
Reason string `json:"reason,omitempty"`
// A readable message indicating details about the transition.
Message string `json:"message,omitempty"`
// The last time this condition was updated.
LastUpdateTime metav1.Time `json:"lastUpdateTime,omitempty"`
// Last time the condition transitioned from one status to another.
LastTransitionTime metav1.Time `json:"lastTransitionTime,omitempty"`
}
// JobConditionType defines all possible types of JobStatus. Can be one of:
// Created, Running, Restarting, Succeeded, or Failed.
type JobConditionType string
const (
// JobCreated means the job has been accepted by the system,
// but one or more of the pods/services has not been started.
// This includes time before pods being scheduled and launched.
JobCreated JobConditionType = "Created"
// JobRunning means all sub-resources (e.g. services/pods) of this job
// have been successfully scheduled and launched.
// The training is running without error.
JobRunning JobConditionType = "Running"
// JobRestarting means one or more sub-resources (e.g. services/pods) of this job
// reached phase failed but maybe restarted according to it's restart policy
// which specified by user in v1.PodTemplateSpec.
// The training is freezing/pending.
JobRestarting JobConditionType = "Restarting"
// JobSucceeded means all sub-resources (e.g. services/pods) of this job
// reached phase have terminated in success.
// The training is complete without error.
JobSucceeded JobConditionType = "Succeeded"
// JobFailed means one or more sub-resources (e.g. services/pods) of this job
// reached phase failed with no restarting.
// The training has failed its execution.
JobFailed JobConditionType = "Failed"
)
// CleanPodPolicy describes how to deal with pods when the job is finished. Can be one
// of: All, Running, or None.
type CleanPodPolicy string
const (
CleanPodPolicyUndefined CleanPodPolicy = ""
CleanPodPolicyAll CleanPodPolicy = "All"
CleanPodPolicyRunning CleanPodPolicy = "Running"
CleanPodPolicyNone CleanPodPolicy = "None"
)
// RestartPolicy describes how the replicas should be restarted.
// Can be one of: Always, OnFailure, Never, or ExitCode.
// If none of the following policies is specified, the default one
// is RestartPolicyAlways.
type RestartPolicy string
const (
RestartPolicyAlways RestartPolicy = "Always"
RestartPolicyOnFailure RestartPolicy = "OnFailure"
RestartPolicyNever RestartPolicy = "Never"
// `ExitCode` policy means that user should add exit code by themselves,
// The job operator will check these exit codes to
// determine the behavior when an error occurs:
// - 1-127: permanent error, do not restart.
// - 128-255: retryable error, will restart the pod.
RestartPolicyExitCode RestartPolicy = "ExitCode"
)