-
Notifications
You must be signed in to change notification settings - Fork 826
/
parallel_node_reconciler.go
233 lines (199 loc) · 7.2 KB
/
parallel_node_reconciler.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
// Copyright 2021 Chaos Mesh 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 controllers
import (
"context"
"fmt"
"time"
"github.com/go-logr/logr"
corev1 "k8s.io/api/core/v1"
"k8s.io/apimachinery/pkg/types"
"k8s.io/client-go/util/retry"
"sigs.k8s.io/controller-runtime/pkg/client"
"sigs.k8s.io/controller-runtime/pkg/reconcile"
"github.com/chaos-mesh/chaos-mesh/api/v1alpha1"
"github.com/chaos-mesh/chaos-mesh/controllers/utils/recorder"
)
// ParallelNodeReconciler watches on nodes which type is Parallel
type ParallelNodeReconciler struct {
*ChildNodesFetcher
kubeClient client.Client
eventRecorder recorder.ChaosRecorder
logger logr.Logger
}
func NewParallelNodeReconciler(kubeClient client.Client, eventRecorder recorder.ChaosRecorder, logger logr.Logger) *ParallelNodeReconciler {
return &ParallelNodeReconciler{
ChildNodesFetcher: NewChildNodesFetcher(kubeClient, logger),
kubeClient: kubeClient,
eventRecorder: eventRecorder,
logger: logger,
}
}
// Reconcile is extremely like the one in SerialNodeReconciler, only allows the parallel schedule, and respawn **all** the children tasks during retry
func (it *ParallelNodeReconciler) Reconcile(ctx context.Context, request reconcile.Request) (reconcile.Result, error) {
startTime := time.Now()
defer func() {
it.logger.V(4).Info("Finished syncing for parallel node",
"node", request.NamespacedName,
"duration", time.Since(startTime),
)
}()
node := v1alpha1.WorkflowNode{}
err := it.kubeClient.Get(ctx, request.NamespacedName, &node)
if err != nil {
return reconcile.Result{}, client.IgnoreNotFound(err)
}
// only resolve parallel nodes
if node.Spec.Type != v1alpha1.TypeParallel {
return reconcile.Result{}, nil
}
it.logger.V(4).Info("resolve parallel node", "node", request)
// make effects, create/remove children nodes
err = it.syncChildNodes(ctx, node)
if err != nil {
return reconcile.Result{}, err
}
// update status
updateError := retry.RetryOnConflict(retry.DefaultRetry, func() error {
nodeNeedUpdate := v1alpha1.WorkflowNode{}
err := it.kubeClient.Get(ctx, request.NamespacedName, &nodeNeedUpdate)
if err != nil {
return err
}
activeChildren, finishedChildren, err := it.fetchChildNodes(ctx, nodeNeedUpdate)
if err != nil {
return err
}
nodeNeedUpdate.Status.FinishedChildren = nil
for _, finishedChild := range finishedChildren {
nodeNeedUpdate.Status.FinishedChildren = append(nodeNeedUpdate.Status.FinishedChildren,
corev1.LocalObjectReference{
Name: finishedChild.Name,
})
}
nodeNeedUpdate.Status.ActiveChildren = nil
for _, activeChild := range activeChildren {
nodeNeedUpdate.Status.ActiveChildren = append(nodeNeedUpdate.Status.ActiveChildren,
corev1.LocalObjectReference{
Name: activeChild.Name,
})
}
// TODO: also check the consistent between spec in task and the spec in child node
if len(finishedChildren) == len(nodeNeedUpdate.Spec.Children) {
if !WorkflowNodeFinished(nodeNeedUpdate.Status) {
it.eventRecorder.Event(&nodeNeedUpdate, recorder.NodeAccomplished{})
}
SetCondition(&nodeNeedUpdate.Status, v1alpha1.WorkflowNodeCondition{
Type: v1alpha1.ConditionAccomplished,
Status: corev1.ConditionTrue,
Reason: "",
})
} else {
SetCondition(&nodeNeedUpdate.Status, v1alpha1.WorkflowNodeCondition{
Type: v1alpha1.ConditionAccomplished,
Status: corev1.ConditionFalse,
Reason: "",
})
}
return it.kubeClient.Status().Update(ctx, &nodeNeedUpdate)
})
if updateError != nil {
it.logger.Error(err, "failed to update the status of node", "node", request)
return reconcile.Result{}, updateError
}
return reconcile.Result{}, nil
}
func (it *ParallelNodeReconciler) syncChildNodes(ctx context.Context, node v1alpha1.WorkflowNode) error {
// empty parallel node
if len(node.Spec.Children) == 0 {
it.logger.V(4).Info("empty parallel node, NOOP",
"node", fmt.Sprintf("%s/%s", node.Namespace, node.Name),
)
return nil
}
if WorkflowNodeFinished(node.Status) {
return nil
}
activeChildNodes, finishedChildNodes, err := it.fetchChildNodes(ctx, node)
if err != nil {
return err
}
existsChildNodes := append(activeChildNodes, finishedChildNodes...)
var taskNamesOfNodes []string
for _, childNode := range existsChildNodes {
taskNamesOfNodes = append(taskNamesOfNodes, getTaskNameFromGeneratedName(childNode.GetName()))
}
var tasksToStartup []string
if len(existsChildNodes) == 0 {
tasksToStartup = node.Spec.Children
}
// TODO: check the specific of task and workflow nodes
// the definition of Spec.Children changed, remove all the existed nodes
if len(existsChildNodes) > 0 && (len(setDifference(taskNamesOfNodes, node.Spec.Children)) > 0 || len(setDifference(node.Spec.Children, taskNamesOfNodes)) > 0) {
tasksToStartup = node.Spec.Children
var nodesToCleanup []string
for _, item := range existsChildNodes {
nodesToCleanup = append(nodesToCleanup, item.Name)
}
it.eventRecorder.Event(&node, recorder.RerunBySpecChanged{CleanedChildrenNode: nodesToCleanup})
for _, childNode := range existsChildNodes {
// best effort deletion
err := it.kubeClient.Delete(ctx, &childNode)
if err != nil {
it.logger.Error(err, "failed to delete outdated child node",
"node", fmt.Sprintf("%s/%s", node.Namespace, node.Name),
"child node", fmt.Sprintf("%s/%s", childNode.Namespace, childNode.Name),
)
}
}
}
if len(tasksToStartup) == 0 {
it.logger.Info("no need to spawn new child node", "node", fmt.Sprintf("%s/%s", node.Namespace, node.Name))
return nil
}
parentWorkflow := v1alpha1.Workflow{}
err = it.kubeClient.Get(ctx, types.NamespacedName{
Namespace: node.Namespace,
Name: node.Spec.WorkflowName,
}, &parentWorkflow)
if err != nil {
it.logger.Error(err, "failed to fetch parent workflow",
"node", fmt.Sprintf("%s/%s", node.Namespace, node.Name),
"workflow name", node.Spec.WorkflowName)
return err
}
childNodes, err := renderNodesByTemplates(&parentWorkflow, &node, tasksToStartup...)
if err != nil {
it.logger.Error(err, "failed to render children childNodes",
"node", fmt.Sprintf("%s/%s", node.Namespace, node.Name))
return err
}
var childrenNames []string
for _, childNode := range childNodes {
err := it.kubeClient.Create(ctx, childNode)
if err != nil {
it.logger.Error(err, "failed to create child node",
"node", fmt.Sprintf("%s/%s", node.Namespace, node.Name),
"child node", childNode)
return err
}
childrenNames = append(childrenNames, childNode.Name)
}
it.eventRecorder.Event(&node, recorder.NodesCreated{ChildNodes: childrenNames})
it.logger.Info("parallel node spawn new child node",
"node", fmt.Sprintf("%s/%s", node.Namespace, node.Name),
"child node", childrenNames)
return nil
}