forked from argoproj/argo-rollouts
/
controller.go
228 lines (198 loc) · 8.33 KB
/
controller.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
package controller
import (
"fmt"
"time"
"github.com/pkg/errors"
log "github.com/sirupsen/logrus"
corev1 "k8s.io/api/core/v1"
"k8s.io/apimachinery/pkg/util/runtime"
utilruntime "k8s.io/apimachinery/pkg/util/runtime"
"k8s.io/apimachinery/pkg/util/wait"
"k8s.io/client-go/dynamic"
appsinformers "k8s.io/client-go/informers/apps/v1"
batchinformers "k8s.io/client-go/informers/batch/v1"
coreinformers "k8s.io/client-go/informers/core/v1"
"k8s.io/client-go/kubernetes"
"k8s.io/client-go/kubernetes/scheme"
typedcorev1 "k8s.io/client-go/kubernetes/typed/core/v1"
"k8s.io/client-go/tools/cache"
"k8s.io/client-go/tools/record"
"k8s.io/client-go/util/workqueue"
"github.com/argoproj/argo-rollouts/analysis"
"github.com/argoproj/argo-rollouts/controller/metrics"
"github.com/argoproj/argo-rollouts/experiments"
clientset "github.com/argoproj/argo-rollouts/pkg/client/clientset/versioned"
rolloutscheme "github.com/argoproj/argo-rollouts/pkg/client/clientset/versioned/scheme"
informers "github.com/argoproj/argo-rollouts/pkg/client/informers/externalversions/rollouts/v1alpha1"
"github.com/argoproj/argo-rollouts/rollout"
"github.com/argoproj/argo-rollouts/service"
)
const controllerAgentName = "rollouts-controller"
const (
// DefaultRolloutResyncPeriod Default time in seconds for rollout resync period
DefaultRolloutResyncPeriod = 15 * 60
// DefaultMetricsPort Default port to expose the metrics endpoint
DefaultMetricsPort = 8090
// DefaultRolloutThreads Default number of rollout worker threads to start with the controller
DefaultRolloutThreads = 10
// DefaultExperimentThreads Default number of experiment worker threads to start with the controller
DefaultExperimentThreads = 10
// DefaultAnalysisThreads Default number of analysis worker threads to start with the controller
DefaultAnalysisThreads = 30
// DefaultServiceThreads Default number of service worker threads to start with the controller
DefaultServiceThreads = 10
)
// Manager is the controller implementation for Argo-Rollout resources
type Manager struct {
metricsServer *metrics.MetricsServer
rolloutController *rollout.RolloutController
experimentController *experiments.ExperimentController
analysisController *analysis.AnalysisController
serviceController *service.ServiceController
rolloutSynced cache.InformerSynced
experimentSynced cache.InformerSynced
analysisRunSynced cache.InformerSynced
analysisTemplateSynced cache.InformerSynced
serviceSynced cache.InformerSynced
jobSynced cache.InformerSynced
replicasSetSynced cache.InformerSynced
rolloutWorkqueue workqueue.RateLimitingInterface
serviceWorkqueue workqueue.RateLimitingInterface
experimentWorkqueue workqueue.RateLimitingInterface
analysisRunWorkqueue workqueue.RateLimitingInterface
defaultIstioVersion string
}
// NewManager returns a new manager to manage all the controllers
func NewManager(
namespace string,
kubeclientset kubernetes.Interface,
argoprojclientset clientset.Interface,
dynamicclientset dynamic.Interface,
replicaSetInformer appsinformers.ReplicaSetInformer,
servicesInformer coreinformers.ServiceInformer,
jobInformer batchinformers.JobInformer,
rolloutsInformer informers.RolloutInformer,
experimentsInformer informers.ExperimentInformer,
analysisRunInformer informers.AnalysisRunInformer,
analysisTemplateInformer informers.AnalysisTemplateInformer,
resyncPeriod time.Duration,
instanceID string,
metricsPort int,
defaultIstioVersion string,
) *Manager {
utilruntime.Must(rolloutscheme.AddToScheme(scheme.Scheme))
log.Info("Creating event broadcaster")
// Create event broadcaster
// Add argo-rollouts custom resources to the default Kubernetes Scheme so Events can be
// logged for argo-rollouts types.
eventBroadcaster := record.NewBroadcaster()
eventBroadcaster.StartLogging(log.Infof)
eventBroadcaster.StartRecordingToSink(&typedcorev1.EventSinkImpl{Interface: kubeclientset.CoreV1().Events("")})
recorder := eventBroadcaster.NewRecorder(scheme.Scheme, corev1.EventSource{Component: controllerAgentName})
metricsAddr := fmt.Sprintf("0.0.0.0:%d", metricsPort)
rolloutWorkqueue := workqueue.NewNamedRateLimitingQueue(workqueue.DefaultControllerRateLimiter(), "Rollouts")
experimentWorkqueue := workqueue.NewNamedRateLimitingQueue(workqueue.DefaultControllerRateLimiter(), "Experiments")
analysisRunWorkqueue := workqueue.NewNamedRateLimitingQueue(workqueue.DefaultControllerRateLimiter(), "AnalysisRuns")
serviceWorkqueue := workqueue.NewNamedRateLimitingQueue(workqueue.DefaultControllerRateLimiter(), "Services")
metricsServer := metrics.NewMetricsServer(metricsAddr, rolloutsInformer.Lister())
rolloutController := rollout.NewRolloutController(
namespace,
kubeclientset,
argoprojclientset,
dynamicclientset,
experimentsInformer,
analysisRunInformer,
analysisTemplateInformer,
replicaSetInformer,
servicesInformer,
rolloutsInformer,
resyncPeriod,
rolloutWorkqueue,
serviceWorkqueue,
metricsServer,
recorder,
defaultIstioVersion)
experimentController := experiments.NewExperimentController(
kubeclientset,
argoprojclientset,
replicaSetInformer,
experimentsInformer,
analysisRunInformer,
analysisTemplateInformer,
resyncPeriod,
rolloutWorkqueue,
experimentWorkqueue,
metricsServer,
recorder)
analysisController := analysis.NewAnalysisController(
kubeclientset,
argoprojclientset,
analysisRunInformer,
jobInformer,
resyncPeriod,
analysisRunWorkqueue,
metricsServer,
recorder)
serviceController := service.NewServiceController(
kubeclientset,
servicesInformer,
rolloutsInformer,
resyncPeriod,
rolloutWorkqueue,
serviceWorkqueue,
metricsServer)
cm := &Manager{
metricsServer: metricsServer,
rolloutSynced: rolloutsInformer.Informer().HasSynced,
serviceSynced: servicesInformer.Informer().HasSynced,
jobSynced: jobInformer.Informer().HasSynced,
experimentSynced: experimentsInformer.Informer().HasSynced,
analysisRunSynced: analysisRunInformer.Informer().HasSynced,
analysisTemplateSynced: analysisTemplateInformer.Informer().HasSynced,
replicasSetSynced: replicaSetInformer.Informer().HasSynced,
rolloutWorkqueue: rolloutWorkqueue,
experimentWorkqueue: experimentWorkqueue,
analysisRunWorkqueue: analysisRunWorkqueue,
serviceWorkqueue: serviceWorkqueue,
rolloutController: rolloutController,
serviceController: serviceController,
experimentController: experimentController,
analysisController: analysisController,
defaultIstioVersion: defaultIstioVersion,
}
return cm
}
// Run will set up the event handlers for types we are interested in, as well
// as syncing informer caches and starting workers. It will block until stopCh
// is closed, at which point it will shutdown the workqueue and wait for
// workers to finish processing their current work items.
func (c *Manager) Run(rolloutThreadiness, serviceThreadiness, experimentThreadiness, analysisThreadiness int, stopCh <-chan struct{}) error {
defer runtime.HandleCrash()
defer c.serviceWorkqueue.ShutDown()
defer c.rolloutWorkqueue.ShutDown()
defer c.experimentWorkqueue.ShutDown()
defer c.analysisRunWorkqueue.ShutDown()
// Wait for the caches to be synced before starting workers
log.Info("Waiting for controller's informer caches to sync")
if ok := cache.WaitForCacheSync(stopCh, c.serviceSynced, c.jobSynced, c.rolloutSynced, c.experimentSynced, c.analysisRunSynced, c.analysisTemplateSynced, c.replicasSetSynced); !ok {
return fmt.Errorf("failed to wait for caches to sync")
}
// Start the informer factories to begin populating the informer caches
log.Info("Starting Controllers")
go wait.Until(func() { c.rolloutController.Run(rolloutThreadiness, stopCh) }, time.Second, stopCh)
go wait.Until(func() { c.serviceController.Run(serviceThreadiness, stopCh) }, time.Second, stopCh)
go wait.Until(func() { c.experimentController.Run(experimentThreadiness, stopCh) }, time.Second, stopCh)
go wait.Until(func() { c.analysisController.Run(analysisThreadiness, stopCh) }, time.Second, stopCh)
log.Info("Started controller")
go func() {
log.Infof("Starting Metric Server at %s", c.metricsServer.Addr)
err := c.metricsServer.ListenAndServe()
if err != nil {
err = errors.Wrap(err, "Starting Metric Server")
log.Fatal(err)
}
}()
<-stopCh
log.Info("Shutting down workers")
return nil
}