summaryrefslogtreecommitdiff
path: root/daemon/monitor.go
blob: 77a14f4aa14ae38c89ec6fcc90d22da3804148ec (plain)
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
package daemon // import "github.com/docker/docker/daemon"

import (
	"context"
	"strconv"
	"time"

	"github.com/docker/docker/api/types"
	"github.com/docker/docker/container"
	"github.com/docker/docker/errdefs"
	libcontainerdtypes "github.com/docker/docker/libcontainerd/types"
	"github.com/docker/docker/restartmanager"
	"github.com/pkg/errors"
	"github.com/sirupsen/logrus"
)

func (daemon *Daemon) setStateCounter(c *container.Container) {
	switch c.StateString() {
	case "paused":
		stateCtr.set(c.ID, "paused")
	case "running":
		stateCtr.set(c.ID, "running")
	default:
		stateCtr.set(c.ID, "stopped")
	}
}

func (daemon *Daemon) handleContainerExit(c *container.Container, e *libcontainerdtypes.EventInfo) error {
	var exitStatus container.ExitStatus
	c.Lock()

	// Health checks will be automatically restarted if/when the
	// container is started again.
	daemon.stopHealthchecks(c)

	tsk, ok := c.Task()
	if ok {
		ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
		es, err := tsk.Delete(ctx)
		cancel()
		if err != nil {
			logrus.WithFields(logrus.Fields{
				logrus.ErrorKey: err,
				"container":     c.ID,
			}).Warn("failed to delete container from containerd")
		} else {
			exitStatus = container.ExitStatus{
				ExitCode: int(es.ExitCode()),
				ExitedAt: es.ExitTime(),
			}
		}
	}

	ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
	c.StreamConfig.Wait(ctx)
	cancel()

	c.Reset(false)

	if e != nil {
		exitStatus.ExitCode = int(e.ExitCode)
		exitStatus.ExitedAt = e.ExitedAt
		if e.Error != nil {
			c.SetError(e.Error)
		}
	}

	daemonShutdown := daemon.IsShuttingDown()
	execDuration := time.Since(c.StartedAt)
	restart, wait, err := c.RestartManager().ShouldRestart(uint32(exitStatus.ExitCode), daemonShutdown || c.HasBeenManuallyStopped, execDuration)
	if err != nil {
		logrus.WithFields(logrus.Fields{
			logrus.ErrorKey:          err,
			"container":              c.ID,
			"restartCount":           c.RestartCount,
			"exitStatus":             exitStatus,
			"daemonShuttingDown":     daemonShutdown,
			"hasBeenManuallyStopped": c.HasBeenManuallyStopped,
			"execDuration":           execDuration,
		}).Warn("ShouldRestart failed, container will not be restarted")
		restart = false
	}

	attributes := map[string]string{
		"exitCode":     strconv.Itoa(exitStatus.ExitCode),
		"execDuration": strconv.Itoa(int(execDuration.Seconds())),
	}
	daemon.Cleanup(c)

	if restart {
		c.RestartCount++
		logrus.WithFields(logrus.Fields{
			"container":     c.ID,
			"restartCount":  c.RestartCount,
			"exitStatus":    exitStatus,
			"manualRestart": c.HasBeenManuallyRestarted,
		}).Debug("Restarting container")
		c.SetRestarting(&exitStatus)
	} else {
		c.SetStopped(&exitStatus)
		if !c.HasBeenManuallyRestarted {
			defer daemon.autoRemove(c)
		}
	}
	defer c.Unlock() // needs to be called before autoRemove

	daemon.setStateCounter(c)
	cpErr := c.CheckpointTo(daemon.containersReplica)

	daemon.LogContainerEventWithAttributes(c, "die", attributes)

	if restart {
		go func() {
			err := <-wait
			if err == nil {
				// daemon.netController is initialized when daemon is restoring containers.
				// But containerStart will use daemon.netController segment.
				// So to avoid panic at startup process, here must wait util daemon restore done.
				daemon.waitForStartupDone()
				if err = daemon.containerStart(context.Background(), c, "", "", false); err != nil {
					logrus.Debugf("failed to restart container: %+v", err)
				}
			}
			if err != nil {
				c.Lock()
				c.SetStopped(&exitStatus)
				daemon.setStateCounter(c)
				c.CheckpointTo(daemon.containersReplica)
				c.Unlock()
				defer daemon.autoRemove(c)
				if err != restartmanager.ErrRestartCanceled {
					logrus.Errorf("restartmanger wait error: %+v", err)
				}
			}
		}()
	}

	return cpErr
}

// ProcessEvent is called by libcontainerd whenever an event occurs
func (daemon *Daemon) ProcessEvent(id string, e libcontainerdtypes.EventType, ei libcontainerdtypes.EventInfo) error {
	c, err := daemon.GetContainer(id)
	if err != nil {
		return errors.Wrapf(err, "could not find container %s", id)
	}

	switch e {
	case libcontainerdtypes.EventOOM:
		// StateOOM is Linux specific and should never be hit on Windows
		if isWindows {
			return errors.New("received StateOOM from libcontainerd on Windows. This should never happen")
		}

		c.Lock()
		defer c.Unlock()
		c.OOMKilled = true
		daemon.updateHealthMonitor(c)
		if err := c.CheckpointTo(daemon.containersReplica); err != nil {
			return err
		}

		daemon.LogContainerEvent(c, "oom")
	case libcontainerdtypes.EventExit:
		if ei.ProcessID == ei.ContainerID {
			return daemon.handleContainerExit(c, &ei)
		}

		exitCode := 127
		if execConfig := c.ExecCommands.Get(ei.ProcessID); execConfig != nil {
			ec := int(ei.ExitCode)
			execConfig.Lock()
			defer execConfig.Unlock()

			// Remove the exec command from the container's store only and not the
			// daemon's store so that the exec command can be inspected. Remove it
			// before mutating execConfig to maintain the invariant that
			// c.ExecCommands only contain execs in the Running state.
			c.ExecCommands.Delete(execConfig.ID)

			execConfig.ExitCode = &ec
			execConfig.Running = false

			ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
			execConfig.StreamConfig.Wait(ctx)
			cancel()

			if err := execConfig.CloseStreams(); err != nil {
				logrus.Errorf("failed to cleanup exec %s streams: %s", c.ID, err)
			}

			exitCode = ec

			go func() {
				if _, err := execConfig.Process.Delete(context.Background()); err != nil {
					logrus.WithFields(logrus.Fields{
						logrus.ErrorKey: err,
						"container":     ei.ContainerID,
						"process":       ei.ProcessID,
					}).Warn("failed to delete process")
				}
			}()
		}
		attributes := map[string]string{
			"execID":   ei.ProcessID,
			"exitCode": strconv.Itoa(exitCode),
		}
		daemon.LogContainerEventWithAttributes(c, "exec_die", attributes)
	case libcontainerdtypes.EventStart:
		c.Lock()
		defer c.Unlock()

		// This is here to handle start not generated by docker
		if !c.Running {
			ctr, err := daemon.containerd.LoadContainer(context.Background(), c.ID)
			if err != nil {
				if errdefs.IsNotFound(err) {
					// The container was started by not-docker and so could have been deleted by
					// not-docker before we got around to loading it from containerd.
					logrus.WithFields(logrus.Fields{
						logrus.ErrorKey: err,
						"container":     c.ID,
					}).Debug("could not load containerd container for start event")
					return nil
				}
				return err
			}
			tsk, err := ctr.Task(context.Background())
			if err != nil {
				if errdefs.IsNotFound(err) {
					logrus.WithFields(logrus.Fields{
						logrus.ErrorKey: err,
						"container":     c.ID,
					}).Debug("failed to load task for externally-started container")
					return nil
				}
				return err
			}
			c.SetRunning(ctr, tsk, false)
			c.HasBeenManuallyStopped = false
			c.HasBeenStartedBefore = true
			daemon.setStateCounter(c)

			daemon.initHealthMonitor(c)

			if err := c.CheckpointTo(daemon.containersReplica); err != nil {
				return err
			}
			daemon.LogContainerEvent(c, "start")
		}

	case libcontainerdtypes.EventPaused:
		c.Lock()
		defer c.Unlock()

		if !c.Paused {
			c.Paused = true
			daemon.setStateCounter(c)
			daemon.updateHealthMonitor(c)
			if err := c.CheckpointTo(daemon.containersReplica); err != nil {
				return err
			}
			daemon.LogContainerEvent(c, "pause")
		}
	case libcontainerdtypes.EventResumed:
		c.Lock()
		defer c.Unlock()

		if c.Paused {
			c.Paused = false
			daemon.setStateCounter(c)
			daemon.updateHealthMonitor(c)

			if err := c.CheckpointTo(daemon.containersReplica); err != nil {
				return err
			}
			daemon.LogContainerEvent(c, "unpause")
		}
	}
	return nil
}

func (daemon *Daemon) autoRemove(c *container.Container) {
	c.Lock()
	ar := c.HostConfig.AutoRemove
	c.Unlock()
	if !ar {
		return
	}

	err := daemon.ContainerRm(c.ID, &types.ContainerRmConfig{ForceRemove: true, RemoveVolume: true})
	if err == nil {
		return
	}
	if c := daemon.containers.Get(c.ID); c == nil {
		return
	}

	logrus.WithFields(logrus.Fields{logrus.ErrorKey: err, "container": c.ID}).Error("error removing container")
}