Fixes a race condition in client events monitoring

In cases where there is high latency (ie, not-local network)
`waitExitOrRemoved` was not receiving events for short-lived containers.
This caused the client to hang while waiting for a notification that the
container has stopped.

This happens because `client.Events()` returns immediately and spins a
goroutine up to process events. The problem here is it returns before
the request to the events endpoint is even made.
Even without high-latency issues, there is no guarantee that the
goroutine is even scheduled by the time the function returns.

Signed-off-by: Brian Goff <cpuguy83@gmail.com>
(cherry picked from commit 47585996bf)
Signed-off-by: Victor Vieux <vieux@docker.com>
Signed-off-by: Vincent Demeester <vincent@sbr.pm>

client/events.go

@@ -22,17 +22,20 @@ func (cli *Client) Events(ctx context.Context, options types.EventsOptions) (<-c
 	messages := make(chan events.Message)
 	errs := make(chan error, 1)
 
+	started := make(chan struct{})
 	go func() {
 		defer close(errs)
 
 		query, err := buildEventsQueryParams(cli.version, options)
 		if err != nil {
+			close(started)
 			errs <- err
 			return
 		}
 
 		resp, err := cli.get(ctx, "/events", query, nil)
 		if err != nil {
+			close(started)
 			errs <- err
 			return
 		}
@@ -40,6 +43,7 @@ func (cli *Client) Events(ctx context.Context, options types.EventsOptions) (<-c
 
 		decoder := json.NewDecoder(resp.body)
 
+		close(started)
 		for {
 			select {
 			case <-ctx.Done():
@@ -61,6 +65,7 @@ func (cli *Client) Events(ctx context.Context, options types.EventsOptions) (<-c
 			}
 		}
 	}()
+	<-started
 
 	return messages, errs
 }