staticaddr: fix crash in loop-in manager shutdown

If the notification manager shuts down, sweepReqs channel is closed and sweepReq
object received from it becomes nil, causing panic in handleLoopInSweepReq. Now
this is fixed by checking if the channel was closed and if it was shutting down
loop-in manager itself.

Author: starius

staticaddr/loopin/manager.go

@@ -205,14 +205,25 @@ func (m *Manager) Run(ctx context.Context) error {
 			case request.respChan <- resp:
 
 			case <-ctx.Done():
-				// Noify subroutines that the main loop has been
-				// canceled.
+				// Notify subroutines that the main loop has
+				// been canceled.
 				close(m.exitChan)
 
 				return ctx.Err()
 			}
 
-		case sweepReq := <-sweepReqs:
+		case sweepReq, ok := <-sweepReqs:
+			if !ok {
+				// The channel has been closed, we'll stop the
+				// loop-in manager.
+				log.Debugf("Stopping loop-in manager " +
+					"(ntfnChan closed)")
+
+				close(m.exitChan)
+
+				return fmt.Errorf("ntfnChan closed")
+			}
+
 			err = m.handleLoopInSweepReq(ctx, sweepReq)
 			if err != nil {
 				log.Errorf("Error handling loop-in sweep "+