admin: fix handling of immediate connection failure to upstream admin

In such case, it didn't reconnect, but thought it was connected,
which eventually led to crashes like:

  Traceback (most recent call last):
    ...
    File "neo/admin/handler.py", line 130, in answerClusterState
      self.app.updateMonitorInformation(None, cluster_state=state)
    File "neo/admin/app.py", line 274, in updateMonitorInformation
      self.upstream_admin_conn.send(Packets.NotifyMonitorInformation(kw))
    File "neo/lib/connection.py", line 565, in send
      raise ConnectionClosed
  neo.lib.connection.ConnectionClosed

neo/admin/app.py

@@ -240,8 +240,13 @@ class Application(BaseApplication, Monitor):
 
     def connectToUpstreamAdmin(self):
         if self.listening_conn: # if running
-            self.upstream_admin_conn = ClientConnection(
-                self, self.upstream_admin_handler, self.upstream_admin)
+            self.upstream_admin_conn = None
+            while True:
+                conn = ClientConnection(
+                    self, self.upstream_admin_handler, self.upstream_admin)
+                if not conn.isClosed():
+                    break
+            self.upstream_admin_conn = conn
 
     def partitionTableUpdated(self):
         pt = self.pt