Getting "DeadReferenceError: Calling Stale Broker" after ipcontroller restart
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
IPython |
Confirmed
|
Medium
|
Vishal Vatsa |
Bug Description
Not Sure that is a bug or an expected behavior.
I am starting the ipcontoller such that it reuses the ipcontroller-
After the restart, if I invoke a DeadReferenceError with the following traceback:
I am using the blocking TaskClient from an ipython shell.
In [8]: tc.spin()
-------
DeadReferenceError Traceback (most recent call last)
/home/vishal/
/usr/lib/
119 scheduler is idle for some reason.
120 """
--> 121 return blockingCallFro
122
123 def queue_status(self, verbose=False):
/usr/lib/
67 @raise: any error raised during the callback chain.
68 """
---> 69 return twisted.
70
71 else:
/usr/lib/
81 result = queue.get()
82 if isinstance(result, failure.Failure):
---> 83 result.
84 return result
85
/usr/lib/
317 information if available.
318 """
--> 319 raise self.type, self.value, self.tb
320
321
DeadReferenceError: Calling Stale Broker
I have confirmed this. It also occurs for the MultiEngineClient when the controller stop and is restarted underneath a live client. The client should have reconnect logic that is triggered on a DeadReferenceError.