Uncaght exception in rabbitpy.IO.io class / channel0 when closing connection without stoping consumers #129
Comments
|
This may work for you. join the thread then close the connection |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Using version 2.0.1
Scenario:
Create a single rabbitpy.connection() and pass it to a thread which will use the connection to create a new channel and consumer.
In the main thread call connection.close() to close the connection.
As one would expect connections begin breaking down at this point and spawned threads throw various exceptions depending on what they were doing at the time, these can be caught in a try block and handled as one would normally do.
The daemon IO thread however while this is ongoing attempts to process a message and triggers an exception. Since the is in its own thread it can not be caught with a try() block anywhere in code.
I'll note in case it is relevant that the message channel0 is attempting to process when this happens appears routinely be a "504): CHANNEL_ERROR - expected 'channel.open'" caused by two attempts to close a single channel.
Sample code can be made to duplicate this by using the code provided in the Multi-threaded Use Notes with the minimal modifications of removing the producer to keep the receiver idle and after the consumer_thread.start() dd a short sleep followed by a connection.close()
It appears this can be avoided by tracking all the queue's globaly in the main thread and calling stop_consuming() on each before the connection.close() call
Debug log:
The text was updated successfully, but these errors were encountered: