WebSocket transport: Unclean disconnects are being treated as clean disconnects #1254

Closed
NTaylorMullen opened this Issue Jan 7, 2013 · 1 comment

Comments

Projects
None yet
2 participants
Contributor

NTaylorMullen commented Jan 7, 2013

Currently when a client loses internet connection the WebSocket transport is being closed as if the disconnect were clean. This is resulting in preemptive disconnects allowing little time for client reconnect.

The correct functionality for this would be for the transport to treat network losses as unclean disconnects and therefore let the Heartbeat timeout connections (this is the functionality for all other ForeverTransports).

NTaylorMullen was assigned Jan 7, 2013

@NTaylorMullen NTaylorMullen added a commit that referenced this issue Jan 8, 2013

@NTaylorMullen NTaylorMullen Added a check in the OperationCanceledException to double check if we…
… should close the socket "cleanly"

- OperationCanceledException is called differently on the .NET client
and the server.
- Also found an issue where if we have an error that's thrown that is
not ObjectDisposedException or OperationCanceledException the WebSocket
would remain open.

#1254
ab63f66

@NTaylorMullen NTaylorMullen added a commit that referenced this issue Jan 8, 2013

@NTaylorMullen NTaylorMullen Added a unit test to verify that the websocket handler causes an uncl…
…ean close when throwing an operationcanceledexception

#1254
89dd1d3

@NTaylorMullen NTaylorMullen added a commit that referenced this issue Jan 9, 2013

@NTaylorMullen NTaylorMullen Created a temporary fix to work more correctly with the WebSocket tra…
…nsport server.

- This covers all test case passes however there are edge cases having
to do with network disconnect that still need to be fixed

#1254
10d2711
Contributor

Xiaohongt commented Jan 9, 2013

verified

Xiaohongt closed this Jan 9, 2013

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment