http://bugzilla.moblin.org/show_bug.cgi?id=3427 --- Comment #12 from Chen Congwu <congwu.chen(a)intel.com&gt; 2009-07-21 03:34:24 --- (In reply to comment #11) So we can register a callback for TransportAgent which will be called regularly, during the callback in EvolutionSyncClient, we can decide a timeout and retry. Yes,it looks better. It's not easy to drop reply at client side; for the real case the reply is never received so the client is blocking; it can not be emulated by getting the reply and ignoring it(the client will not block).This can be emulated by a tcp proxy which reads the request but does not reply. Duplicate requests in different sessions should have already been covered by existing interrupt tests; but not duplicate request in the same session. This can be added of course. The two scenarios can be tested separately as above, but is not easy to combine. I think the message should have reached the server in this case, if the request sending is not successful, the transport should complain (TCP should work here), right? -- Configure bugmail: http://bugzilla.moblin.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching someone on the CC list of the bug.