If a registered extension responds to a vCD request in more than 10 seconds we are seeing erratic vCD behavior. We see that the vCD cell does not retreive messages that took more than 10 seconds to be pushed in the amqp queue. Also on subsequent request vCD retreives the old message and fails the new REST request while the correct response is still queued in the amqp queue. The only way to fix this issue is by purging the amqp queue and re-issuing the REST request. vCD should fix this issue by continuing to retreive messages from the queue until a message with the right correlation id is found.
Is this is known bug in vCD? Are there any plans to fix it?