Check the logs on the MID server itself to see if any recorded processing is shedding light on what's causing the stall.

As a matter of interest: did your MID exhibit the same behaviour before you did this change, or has it always stalled like that?