We have an issue similar to that, but it may have more to do with the firewall than it does Cerner.
Typically, the only way that we can get it to start working again is just restart both ends around the same time.
We also have an alert setup that checks if there are a bunch of messages queued up in one of our feeds to Cerner, and it restarts all of the Cerner connections then, assuming something could be wrong with the firewall.
We have to clear xlates in the firewall every once in awhile, pretty much anytime we need to NAT a new device to communicate with Cerner. So, we have to restart the connections pretty frequently.
