Here's the scenario:
2 Exchange 2013 Servers configured with combined CAS and Mailbox roles.
While simulating an outage on one server (ServerA) by restarting it, any inbound mail (e.g. from the Internet) sits in the "SMTP Delivery to Mailbox" queue in Retry on ServerB until the ServerA is back up. I would expect the mail to be delivered successfully as the mailbox databases all mounted successfully on ServerB when ServerA went down.
It's looks almost as if the "SMTP Delivery to Mailbox" queue has somehow cached the active mailbox information, i.e. "When I last checked the mailbox databases were active on ServerA, so I'll keep trying to deliver there".
Any thoughts on what is going on here?
Tony www.activedir.org Blog: www.open-a-socket.com