Up until recently, I had one Exchange 2010 server running the mailbox, client access, and hub transport roles. There was one send connector for the * address space that forwarded all mail to our hosted relay service.
Today, I added an additional Exchange 2010 server that also hosted the mailbox, client access, and hub transport roles. I did NOT add it to the existing send connector, so it was not listed as the source server in any send connector. Since I had just set it up today,(ran setup, installed the latest patches, and that's it) it did not have any mailboxes on it.
After a few hours, I noticed that mail destined for the internet was queuing up on this new server in the queue called "SMTP Relay in Active Directory Site". This new server did not have the ability to send mail to the Internet since the firewall blocked it (by design, I don't want it sending mail to the Internet).
My question is: why was the existing server trying to route outbound mail through the new server? My understanding of the Exchange routing is that if the mailbox role on the old server wants to send mail, it will always contact the hub transport role on that same server. According to http://technet.microsoft.com/en-us/library/aa998825.aspx
: "If the local Mailbox server is also running the Hub Transport server role and it is not participating in a database availability group (DAG), the local server is notified. If the local Microsoft Exchange Transport service isn't running or the local Hub Transport server can't process new mail submissions because of back pressure, another available Hub Transport server is notified."
I know the hub transport role worked fine on the old server because it had been sending mail without any issues up until I installed the new server.
I will be setting up additional servers and want to understand why this is happening and how to avoid it.