Mail Routing for Exchange Server DAG within two sites

Question

I've two Exchange Server 2019 configured with DAG. Exchange Servers are located at different sites. I've databases active on both servers like, DB01 is active on EX1 and DB02 is active on EX2. Now what's happening if EX2 receives an email for user Joe which is sitting on DB02 should get email instead EX2 route email to EX1 then EX1 route back it to EX2 then finally Joe receives the email.

Regards,

hi,

I've two Exchange Server 2019 configured with DAG. Exchange Servers are located at different sites. I've databases active on both servers like, DB01 is active on EX1 and DB02 is active on EX2. Now what's happening if EX2 receives an email for user Joe which is sitting on DB02 should get email instead EX2 route email to EX1 then EX1 route back it to EX2 then finally Joe receives the email.

Answer 1

I think your concern is that the Front End Transport and Transport service are picking random servers to handle the message, unlike Mailbox Transport service which always happens where the mailbox is (fixed).

Because apparently, when a message lands in your environment, it would land to either of the two servers no matter where the user is -> then is passed to either of the two servers to do some transport tasks such as categorizing (AGENT in message tracking).

Refer to the mail flow picture in https://learn.microsoft.com/en-us/exchange/mail-flow/mail-flow?view=exchserver-2019

This is annoying because if we look at this, we will think that, oh this is not efficient!

However, this is by design because both Front End Transport and Transport service work based on routing tables.

Like the Transport service, the Front End Transport service loads routing tables based on information from Active Directory, and uses delivery groups to determine how to route messages. https://learn.microsoft.com/en-us/exchange/mail-routing-exchange-2013-help

Exchange server reads from Active Directory for: 1. Active Directory sites 2. IP site links 3. All Exchange servers And then routing tables are updated with the collection of above information retrieved.

No rule set there for this process makes us feel the message tracking looks not efficient, however, this is also the key why it is quick (we will never have a problem where the mail delays because of processing which is the next hop) and stable (when reading from AD).

In some environment, I've seen customized routing table - this requires some developing work, but to be honest, I did not see necessity there as it makes the mail flow super complicated.

Simple is Beautiful.