none
Question Regarding Mail Routing within Active Directory Site

    Question

  • See screenshot below. 

    I have 3 Exchange Servers in the same AD site. Two of the servers (one of them is named sbdmsxp02) are running Exchange 2013 and are members of a DAG. The third Exchange Server (SBDEMSI) is running 2016 and is not a member of the DAG. There is a send connector that relays email to an external smart host. This is the only send connector that can route to the internet. The only source server on this send connector is the Exchange 2016 server (SBDEMSI).

    I have an internal application that sends messages to my clients. The problem is that the application generates a message, and sends the message to the 2016 server (SBDEMSI). The 2016 server than sends the message to one of the 2013 servers (SBDMSXP02). This 2013 server then sends the message back to the 2016 server, before the 2016 server uses the send connector to route email to the internet. What would cause this and how do I prevent it from happening?

     


    Cheers,

    Ryan

    Microsoft Server Engineer

    Blog:   Twitter:   LinkedIn:   

    Please remember to mark the replies as answers if they help.

    Note: Posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Sunday, September 24, 2017 2:44 AM

Answers

  • I have to tell people this all the time.  Don't shut down Exchange servers to see if anything will break because something will always break.  If the server is in the topology, Exchange expects it to be up.

    If you have any send connectors with the server as a source transport server, change them so it's not.  If you have anything pointing inbound traffic to the server, redirect them.  You're welcome to turn up protocol logging on all receive connectors to see if any non-Exchange server traffic is coming in if you want.  You can also look at the IIS logs to see if anything is trying to use the CAS functions.  When you're reasonably satisfied that nothing is using the server, uninstall it.


    Ed Crowley MVP "There are seldom good technological solutions to behavioral problems."
    Celebrating 20 years of providing Exchange peer support!

    • Proposed as answer by Alex Sun MS Friday, September 29, 2017 9:42 AM
    • Marked as answer by Ryan Nemeth Monday, October 9, 2017 3:24 PM
    Wednesday, September 27, 2017 4:42 AM
    Moderator

All replies

  • Why do you want to prevent this from happening?

    It could be that the message is being sent between hosts when it goes from front-end to back-end transport, but you didn't show the port that's being used.  In any case, this is completely normal.  Any Exchange transport server in the site will be randomly selected for the next hop.  If for whatever reason you don't want the Exchange 2016 server to be used this way, you could move it to a different AD site.


    Ed Crowley MVP "There are seldom good technological solutions to behavioral problems."
    Celebrating 20 years of providing Exchange peer support!

    • Proposed as answer by Alex Sun MS Monday, September 25, 2017 6:57 AM
    • Unproposed as answer by Ryan Nemeth Tuesday, September 26, 2017 7:59 PM
    Sunday, September 24, 2017 4:04 PM
    Moderator
  • Thanks Ed. You've helped me out in other posts regarding Exchange and I greatly appreciate you taking the time to respond.

    What I'm really trying to accomplish here is retiring the Exchange 2013 servers. I'm fairly new to this environment, and unfortunately it is very large but not well documented. We have several web servers (and possibly other servers) that were relaying messages through the 2013 servers. I have since updated all of these servers to relay through the Exchange 2016 server. 

    I was hoping to just shutdown the Exchange 2013 servers for a few days to verify nothing else in the environment is dependent upon them. However, whenever I shut down the two Exchange 2013 servers, the ecp and owa on the Exchange 2016 server becomes unavailable. I get an HTTP 500 when trying to browse to either of these web pages. These servers are all in the same AD site. I did some troubleshooting and found that stopping the AD Topology Service on either of the Exchange 2013 servers causes this to happen. Any idea why this would be? 


    Cheers,

    Ryan

    Microsoft Server Engineer

    Blog:   Twitter:   LinkedIn:   

    Please remember to mark the replies as answers if they help.

    Note: Posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Tuesday, September 26, 2017 6:42 PM
  • I have to tell people this all the time.  Don't shut down Exchange servers to see if anything will break because something will always break.  If the server is in the topology, Exchange expects it to be up.

    If you have any send connectors with the server as a source transport server, change them so it's not.  If you have anything pointing inbound traffic to the server, redirect them.  You're welcome to turn up protocol logging on all receive connectors to see if any non-Exchange server traffic is coming in if you want.  You can also look at the IIS logs to see if anything is trying to use the CAS functions.  When you're reasonably satisfied that nothing is using the server, uninstall it.


    Ed Crowley MVP "There are seldom good technological solutions to behavioral problems."
    Celebrating 20 years of providing Exchange peer support!

    • Proposed as answer by Alex Sun MS Friday, September 29, 2017 9:42 AM
    • Marked as answer by Ryan Nemeth Monday, October 9, 2017 3:24 PM
    Wednesday, September 27, 2017 4:42 AM
    Moderator