locked
Behavior for Multiple Send Connectors with Same Address Space, Same Cost, but Different Source Servers RRS feed

  • Question

  • We have multiple Send Connectors with an Address Space of "*" and all of these connectors have the same cost.  

    What is different is that these connectors have different Source Servers defined. 

    Each of these connectors is intended for a specific site.  As an example, one connector contains the source servers for site A, the next connector contains source servers for site B, and the next connector for site C, etc.  

    Even though the connectors have the same address space and cost, Exchange is somehow able to use the correct connector that contains the local source servers where the message is being sent from.  My question is...how is this being done when the send connectors have the same address space and cost?  

    Thursday, January 18, 2018 3:15 AM

Answers

  • I agree, it should be scoped.  It's working, just want to understand why as we are in the process of migrating from Exchange 2010 to 2016.  I should also mention that each of the source servers is an Exchange 2010 multi-role server in a DAG.  

    So these are separate DAGs in each AD site? Then makes sense then. That's the transport boundary. 

    https://technet.microsoft.com/en-us/library/jj657506(v=exchg.160).aspx

    • Marked as answer by TGIF2011 Thursday, January 18, 2018 3:13 PM
    Thursday, January 18, 2018 3:07 PM

All replies

  • We have multiple Send Connectors with an Address Space of "*" and all of these connectors have the same cost.  

    What is different is that these connectors have different Source Servers defined. 

    Each of these connectors is intended for a specific site.  As an example, one connector contains the source servers for site A, the next connector contains source servers for site B, and the next connector for site C, etc.  

    Even though the connectors have the same address space and cost, Exchange is somehow able to use the correct connector that contains the local source servers where the message is being sent from.  My question is...how is this being done when the send connectors have the same address space and cost?  

    Are the connectors scoped? 

    https://technet.microsoft.com/en-us/library/jj657457(v=exchg.160).aspx

    If you select Scoped send connector, the connector is only usable by other transport servers in the same Active Directory site.

    Thursday, January 18, 2018 1:20 PM
  • The connectors are not scoped.  Could it be something related to AD sites?
    Thursday, January 18, 2018 2:40 PM
  • The connectors are not scoped.  Could it be something related to AD sites?

    Well, if each connector is intended for a specific site, then they should be scoped. Is it causing you a problem? I mean it sounds as if its working.
    Thursday, January 18, 2018 2:48 PM
  • I agree, it should be scoped.  It's working, just want to understand why as we are in the process of migrating from Exchange 2010 to 2016.  I should also mention that each of the source servers is an Exchange 2010 multi-role server in a DAG.  
    Thursday, January 18, 2018 2:53 PM
  • I agree, it should be scoped.  It's working, just want to understand why as we are in the process of migrating from Exchange 2010 to 2016.  I should also mention that each of the source servers is an Exchange 2010 multi-role server in a DAG.  

    So these are separate DAGs in each AD site? Then makes sense then. That's the transport boundary. 

    https://technet.microsoft.com/en-us/library/jj657506(v=exchg.160).aspx

    • Marked as answer by TGIF2011 Thursday, January 18, 2018 3:13 PM
    Thursday, January 18, 2018 3:07 PM
  • OK that makes sense.  Thanks.  The DAGs are stretched, but each DAG has its own primary production site.  
    Thursday, January 18, 2018 3:12 PM