Datasheet

30
Chapter 1

Designing and Planning Messaging Services
Hub Transport Server Routing Rules
As explained before, Hub Transport server is the only Exchange Server 2007 server role
to route messages within an Exchange organization. Of course, the Edge server role can also
route messages, but only to and from the Internet.
Another difference between Exchange Server 2007 and previous Exchange ver-
sions is that any message must now be sent through a Hub Transport server,
even if the recipient is on the same Mailbox server as the sender.
Internal message routing in Exchange Server 2007 uses SMTP as its primary message-trans-
port protocol and is based on the following two basic rules:
If the message target recipient is within the same AD site the Hub Transport server delivers
the message directly to the Mailbox server where the recipient mailbox resides.
If the message is targeted to a recipient located in a different AD site the Hub Transport server
sends it directly to a Hub Transport server in the target AD site. This means that the message
does not relay to each AD site along the least-cost routing path as previous versions of Exchange
did! It will choose the target Hub Transport server using round-robin load-balancing mecha-
nisms. Only if the preferred Hub Transport server becomes unavailable will it choose another
Hub Transport server.
As most large-scale network environments are complex, some situations require special con-
figurations. What happens when the target AD site is offline due to network problems? Or what
about firewall settings where network traffic is forced to flow through specific AD sites? These
issues are covered in the following paragraphs.
Queue at Point of Failure
Exchange Server 2007 uses the least-cost routing path information when no Hub Transport
server in the destination AD site responds due to network issues or server outages. In this situ-
ation the least-cost routing path will be used in reverse order: from the destination AD site to the
source AD site. All AD sites will be contacted along this path, and if a Hub Transport server is
available, the message will be queued there in a retry state. Thus the message will be delivered
to a Hub Transport server that seems to be the closest one to the target Hub Transport server
from the IP site link cost perspective. This is called queue at point of failure.
For example, say you have Site 1, Site 2, and Site 3 that have a cost of 5 and are connected
Site 1 to Site 2, and Site 2 to Site 3. Hub Transport servers exist in all three sites. Let’s assume
that we have a message being sent from Site 1 to Site 3. Under normal behavior the Hub Trans-
port server in Site 1 would send it directly to the server in Site 3. However, if the server in Site 3
is offline, the message cannot be sent directly. Queue at point of failure would be when the Hub
Transport server delivers the message to the server in Site 2 for queuing, as it knows that this site
is closer to the target site (Site 3) than its own server is.
Hub Sites
One way to interfere with the least-cost routing path is by defining hub sites through which all
message flow must be relayed. You can think of this situation as a former hub-and-spoke design
with a messaging backbone.
81461.book Page 30 Wednesday, December 12, 2007 4:49 PM