Hi,

Don't know if anyone got this problem yet but I am currently testing a migration from Exchange 2010 to Exchange 2013 (in a lab) and I have a SMTP issue.

First I had one w2k8r2 domain controller and one w2k8r2 member with Exchange 2010 on it (with one valid domain and ssl certificate). No Edge Transport Server.

I recently added a Windows 2012 server and Exchange 2013 server in this lab domain and started migrating things to see if everything worked (I moved the SSL certificate, made the Exchange 2013 receive connectors active and also moved a mailbox).

However, there is one item that eludes me at this point: the new server can't send emails out (but it receives just fine).

On my send connector, I removed the Exchange 2010 server and added the new Exchange 2013 server instead (all other values remained the same).

Now when I try to send emails out with this new Exchange 2013 connector, I am getting a message that says (in the Exchange Queue Viewer): "A local loop was detected".

So I started looking at the problem deeper and found out that the transport role is not installed (mailbox, client access and management tools are installed).

How do I add the role?

Seems like the Transport role has been integrated inside the Mailbox Server. Now I need to find out why emails are stuck in a loop. I wonder what is missing in my setup. 

Hello,

I recommend you use get-sendconnector | fl cmdlet to check you use DNS MX record to route mail or smart hosts to route mail. If you use DNS MX record to route mail, please use nslookup to check if the record is correct.

In order to troubleshooting your issue, please post the NDR information.

If you have any feedback on our support, please click here

I fixed the issue but I am not too sure how... First, I ran a "setup /m:upgrade" in Exchange 2013 and then emails started to be sent. I was also having "Subject: Undeliverable: Inbound proxy probe" cryptic messages in the Queue Viewer but I found out that my webmail.domain.com address was not resolving correctly (in my public DNS service).

With these two actions, the server is now working A1. All I have left to do is to decommission the Exchange 2010 server and the cycle will be complete.

Thanks.

Hello,

I'm glad to hear that.

This issue seems occured by wrong DNS MX record.

At present, there is no technet article or blog to explain upgrade exchange 2010 to exchange 2013 in detail.

If you have any feedback on our support, please click here

So bizarre. The setup upgrade fixed the whole thing. Now everything is flowing correctly, direct MX or Smart Host. I just can't explain it.

Hello,

I'm Sorry.

There is no related article to explain this point.

If you have no other issues, I will make your issue as answer.

If you have any feedback on our support, please click here