Hi
I’m having difficulties installing Exchange 2013 on a Windows 2012 R2 Standard Server in a Windows 2012 R2 Active Directory environment. I’ve already lost 3 days installing the Exchange server and the Active Directory several times without succeeding.
I have followed the instructions on the Microsoft web sites (Prerequisites, etc.) but nothing seems to work. Or at least what I was trying to achieve didn’t work.
As I’m currently not having full access to the systems I can only try to describe what I did, what worked and what the issue is. I’m thankful for every input.
I set up a new domain controller (the only one in the network, as there are only a few machines and redundancy isn’t a requirement at the moment and I wanted to keep things as simple as possible at the beginning) with the features of a Windows 2012 R2 AD DS server. Afterwards, I’ve installed the Windows 2012 R2 Standard server for the Exchange 2013. I’ve installed the Prerequisites, created a new domain account, which is group member of enterprise and schema admins and joined the domain with this machine. Afterwards I’ve installed the Exchange 2013 (SP1) Server with mailbox and client access roles while checking for updates (none available according to the installer). Exchange installed successfully. On the DC I’ll see all the new items that have been added and how the user’s / machine’s groups have changed. Furthermore, it is possible to access the ecp and owa web sites (as well as the other default sites that have been installed, but I wasn’t playing around with them later). I’ve created a mailbox for a test domain user and was able to access it via the owa web interface. So far so good (at least I think so…).
Before I describe the issue I’m experiencing I probably should add a few things to the network topology first. The DC and Exchange are in the same network zone and can directly communicate with each other. Between this server zone and the client zone is a firewall (obviously). Several ports have been opened to allow the clients to interact with the DC and the Exchange. I had to edit 2 registry entries on the DC so that the RPC calls will return a static port instead of a random port as I didn’t want to open a whole port range for. This is described in http://support.microsoft.com/kb/224196 and works as expected.
And now comes to frustrating part where I’ve lost a lot of time. Likely a quite easy to fix if you know what the issue is:
I’ve tried to connect to the Exchange with Outlook on from the domain’s test user’s client machine. Unsuccessfully. Btw. What I want to achieve is a native Outlook connection (RPC based) and not RPC over HTTP or anything else. I know that Outlook asks the Exchange server for certain ports via RPC calls, so I was going to adjust these ports on the Exchange to be static (same reason as above: firewall) as described in http://social.technet.microsoft.com/wiki/contents/articles/864.configure-static-rpc-ports-on-an-exchange-2010-client-access-server.aspx and similar articles. Opened the ports on the firewall and tried again to connect with Outlook. Again, unsuccessfully. I disabled the firewall to check if there is anything wrong with the configuration and to ensure that the firewall isn’t the problem. Didn’t work as well. So I started Wireshark to see what is going on. Outlook asks the Exchange server via RPC for certain ports, but Exchange is unable to provide these ports (the response contains an error code instead of a port). So I took a look at the Exchange server again and realized that the MS Exchange RPC Service is running (port 6001) but the ports defined through the registry aren’t (as you would see in the last screenshot of the link I’ve provided). So one might think now that I probably have messed up the registry or something else. Well, very unlikely as I’ve installed the whole AD and Exchange a second time without reconfiguring ports and the behavior was exactly the same (Exchange didn’t return the RPC ports).
So I googled a bit and found the following article:
http://blogs.technet.com/b/mspfe/archive/2011/04/12/troubleshooting_2d00_microsoft_2d00_exchange_2d00_outlook_2d00_connectivity_2d00_issues.aspx
In chapter “2.2 – Verifying Exchange Client Access Ports” is a sample excerpt of an RPCDump output. I ran the same command on my Exchange server and had a slightly different result, which I believe shows what the issue is. Unfortunately, I’ve no idea what exactly the issue is and how to fix it.
Example from the technet.com article:
LAB-E2K10-CSHT[63534] [1544f5e0-613c-11d1-93df-00c04fd7bd09] Microsoft Exchange RFR Interface :YES
I searched the same entry in the RPCDump output on my server. The difference was:
Microsoft Exchange RFR Interface :NO
Same applied to the other entries. So I believe there’s an issue with the RPC binding for the Exchange on that server. Unfortunately, I haven’t found a solution on the Internet so far. And most likely, this isn’t a big issue and I’ve simply forgotten a certain configuration step.
Any suggestions what the issue is and how to fix it? Thanks a lot!
EDIT: Exchange 2013 SP1 (&CU5) do not support plain RPC anymore, which is likely the reason for this behavior. I’m going to reconfigure the server and check if the issue is solved by using RPC/MAPI over HTTP.