none
RPC error when configuring Exchange 2013 servers in 2nd site

    Question

  • Hello. I'm running into an error when trying to configure any of my Exchange 2013 servers in my 2nd AD site. To get into the loop of what my server structure looks like, please check below:

    Site 1 servers:

    DC1 - Domain Controller
    DC2 - Domain Controller
    CAS1 - CAS server
    CAS2 - CAS server
    MBX1 - Mailbox server
    MBX2 - Mailbox server
    MATHAFTMG - TMG server

    Site 2 servers:

    CCCDC1 - Domain Controller
    CCCDC2 - Domain Controller
    CCCCAS1 - CAS server
    CCCCAS2 - CAS server
    CCCMBX1 - MBX server
    CCCMBX2 - MBX server
    CCCTMG - TMG server

    Currently I have a site-to-site vpn connection between site 1 and site 2 TMG servers via Internet connection; I can access the servers of the other site perfectly (whether I am in Site 1 or Site 2).

    All user mailboxes are currently in Site 1 MBX servers; when users are in Site 2, they connect to the CAS servers in Site 1 to access their mailboxes.

    Many users will stay permanently in Site 2, so it makes sense to have Exchange servers in Site 2 to provide faster access to mailboxes. I created the Site 2 domain controllers, and made sure AD replication is working; and it is. I then added the MBX servers and CAS servers in Site 2 in this order: CCCMBX1, then CCCCAS1, then CCCMBX2, then CCCCAS2.

    All Exchange servers in Site 2 installed beautifully. But then I tried to access the servers via ECP to proceed with the configuration. In ECP, I click on the server link, and all Exchange servers in both sites appear. If I try to configure the virtual directories of Site 1 CAS servers, no problem. But when I try to configure virtual directories of Site 2 CAS servers, I get this error message: The task wasn't able to connect to IIS on the server 'CCCCAS1.domain.com'. Make sure that the server exists and can be reached from this computer: The RPC server is unavailable.

    The virtual directories issue is just an example. Same thing happens if I try to configure Outlook Anywhere for Site 2 CAS servers.

    Users connect to Site 1 CAS servers via mail.domain.com. I have the A record mail.domain.com pointing to the IP address of CAS1 server, and another A record mail.domain.com pointing to the IP address of CAS2 server. Not the best load balancing going on here, but it works great with Exchange 2013.

    From mail.domain.com I can access OWA and ECP internally and externally; no problems there. From ECP I can access and configure any Site 1 Exchange 2013 servers.

    The only problem is when I access ECP to configure the Site 2 Exchange 2013 servers, I get the same error message: The task wasn't able to connect to IIS on the server '<server name>.domain.com'. Make sure that the server exists and can be reached from this computer: The RPC server is unavailable.

    Even if I try to access a Site 2 Exchange 2013 server via https://localhost/ecp to configure it, it get the same error message.

    I updated all Exchange 2013 servers in both sites to CU2 v2 and rebooted the servers in the proper order; problem still there.

    Any clue what might the problem be?

    Thank you!

    Saturday, September 07, 2013 7:11 AM

Answers

  • Make sure that all firewalls and TMG servers or whatever else there is does not interfere with all protocols required between the Exchange servers.

    Ed Crowley MVP "There are seldom good technological solutions to behavioral problems."

    Monday, September 09, 2013 12:32 AM
  • Hi,

    Let’s try the following test firstly.

    1.Create a new admin user in site 2.
    2.Give it the same permission with the user as the user1.
    3.Log on the new user on the CAS in Site 2.
    4.Access ECP through the account and check if everything is well.

    And we can check application logs on the CAS server in site 2 and confirm if there is some error.

    If you have any question, please feel free to let me know.

    Thanks,

    Angela Shi

     


    Tuesday, September 10, 2013 12:56 AM

All replies

  • Hello. I'm running into an error when trying to configure any of my Exchange 2013 servers in my 2nd AD site. To get into the loop of what my server structure looks like, please check below:

    Site 1 servers:

    DC1 - Domain Controller
    DC2 - Domain Controller
    CAS1 - CAS server
    CAS2 - CAS server
    MBX1 - Mailbox server
    MBX2 - Mailbox server
    MATHAFTMG - TMG server

    Site 2 servers:

    CCCDC1 - Domain Controller
    CCCDC2 - Domain Controller
    CCCCAS1 - CAS server
    CCCCAS2 - CAS server
    CCCMBX1 - MBX server
    CCCMBX2 - MBX server
    CCCTMG - TMG server

    Currently I have a site-to-site vpn connection between site 1 and site 2 TMG servers via Internet connection; I can access the servers of the other site perfectly (whether I am in Site 1 or Site 2).

    All user mailboxes are currently in Site 1 MBX servers; when users are in Site 2, they connect to the CAS servers in Site 1 to access their mailboxes.

    Many users will stay permanently in Site 2, so it makes sense to have Exchange servers in Site 2 to provide faster access to mailboxes. I created the Site 2 domain controllers, and made sure AD replication is working; and it is. I then added the MBX servers and CAS servers in Site 2 in this order: CCCMBX1, then CCCCAS1, then CCCMBX2, then CCCCAS2.

    All Exchange servers in Site 2 installed beautifully. But then I tried to access the servers via ECP to proceed with the configuration. In ECP, I click on the server link, and all Exchange servers in both sites appear. If I try to configure the virtual directories of Site 1 CAS servers, no problem. But when I try to configure virtual directories of Site 2 CAS servers, I get this error message: The task wasn't able to connect to IIS on the server 'CCCCAS1.domain.com'. Make sure that the server exists and can be reached from this computer: The RPC server is unavailable.

    The virtual directories issue is just an example. Same thing happens if I try to configure Outlook Anywhere for Site 2 CAS servers.

    Users connect to Site 1 CAS servers via mail.domain.com. I have the A record mail.domain.com pointing to the IP address of CAS1 server, and another A record mail.domain.com pointing to the IP address of CAS2 server. Not the best load balancing going on here, but it works great with Exchange 2013.

    From mail.domain.com I can access OWA and ECP internally and externally; no problems there. From ECP I can access and configure any Site 1 Exchange 2013 servers.

    The only problem is when I access ECP to configure the Site 2 Exchange 2013 servers, I get the same error message: The task wasn't able to connect to IIS on the server '<server name>.domain.com'. Make sure that the server exists and can be reached from this computer: The RPC server is unavailable.

    Even if I try to access a Site 2 Exchange 2013 server via https://localhost/ecp to configure it, it get the same error message.

    I updated all Exchange 2013 servers in both sites to CU2 v2 and rebooted the servers in the proper order; problem still there.

    Any clue what might the problem be?

    Thank you!

    Saturday, September 07, 2013 6:39 AM
  • The first thing to check is that there are no firewalls between the sites blocking RPC.

    Ed Crowley MVP "There are seldom good technological solutions to behavioral problems."

    Saturday, September 07, 2013 5:49 PM
  • Responded to your post in the correct (Exchange 2013) forum.

    Ed Crowley MVP "There are seldom good technological solutions to behavioral problems."

    Saturday, September 07, 2013 5:50 PM
  • You mean on the TMG servers in each site? I setup the site-to-site vpn and all servers in site 1 can ping all servers in site 2 and vice versa. When I setup the TMG servers, I created a firewall rule on site 1 TMG to allow all outbound traffic from site 1 to site 2, and a firewall rule on site 2 TMG to allow all outbound traffic from site 2 to site 1.The exchange servers in each site can ping every other server in the other site.

    I tried running Get-OutlookAnywhere from CAS1 server in Site 1, I get information about CAS1 and CAS2 (both are in Site 1), but it gives me same error about CCCCAS1 and CCCCAS2 (both are in Site 2).

    I also tried running Get-OutlookAnywhere from CCCCAS1 server in Site 2, I get information about CCCCAS1 and CCCCAS2 (which are in Site 2), but it gives me same error about CAS1 and CAS2 (both are in Site 1).

    So it seems servers of a certain site can only read information of other servers in the same site, but not in the other site (Exchange power shell giving IISNotReachableException).

    Maybe my ISP is blocking RPC?

    Saturday, September 07, 2013 6:36 PM
  • Make sure that all firewalls and TMG servers or whatever else there is does not interfere with all protocols required between the Exchange servers.

    Ed Crowley MVP "There are seldom good technological solutions to behavioral problems."

    Monday, September 09, 2013 12:32 AM
  • Hi,

    Let’s try the following test firstly.

    1.Create a new admin user in site 2.
    2.Give it the same permission with the user as the user1.
    3.Log on the new user on the CAS in Site 2.
    4.Access ECP through the account and check if everything is well.

    And we can check application logs on the CAS server in site 2 and confirm if there is some error.

    If you have any question, please feel free to let me know.

    Thanks,

    Angela Shi

     


    Tuesday, September 10, 2013 12:56 AM
  • On each site's TMG server, I added 2 rules. First rule, open several ports (including Exchange RPC, port 135) between Site 1 servers (From) and Site 2 servers (Destination). Second rule, open several ports (including Exchange RPC, port 135) between Site 2 servers (From) and Site 1 servers (Destination). I tried accessing mail.domain.com to edit Site 2 exchange servers, and it seems that communication went through. I have not seen the RPC error anymore, but retrieving the server information is just slow. I suspected that my Internet connection might be slow, but when I try to access the Site 1 servers from any location (from home for example), the servers refresh very fast. Accessing Site 2 servers now work, but the refresh time is very slow. Any ideas how to improve refresh time? Thanks!
    Saturday, September 14, 2013 10:49 AM
  • Same issue I am also facing in my secondary site error is also same can you explain the ports which required between exchange 2013 servers.
    Thursday, May 22, 2014 5:16 AM