Ask a questionAsk a question
 

QuestionSrv 2008 RDP breaks

  • Tuesday, March 03, 2009 7:13 PMNathan Hammes Users MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     

    Here's the scenario - 2 node 2008 SCC.  Everything is working with a single NIC active on each node (mixed communication). When I plug in the second NIC (heartbeat - which is configured with a private address) on each host, I can RDP to one host but RDP to the other does not work anymore.  The server is still online - services running, responding to ICMP, WMI access, etc - just can't RDP into it.  I have checked and changed NIC binding orders, checked for new updates, rebooted the server, checked Windows services - nothing looks wrong and nothing seems to resolve the issue.  Except unplugging the second NIC on the "affected" server - if I unplug the second NIC, I can now RDP into the server again.

    Obvioulsy I am concerned and do not want to install Exchange on the cluster yet until the heartbeat is connected and working properly along with RDP to the server...

    Any ideas would be greatly appreciated.  Thanks!

    Nate


    Nate

All Replies

  • Wednesday, March 11, 2009 7:12 PMRobert Smit [MVP]MVPUsers MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     
    Hi Nathan,

    Is the IP config correct on the private there is only a IP and a mask and no DNS , WINS, No register in DNS, NO entries in the DNS ?
    This should only be on the lan nic. check your binding order.

    This must be a network setting item , should all work in the basic.

    let us know the IP settings check the netbios ,AD-DNS default gateway.
    Greetings, Robert Smit http://fiberman.spaces.live.com/
  • Thursday, March 12, 2009 12:01 AMMSD5 Users MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     
    Do you have any static routes configured on the server?  Try running route print and see if there any persistent routes.
  • Thursday, June 04, 2009 6:37 PMDor Nosreap Users MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     

    We're experiencing the same issue.  Did anyone identify a resolution?

  • Wednesday, July 01, 2009 9:33 AMMplant Users MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     
    Ditto, exactly the same problem here. Win Server 2008 EE with 4 active NIC connections:

    2 x iSCSI (no gateway, DNS, WINS etc)
    1 x Cluster hearbeat (no gateway, DNS, WINS etc)
    1 x LAN connection for RDP etc (gateway, DNS, the usual)

    As soon as either iSCSI port is enabled, RDP disconnects. the LAN connection is set at the highest in the bind order, and TS Configuration has been set to bind RDP-Tcp to the LAN connection only.

    A solution greatly appreciated!
  • Thursday, July 02, 2009 9:59 AMEdwin vMierloMVP, ModeratorUsers MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     
    post "ipconfig /all" from both nodes to this forum please

    thanks,
    Edwin.
  • Tuesday, July 07, 2009 11:53 PMG_Rowley Users MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     
    I am also having the same problem but I also lose internet connectivity. I have a 2 node failover cluster using iSCSI. I can still access the nodes through another server that shares a private network. The public NIC's are obtaining IP addresses (this is still in testing phase) but I have hardset the DNS settings. I tried turning off the Large Send Offload on the public NIC and was able to get access for a short time without going through the private network. But now it no longer works.

    Does anyone have any ideas?