locked
OCS MSMQ using SNAT Load Balancer doesn't work (Archiving) RRS feed

  • Question

  • I have a problem whereby my enterprise pool using SNAT and outgoing packets have the new IP of the load balancer.  All of this works fine.  My SQL Archinging Server/Agent lie on one computer outside of this load balancer pool.  When I start front end services, archiving works fine for the first server  but the second one does not.

    Looking into the outbound queues I see that the Front End servers identify the SQL by the netbios name using DIRECT=OSTongue TiedERVNAME   however the SQL Archiving server's outbound shows DIRECT=TCP:[ip of load balancer]
    So the first connection holds this open and the second one can't connect leaving unacknowledged messages in the queue.

    Has anyone had a similar problem?  I've disabled NameValidation for OS but that didn't help.  So right now I can only use one server in my pool or else it won't archive half the messages.
    Thursday, August 14, 2008 5:02 PM

All replies

  • The OCS front end servers must see their backend services untranslated by NAT ... e.g. MSMQ, SQL, Active Directory. NAT only plays a role between the client->load balancer->front ends. The clients use a LB VIP address and certificate/DNS name, then the LB "sprays" sessions to the FE's.
    Thursday, January 15, 2009 12:32 AM