This alert represents the total end-to-end processing time for a client RPC request after the request reaches Microsoft Exchange Server 2010 Client Access Server (CAS) middletier. If the RPC requests take a long time to process, you may experience one or more of the following symptoms in the Exchange organization:

• MAPI clients such as Outlook hangs, and the following notification appears on the client computers:
  
  

  Outlook is retrieving data from Microsoft Exchange Server <ServerName>

• Devices such as BlackBerry devices may drop their connections to the server or experience timeout errors when they send and receive messages. In this scenario, information that resembles the following may be logged on the BlackBerry Enterprise Server (BES):
  
  

  MAPI call failed. Error 'Network problems are preventing connection to the Microsoft Exchange Server computer.

One of the client access improvements that Microsoft Exchange Server 2010 includes is to support MAPI connections to the Client Access server role. This change required creating the Microsoft Exchange RPC Client Access service on the Client Access server.

The RPC Client Access service establishes an RPC endpoint on a Client Access server to which MAPI clients such as Microsoft Office Outlook can connect.

Note:
In earlier versions of Exchange, MAPI clients create an RPC connection to servers running the Mailbox server role.

In Exchange 2010, the Client Access servers communicate with Mailbox servers to give MAPI clients access to Exchange mailboxes. This new functionality gives Exchange 2010 the following advantages over earlier versions of Exchange:

• It isolates the information store from direct client access.
  
  
• It provides greater integration with the Exchange 2010 high availability strategy.
  
  
• It provides a more reliable system in the event that an issue affects the information store.
  
  

For more information, see Understanding RPC Client Access.

To provide greater scalability and redundancy for MAPI clients, CAS servers may be configured in a Network Load Balancing middletier array that is connected to the mailbox servers.

The RPC Average Latency (End To End) counter represents the averaged latency for the RpcClientAccess service to process the RPC requests from MAPI on the Middle-Tier (MOMT) to the mailbox servers. This latency is the average number of milliseconds for the last 1,024 packets.

This alert may be triggered when any of the following conditions are true:

• A processing delay exists within MAPI on the middletier.
  
  
• A high network latency exists between the CAS servers and mailbox servers.
  
  
• The information store takes a long time to process requests.
  
  

To resolve this problem, do one or more of the following:

• Review the Application log and System log on your Exchange 2010 servers for related events. For example, events that occur immediately before and after this event may provide more information about the root cause of this error.
  
  
• To review detailed information about the cause of this alert, use the Operations Console in Operations Manager. For more information, see the "Introduction" section in this topic.
  
  
• Examine networking devices that are positioned between the CAS servers and mailbox servers. Specifically, look for the following issues:
  
  
  • Determine whether all network devices and network adapters are functioning correctly.
    
    
  • Determine whether any network devices have an idle timeout value that is set too low. To work around an idle-timeout issue, you may want to change the KeepAlive behavior on the CAS server or on a mailbox server. For more information, see Microsoft Knowledge Base article 314053, TCP/IP and NBT configuration parameters for Windows XP.
    
    
  • Verify that the MTU size limit that is set on any network device is not lower than the default settings in Windows.
    
    
• Increase the number of Exchange 2010 CAS servers in the middletier NLB cluster.
  
  
• Examine performance counter information on the mailbox servers to determine whether one or more servers are overloaded.
  
  
• Resolve your issue by using self-support options, assisted support options, and other resources. You can access these resources from the Exchange Server Solutions Center. From this page, click Self-Support Options in the navigation pane to use self-help options. Self-help options include searching the Microsoft Knowledge Base, posting a question at the Exchange Server forums, and others. Alternatively, in the navigation pane, you can click Assisted Support Options to contact a Microsoft support professional. Because your organization may have a specific procedure for directly contacting Microsoft Technical Support, be sure to review your organization's guidelines first.
  
  

If you are not already doing so, consider running the Exchange tools created to help you analyze and troubleshoot your Exchange environment. These tools can help make sure that your configuration aligns with Microsoft best practices. They can also help you identify and resolve performance issues, improve mail flow, and better manage disaster recovery scenarios. To run these tools, go to the Toolbox node of the Exchange Management Console. To learn more about these tools, see Managing Tools in the Toolbox.