Example: Unexpected switchover to primary server

Part of configuring a redundant server pair includes choosing a switchover option. One of the options is to continue using the secondary server, even when the primary becomes available again.
With this option, clients remain connected to the active server until the server is manually switched or until the active server fails. Upon reconnection, however, the health monitoring service decides which server in a redundant pair is the active one.
The service bases this decision upon which server has the most clients connected to it. Some computers may be using a data server behind the scenes, without displaying data to users. The health monitoring service is aware of these clients and may consider the primary server to be the active one, when the primary server is reconnected.
If the primary server is considered to be the active server, then clients switch over to it, even though the redundant server pair has been configured to continue using the secondary server.
Consider this example:
FTSP Linked Icons PNG File Template
Computer 1: Primary server
. Installed software:
FactoryTalk Services Platform
, hosting Network Directory Server.
FactoryTalk View SE
 and
FactoryTalk Linx
 are installed and configured with redundant server pairs on Computer 2.
Computer 2: Secondary server
. Installed software:
FactoryTalk Services Platform
.
FactoryTalk View SE
 and
FactoryTalk Linx
 are installed and configured with redundant server pairs on Computer 1.
Computer 3: Client computer
. Installed software:
FactoryTalk Services Platform
, hosting a Network Directory client;
FactoryTalk View SE
 client.
Switch over example
Assume that the redundant server pairs on computers 1 and 2 are configured to continue using the secondary server, even when the primary server becomes available again.
FTSP Linked Icons PNG File Template
  1. Computer 1, the primary server, is disconnected from the network. Computer 2 becomes the active server. Computer 3 switches over to the redundant
    FactoryTalk View SE
     and
    FactoryTalk Linx
     servers on Computer 2. Because the Network Directory Server resides only on Computer 1, Computer 3 uses its local directory cache until it reconnects to Computer 1.
  2. Computer 1 is reconnected to the network.
    Because the redundancy option is configured to remain connected to the secondary server, the expectation is that Computer 2 remains the active server and that a switch back to the primary server would not occur.
    However, when a server reconnects to the
    FactoryTalk Directory
    , the health monitoring service determines which of the two computers is servicing the most clients and the computer with the most clients is designated the active one. In this example, the health monitoring service finds that Computer 3 is a client of Computer 2. It also finds that an invisible
    FactoryTalk
     service has remained a client of Computer 1, because Computer 1 hosted the Network Directory Server. When two redundant servers have the same number of clients, the primary server is considered the active server.
  3. The health monitoring service considers Computer 1 to be the Active server and switches all client computers over to it.
FTSP Linked Icons PNG File Template
Provide Feedback
Have questions or feedback about this documentation? Please submit your feedback here.
Normal