![]() |
Copyright © TIBCO Software Inc. All Rights Reserved |
The typical activity switches for the TIBCO Messaging Appliance P-7500 redundancy facility are:
Note: Many other activity switch scenarios are possible, but they are beyond the scope of this chapter. For more information, contact TIBCO.When a TIBCO Messaging Appliance P-7500 system is described as failed, the system is not able to participate in message routing and forwarding for a period of time. This might be a transient outage, which is relatively short-lived, or a longer-term outage, and includes both planned and unplanned causes such as:
• physical link to the system is down (could be transient or longer-term, planned or unplanned)Figure 10 system has been placed in the standby state as a result of the network operator entering the release-activity Router Redundancy CONFIG command through the P-7500 Command Line Interface (CLI) (longer-term outage, planned) Active/Active Redundancy PairingIn an active/active network configuration, such as that illustrated in Figure 10, the following series of operations occur if one of the systems (in the following example P-7500 2) goes out of service for any reason:
2. P-7500 1 becomes active for the IP address 192.168.171.133/19. When this occurs, the backup virtual router for P-7500 1 goes from an idle state to an active state.
3. P-7500 1 increases the priority of its VRRP advertisements for the IP address 192.168.171.133/19 to Assert-Activity. It also sends out a gratuitous ARP to let other network devices know that it is taking over the IP address. At this point, network traffic to the IP address 192.168.171.133/19 flows to P-7500 1.
4. As this redundant pair is interacting with RV services on other networks, the backup RVGD and the backup eth2 interface (eth2:2) for P-7500 1 become active.
5. RV messages directed to the IP address 192.168.160.30/19 now flow to eth2:2 on P-7500 1. P-7500 also sends out a gratuitous ARP through eth2:2 to let other network devices know that it is taking over 192.168.160.30/19. As the backup for eth2:1 on P-7500 2, eth2:2 on P-7500 1 uses the same IP address.
6. Clients that had already established TCP connections to P-7500 2 receive error messages back from P-7500 1 because the TCP connections are not present on P-7500 1. This causes the clients to tear down the existing TCP sessions and immediately establish new TCP sessions. These new connections are accepted by P-7500 1.
7. P-7500 1 accepts subscription updates and published messages from its own clients and those that typically have service from P-7500 2.Figure 11 illustrates a common recovery activity switch scenario, where system P-7500 2 has failed and its mate system, P-7500 1, is acting on behalf of P-7500 2.Figure 11Failure Activity Switch Behavior
1. As P-7500 2 starts up, its primary virtual router uses a VRRP advertisement of Primary-Reconcile, and its backup virtual router uses a VRRP advertisement of Backup-Reconcile. These advertisements indicate that the virtual routers are initializing, but are not yet ready to take on client activity.
2. Once P-7500 2 has fully started, its primary virtual router increases the priority of its VRRP advertisements to Assert-Activity. This priority level causes the backup virtual router of its mate system to relinquish activity. P-7500 2 then sends a gratuitous ARP to let the other network services know it is taking over the IP address 192.168.171.133/19 for client connections.
3. P-7500 1 disconnects any TCP sessions that it had established for clients receiving service from IP address 192.168.171.133/19 and for RV messages flowing through eth2:2 on IP address 192.168.160.30/19.
4. The primary virtual router for P-7500 1 advertises a local priority level of Active, and the backup virtual router advertises a local priority level of Backup.
Note: Through the auto-revert Router Redundancy CONFIG command, the network administrator can configure whether the standby system in a redundancy pair should automatically revert back at this point to the primary when the primary comes back online after a service outage. Refer to “Enabling the Backup System to Automatically Revert Activity” on page 167 for details.
5. At this point, the redundancy state of P-7500 2 causes eth2:2 and the backup RVGD to become idle.
6. P-7500 2 accepts new TCP connections to IP address 192.168.171.133/19 and delivers messages to, and accepts messages from, clients using this IP address.
![]() |
Copyright © TIBCO Software Inc. All Rights Reserved |