WebSphere Application Server V8.5 Administration and Configuration Guide for the Full Profile

674 WebSphere Application Server V8.5 Administration and Configuration Guide for the Full Profile

4. Click Show servers to display core group servers that are hosting active high-availability

group members. To display high-availability groups, specify the name=value pairs or * in

the Group name properties field, and click Show groups, as shown in Figure 21-39.

Figure 18-39 Group name properties

18.3 Failover and failback

Failover is one of the techniques of fault tolerance. It is used to make the system to some

degree survive faults and failures.

In this section, we discuss WebSphere Application Server for z/OS failover capabilities.

18.3.1 High availability and failover of singletons

Singleton failover is a cluster-based service. High availability manager has to be enabled on

all the members. Failover and subsequent failback depend on the options selected in the high

availability service policy. The Preferred servers list option is provided to allow the service to

run exclusively on the servers in the list in a preferred order. External resources and any other

dependencies must be available to all the members that are to run the service in case of a

failure of the primary. External clustering software can be used to complement the failover

and failback processing.

WebSphere Application Server provides failover and recovery for the following singleton

services:

 Transaction service

 Service integration bus

Transaction service

Transaction service peer recovery processing can only take place between members of the

same server cluster. The Transaction Manager supports three different high-availability

policies to achieve the recovery of transaction logs in a highly-available manner:

 One of N policy

This policy is the default style of peer recovery initiation. If an application server fails, the

high availability manager selects another server to perform peer recovery processing on

behalf of the failed server.

Chapter 18. Clustering and high availability 675

 Static policy

This style of peer recovery must be explicitly configured. If an application server fails, the

operator can use the Administrative Console to select another server to perform the

recovery processing.

 No Operation policy

This style of peer recovery must be explicitly configured. It indicates that external

clustering software is monitoring the Transaction Manager and will fail over to an

application server that is configured by the external clustering software to perform the

recovery processing.

For more information about transactional high availability, go to the following website:

http://pic.dhe.ibm.com/infocenter/wasinfo/v8r5/index.jsp?topic=%2Fcom.ibm.websp

here.zseries.doc%2Fae%2Fcjta_trans_ha.htmll

To enable transaction service log recovery, complete the following steps:

1. Click Servers  Server types  WebSphere application servers.

2. Select your desired application server, as shown in Figure 21-40.

Figure 18-40 Select the application server

3. Under the Container Settings section, expand Container Services and then click

Transaction service, as shown in Figure 21-41.

Figure 18-41 Transaction service

676 WebSphere Application Server V8.5 Administration and Configuration Guide for the Full Profile

4. Specify the Transaction log directory value, as shown in Figure 21-42. The directory

specified must be unique in the cluster and accessible to all the cluster members. It must

be allocated on a fast disk.

Figure 18-42 Specify the Transaction log directory

5. Click Apply.

6. Click Review, click Synchronize changes with nodes, and click Save. Alternatively, click

System administration  Nodes, and click Synchronize with the appropriate node or

nodes selected, or enable the Synchronize changes with Nodes option in System

administration  Console Preferences.

7. Repeat steps 2 on page 181 to 6 for all the cluster members.

8. Click Servers  Clusters  WebSphere application server clusters.

9. Select your desired application server cluster, as shown in Figure 21-43.

Figure 18-43 Select application server cluster

Chapter 18. Clustering and high availability 677

10.Enable the Enable failover of transaction log recovery option, as shown in

Figure 21-44. Click Apply.

Figure 18-44 Enable failover

11.Click Review, click Synchronize changes with nodes, and click Save. Alternatively, click

System administration  Nodes, and click Synchronize with the appropriate node or

nodes selected, or enable the Synchronize changes with Nodes option in System

administration  Console Preferences.

12.Restart the affected servers by clicking Servers  Server Types  WebSphere

application servers.

Service integration bus

For bus messaging engines, use a policy type of One of N. Thus, although the messaging

engine can be defined on every server in the cluster, the high availability manager ensures

that it is active only on one of the servers in the group and will always be active on one of the

servers as long as one is available.

For more information about making the SIBus highly available, refer to the following website:

http://pic.dhe.ibm.com/infocenter/wasinfo/v8r5/index.jsp?topic=%2Fcom.ibm.webspher

e.zseries.doc%2Fae%2Fcjt0010_.html

To configure a service integration bus (SIBus) with data store persistence:

1. Click Service integration  Buses.