Create a policy for a high availability group

Every high availability group has to have an associated policy. This policy determines which members of a high availability group to put in the active state.

We should know:

• The name of the core group to associate with the new policy.

• The name of the high availability group that we want this policy to control.

• The function, such as transaction log recovery or messaging engine, associated with this high availability group.

• The policy types, such as One of N or Static, that this function supports.

• The type of policy we want to create.

• The policy settings, such as failback, and preferred servers only, to configure for this policy.

The product includes default policies that are already associated with the high availability groups some of the product components use. If these default policies do not meet the requirements of the installation, IBM recommends that we create a new policy instead of changing one of the default policies. The creation of new policies provides you with the capability to tailor the policy settings to the installations requirements while giving us the option to revert back to the default policy.

Tasks

1. In the administrative console, click...
   Servers > Core Groups > Core group settings > core_group_name > Policies > New

   

2. Select the new policy we want in effect for a specific high availability group. If defining a new policy, the policy options are:

   All active policy	All of the group members are activated.
   M of N policy	M group members are activated. The number represented by M is defined as part of the policy details.
   No operation policy	No group members are activated.
   One of N policy	Only one group member is activated.
   Static policy	The active members of a group are statically configured.

   For messaging engines in service integration, only the Static, One of N, and No operation policies apply.

3. Click Next.

4. Specify a name for the policy in the Name field. The name must be unique within the scope of the core group. Make the name meaningful to other administrators.

5. Optional: Specify a Description of the policy in the description field. This description might include the name of the associated core group.

6. Specify a value for the Is alive timer field, if the default value is too long or too short a time period. This value determines how frequently the high availability manager checks the health of the high availability group members. The default is 0 seconds.

   • If we specify -1 (minus 1), the Is alive timer is disabled.

   • If we specify 0 (zero), the value that is specified for the Is alive timer at the core group services level is used for high availability groups associated with this policy.

   • If we specify an integer between 1 and 2147483647, inclusive, this value is used for the high availability groups associated with this policy.

7. Make sure the Quorum field is not selected. We should not enable Quorum unless we are explicitly instructed to do so in the documentation for some other product.

8. Select the Failback field to have the high availability manager make the most preferred member the currently active member whenever this action is possible. This option is available only for M of N and One of N policies.

9. Select the Preferred servers only field if we want the high availability manager to only activate group members on servers contained in the Preferred servers list. This option is available only for M of N and One of N policies. If we select this option, configure a list of preferred servers. A description of how to set up this list is provided in a later optional step.

10. Specify the number of group members that we want active in the Number of active members field. This option is available only for an M of N policy.

11. Click Apply and then select Match criteria.

12. On the next panel, click New and then configure the match criterion for this policy.

    1. In the Name field, specify the name of one of the name-value pairs contained in the name of the high availability group to associate with this policy.

    2. In the Value field, specify the value of the name-value pair we specified in the Name field.

    3. Optional: In the Description field, add a description of this match criterion. For example, we might specify First attribute to indicate that this name-value pair matches the first attribute contained in the group name.

    4. Click OK.

    5. Repeat these steps for each additional attribute we want to include as part of our match criterion.

    We should set the match criterion for a new policy to two or more of the name-value attributes contained in the name of the high availability group to ensure that this policy is used instead of one of the product default policies.

13. In the Additional Properties section, select the Static group servers field to configure the list of servers that we want activated. This option is available only for Static policies. Click Add to move core group servers into the list of Static group servers, and then Click OK after completing the list.

14. Optional: Under Additional Properties, select Preferred servers and select the preferred servers for this policy. This option is available only if we selected the Preferred servers only field for M of N and One of N policies. If we do not set up this list, no group members are activated.

    Click Add to move core group servers into the list of preferred servers.

    Select specific servers in the list and click Move up and Move down to adjust the order of the servers within the list. Make sure that the most preferred server is at the beginning of the list and the least preferred server is at the end of the list.

    After completing the preferred servers list, click OK.

    Use caution when selecting preferred servers. The product cannot detect if we select an inappropriate server as a preferred server. For example, if the policy affects a messaging engine or transaction service, only select preferred servers from the messaging engine cluster. Similarly, if the policy affects a transaction service, only select preferred servers from the transaction service cluster.

15. Click OK and then click Review.

16. Select Synchronize changes with nodes> Save.

The new policy goes into effect after it is saved and synchronized. We do not have to stop and restart the affected application servers.

What to do next

We can change the Failback and Preferred servers only options for this policy without stopping and restarting the affected application servers.

We can create or update the list of preferred servers that for this policy without stopping and restarting the affected application servers.

Subtopics

• Core group policies
  
• Core group policy settings
  
• New core group policy definition
  
• Preferred servers
  
• Match criteria collection
  
• Match criteria settings
  
• Static group servers collection
  

Related:

High availability groups

High availability group policy modification guidelines

High availability group policies

High availability manager

Select the policy for a high availability group

Configure a core group policy for messaging engines