System configuration changes using the smart workload recovery feature include.
Cluster application configuration changes
Add/Remove/Modify Cluster Applications
Add/Remove/Modify Resources to Cluster Applications
Add/Remove Cluster Nodes
Change the configuration of an instance on a cluster node
Change settings of switching mechanism and resource monitoring mechanism
For instructions on modifying the cluster application configuration, see "Chapter 10 Configuration Change of Cluster Applications" in the "PRIMECLUSTER Installation and Administration Guide". However, you must do the following before stopping and after starting RMS.
Disables tags on the instance.
Use the AWS Management Console to set the value of the tag "fujitsu.pclswr.is_recovery_target" for the instance to false.
Disables CloudWatch Agent.
Disables monitoring of RMS by the CloudWatch agent and changes the setting to log monitoring only. For more information, see "A.3.1.2 Operation when RMS startup is stopped / RMS monitoring is disabled".
Enables CloudWatch Agent.
Enable RMS monitoring in CloudWatch Agent configuration file. For more information, see "A.3.1.2 Operation when RMS startup is stopped / RMS monitoring is disabled".
Enables tags on the instance.
Use the AWS Management Console to set the value of the tag "fujitsu.pclswr.is_recovery_target" for the instance to true.
If you want to add or remove cluster nodes, see "A.5.4 Procedure for Adding Cluster Node" and "A.5.5 Procedure for Removing Cluster Node".
For instructions on changing the configuration of instances on a cluster node, see the official AWS documentation.
For instructions on how to change the configuration of switcher and resource monitor, see the official AWS documentation. Note that operation is not guaranteed if the setting is changed to a value other than that described in "A.2.4 Construction of Switcher and Resource monitor".
Note
If you made configuration changes to your instance (such as changing the instance type), you must recreate the AMI. For instructions about creating an AMI in the operating instance, see Create an AMI "A.5.8 Procedure for Getting AMI in Operation".
Always perform a switch test after a configuration change to detect misconfiguration settings.