Performing LPM operation on a cluster node and modifying LPM variables

Edit online

PowerHA® SystemMirror® automates some of the Live Partition Mobility (LPM) steps by registering a script with the LPM framework.

PowerHA SystemMirror listens to LPM events and automates steps in PowerHA SystemMirror to handle the LPAR freeze that might occur during the LPM process. As part of the automation, PowerHA SystemMirror provides a few variables that can be changed based on the requirements for your environment.

You can change the following LPM variables in PowerHA SystemMirror that provide LPM automation:
Variable Description Usage options
  • Command line: HEARTBEAT_FREQUENCY_DURING_LPM
  • SMIT interface: Node Failure Detection Timeout during LPM
 You can use this variable to increase in seconds the node failure detection time (in seconds) across the cluster, during the LPM process. Ensure that the LPM process that is related to LPAR freeze times does not exceed the node failure detection time. The LPAR freeze time is different and it is a smaller value as compared to the entire duration needed for the LPM process. If you do not specify a value for this variable, then the maximum value of 600 seconds is used during the LPM process.
  • Command line: clmgr modify cluster HEARTBEAT_FREQUENCY_DURING_LPM=<node_timeout>
  • SMIT interface: Custom Cluster Configuration > Cluster Nodes and Networks > Manage the Cluster > Cluster heartbeat settings and select the Node Failure Detection Timeout during LPM field.
  • Command line: LPM_POLICY
  • SMIT interface: LPM Node Policy
 You can use this variable to set the resource groups on the LPAR to be in an unmanage state or in a manage state during the LPM process. The default value for this variable is manage.
Note: Do not modify any applications during the LPM process. While the cluster is in an unmanage state, PowerHA SystemMirror does not monitor any applications. If an LPAR failure occurs during the LPM process, you can start the application on a standby node.

You can reduce the occurrence of an unwanted cluster event by disabling cluster services during the LPM process. To disable cluster services, specify unmanage for this variable.
  • Command line: clmgr modify cluster LPM_POLICY=unmanage
  • SMIT interface: Custom Cluster Configuration > Cluster Nodes and Networks > Manage the Cluster > Cluster heartbeat settings and select the LPM Node Policy field.
Note: You must verify and synchronize the cluster for any changes that you make to the LPM variables.
If you have not set the LPM_POLICY variable to unmanage, you can manually stop cluster services and perform the LPM operation on a cluster node by completing the following steps:
  1. From the command line, enter smitty cl_stop.
  2. In the SMIT interface, select PowerHA SystemMirror Services > Stop Cluster Services, and press Enter.
  3. Select the Select an Action on Resource Groups field and press F4.
  4. From the list, select Unmanage Resource Groups and press Enter to stop cluster services.
  5. Perform the LPM process.
  6. After the LPM process is complete, from the command line, enter smitty clstart.
  7. Select the value for the Manage Resource Groups field as Automatically and press Enter.