Planning for HyperSwap for PowerHA SystemMirror

Before you can install HyperSwap® for PowerHA® SystemMirror®, you must verify that your environment is working correctly.

To correctly plan for a HyperSwap configuration, review the following information:
  • Storage level Peer-to-Peer Remote Copy (PPRC) relationships and PPRC paths must be defined before you configure HyperSwap for PowerHA SystemMirror or In-band PPRC Metro Mirror for PowerHA SystemMirror.
  • HyperSwap for PowerHA SystemMirror is only supported on IBM® DS8800, or later, storage systems.
  • Concurrent workloads across sites, such as Oracle Real Application Clusters (RAC), and concurrent resource groups are supported in stretched clusters and linked clusters that are using HyperSwap enabled mirror groups. PowerHA SystemMirror does not support concurrent resource groups for other PPRC replications solutions.
  • You can configure HyperSwap enabled mirror groups for automatic resynchronization when a replication failure occurs. Error logs are created when the replication failure occurs, and the log file can be used to identify the cause of the failure.
  • To maintain consistency group schematics, suspended operations on a storage device must function on the entire logical subsystem (LSS). For example, if a single IBM DS8800 LSS contains PPRC volumes from more than one application and if one of the replication connections breaks, all PPRC paths are removed. If the applications are not managed by PowerHA SystemMirror, the PPRC paths must be manually re-created or refreshed after the replication connection is reestablished.
  • Refresh a mirror group in the cluster after any storage-level PPRC configuration changes or if a PPRC path is broken. When you refresh a mirror group, you are re-creating the paths for the mirror group and refreshing any existing configuration changes. Do not run any HyperSwap functions when you are refreshing a mirror group.
  • To add a node to a mirror group, you must perform configuration operations from a node where all disks are accessible.
  • Metro Mirror (in-band) functions, including HyperSwap, are supported in Virtual I/O Server (VIOS) configurations by the N-Port ID Virtualization (NPIV) method of disk management.
  • Metro Mirror (in-band) functions, including HyperSwap, are not supported by the virtual SCSI (VSCSI) method of disk management.
  • To use Live Partition Mobility (LPM), you must bring the resource group that contains the mirror group into an unmanaged state by using the C-SPOC utility to stop cluster services with the Unmanage Resource Groups option. After you complete the LPM configuration process, you must bring the resource group back online by using SMIT. This process brings all mirror groups and resource groups back online.
  • Disk replication relationships must adhere to a one-to-one relationship between the underlying LSS. An LSS that is already part of a mirror group, cannot be part of another mirror group.
  • Repository disks require that you specify an alternate disk or a disk that is not configured to use the HyperSwap function when you set HyperSwap property to Disable.
  • SCSI reservations are not supported for devices that use the HyperSwap function.
  • You must verify and synchronize the cluster when you change the cluster configuration. If you change the mirror group configuration while cluster services are active (DARE), those changes might be interpreted as failures, which result in unwanted cluster events. You must disable the HyperSwap function before you change any settings in an active cluster environment.
  • A node in an unmanaged state does not respond or perform any swap or resync functions. However, the I/O to the node continues. Thus, having a node in an unmanaged state is ideal for performing LPM on the node. After you complete the LPM process, you can move the node from unmanaged to online state by using SMIT. If an unplanned event occurs while the node is in an unmanaged state, the node is moved to a halt state.

Fileset requirements

To run HyperSwap for PowerHA SystemMirror, you must install the following filesets on all nodes in the cluster:
Table 1. HyperSwap fileset requirements for PowerHA SystemMirror
Fileset name Minimum fileset level for AIX® Version 7.1 Minimum fileset level for AIX Version 6.1
cluster.es.genxd.cmds 7.1.2.0 6.1.8.0
devices.common.IBM.storfwork.rte 7.1.2.0 6.1.8.0
devices.common.IBM.mpio.rte 7.1.2.0 6.1.8.0
devices.fcp.disk.rte 7.1.2.0 6.1.8.0