Pdisks

The IBM Storage Scale RAID pdisk is an abstraction of a physical disk. A pdisk corresponds to exactly one physical disk and belongs to exactly one declustered array within exactly one recovery group. Before discussing how declustered arrays collect pdisks into groups, it will be useful to describe the characteristics of pdisks.

A recovery group can contain a maximum of 512 pdisks. A declustered array within a recovery group can contain a maximum of 512 pdisks. The name of a pdisk must be unique within a recovery group; that is, two recovery groups can each contain a pdisk named disk10, but a recovery group cannot contain two pdisks named disk10, even if they are in different declustered arrays.

A pdisk is usually created using the mmcrrecoverygroup command, whereby it is assigned to a declustered array within a newly created recovery group. In unusual situations, pdisks can also be created and assigned to a declustered array of an existing recovery group by using the mmaddpdisk command.

To create a pdisk, a stanza must be supplied to the mmcrrecoverygroup or mmaddpdisk commands specifying the pdisk name, the declustered array name to which it is assigned, and a block device special file name for the entire physical disk as it is configured by the operating system on the active recovery group server. A sample pdisk creation stanza follows: 

%pdisk: pdiskName=c073d1
        device=/dev/hdisk192
        da=DA1
        nPathActive=2
        nPathTotal=4
Other stanza parameters might be present. For more information about pdisk stanza parameters, see Pdisk stanza format.

The device name for a pdisk must refer to the entirety of a single physical disk; pdisks should not be created using virtualized or software-based disks (for example, logical volumes, disk partitions, logical units from other RAID controllers, or network-attached disks). The exception to this rule are non-volatile RAM (NVRAM) volumes used for the log tip vdisk, which is described in Log vdisks. For a pdisk to be created successfully, the physical disk must be present and functional at the specified device name on the active server. The physical disk must also be present on the standby recovery group server, if one is configured. The physical disk block device special name on the standby server will almost certainly be different and will be discovered automatically by IBM Storage Scale.

The attributes of a pdisk include the physical disk's unique worldwide name (WWN), its field replaceable unit (FRU) code, and its physical location code. Pdisk attributes can be displayed using the mmlspdisk command; of particular interest here are the pdisk device paths and the pdisk states.

Pdisks that have failed and have been marked for replacement by the disk hospital are replaced using the mmchcarrier command. In unusual situations, pdisks can be added or deleted using the mmaddpdisk or mmdelpdisk commands. When deleted, either through replacement or the mmdelpdisk command, the pdisk abstraction will only cease to exist when all of the data it contained has been rebuilt onto spare space (even though the physical disk might have been removed from the system).

Pdisks are normally under the control of IBM Storage Scale RAID and the disk hospital. In some situations, however, the mmchpdisk command can be used to manipulate pdisks directly. For example, if a pdisk has to be removed temporarily to allow for hardware maintenance on other parts of the system, you can use the mmchpdisk --begin-service-drain command to drain the data before removing the pdisk. After bringing the pdisk back online, you can use the mmchpdisk --end-service-drain command to return the drained data to the pdisk.
Note: This process requires that there be sufficient spare space in the declustered array for the data that is to be drained. If the available spare space is insufficient, it can be increased with the mmchrecoverygroup command.