• HP recommends that you create separate managed sets for source and destination D R groups.
When you conduct failover operations, you can per form the actions that correspond to the new
roles of the managed set.
• Source and d estination DR groups (from different DR group pairs) can be m embers of the
same managed set.
• The source and destination DR group in a DR group pair cannot be members of the same
managed set.
• If you plan to use DR group managed sets for failover operations, ensure the managed sets are
controlled by the same managem ent server at the time of failover.
Normalization
Normalizati
on is a replication background process which verifies, on a block-by-block basis, that data is
identical on a source virtual disk a nd its replica. When the source and its replica have identical data,
they are said to be normalize d or in a normalized state. The following t ypes of normalization can apply.
DR group normalization
DR group normalization is a process which verifies tha t data is identical on source virtual disks and
their remote copies.
Some actions should not be performed du ring remote copy normalization. When creating jobs, use wait
commands to ensu re that normalization is completed. See WaitDrGroupNormalization.
The disks in a DR group ca nnot become normalized while any of the following conditions are in effect:
• Remote replication is suspended. See suspension state.
• An intersite link is down.
• In standard asynchronous write mode, or synchronous write mode, its DR group log is logging or
merging. See write mode and DR group log.
• In enhanced asynchronous write mode, its DR group log is logging, merging, or contains any
transactions to be merged.
Mirrorclone normalization
Mirrorcl
one n ormalization is a process which verifies that da ta is identical on a source virtual disk and its
synchronized m irrorclone. See synchronized m irrorclones.
Some actions s hould not be performed during mirrorclone normalization. When creating jobs, use
wait comm
ands to ensure that normalization is completed. See WaitStorageVolum eNormalization,
WaitStorageVolumesNormalization,andWaitVolumeGroupNormalization.
Snapclone normalization (unsharing)
The snapclone normalization is a process which verifies that data is identical on a source virtual d isk
and its snapclone, before the snapclone c an become an independent disk (a point-in-time copy). See
snapclones. Snapclone normalization is also called unsharing.
Some actions should not be performed during s napclone normalization. When creating jobs, use
wait com mands to ensure that normalization is completed. See WaitStorageVolum eNormalization,
WaitStorageVolumesNormalization,andWaitVolumeGroupNormalization.
Operat
ional state - blocked
Blocked is an operational state of a DR group, or a virtual disk, that indicates if the array has detected
a potential replication issue and is preventing presentation of impacted source virtual d isks. See
prese
ntation to hosts.
The presentation of virtual disks in a source DR group is blocked after power is cycled off and on to
both controllers in the source array and:
108 DR Groups