Administrator Guide

A volume can become out of sync if synchronous replication is paused, or if one of the volumes becomes unavailable or has no
free space. The volume can become out of sync when the snapshot reserve in the SyncAlternate volume is full, but only when
the snapshot space recovery policy sets volumes oine when the snapshot reserve is depleted.
Whenever the volume’s state changes to paused or out of sync, the group creates a snapshot of the volume that reects the
volume’s contents at the point in time when the state changed. This snapshot resides in the active pool. If not enough space in the
snapshot reserve is available for the snapshot, the group does not create the snapshot.
NOTE: When synchronous replication is rst enabled, or at any other time when data is being written to both volumes to
become in sync, performance degradation might occur. This eect increases with the quantity of tracked changes, but it
is signicantly reduced after the volumes become in sync. Also, depending on the quantity of tracked changes, activity
within the group, and available network bandwidth, an extended period of time might elapse before the two volumes
become in sync again. The Group Manager GUI displays the status of this operation.
About System Snapshots and SyncRep
System snapshots are a special type of snapshot that are automatically created by the system during management of synchronous
replication (SyncRep) volumes. System snapshots reside in the same snapshot reserve space as other snapshots. Although you
cannot modify their attributes, they are visible in the Group Manager GUI or CLI, and you can create clones of them. You can clone a
system snapshot if you need access to the volume’s data as it existed at the point in time where the volume went out of sync.
System snapshots are of two types, protected and failback.
Protected System Snapshots
Protected snapshots are created before the system writes tracked changes to the SyncAlternate volume to keep the volume in sync.
Protected snapshots reside in the alternate pool. They represent the state of the SyncAlternate volume at the point in time when the
volume went out of sync.
You cannot delete protected snapshots; the system deletes them automatically after the volume becomes in sync.
If not enough space in the snapshot reserve is available for the protected snapshot, the system will not be able to put the volume
back in sync unless the group’s snapshot space recovery policy has snapshot borrowing enabled.
Failback System Snapshots
Failback snapshots are created in the active pool when the volume is paused or goes out of sync. If not enough snapshot space is
available to contain the failback snapshot, you cannot fail back to the SyncActive volume unless the group’s snapshot space recovery
policy has snapshot borrowing enabled.
About Synchronous Replication and Snapshots
You can create snapshots of synchronous replication volumes. You can access the data in the snapshots by setting the snapshot
online or cloning it.
With synchronous replication volumes, the snapshot reserve percentage you specify for a volume is applied to both the SyncActive
and SyncAlternate pools. For example, if you are creating a 20GB volume and setting the snapshot reserve to 200 percent, you need
to have:
• 20GB for the volume reserve in the SyncActive pool
• 20GB for the snapshot reserve in the SyncActive pool
• 20GB for the volume reserve in the SyncAlternate pool
• 20GB for the snapshot reserve in the SyncAlternate pool
About Synchronous Replication
303