Failover And Failback Operations - IBM DS8000 User Manual

Failover and failback operations

The failover operation is the process of switching production to a backup facility
(normally your recovery site). A failback operation is the process of returning
production to its original location after a disaster or a scheduled maintenance
period.
There are times, both planned and unplanned, when it is necessary to suspend
disk mirroring and to make use of the secondary storage unit in your
configuration. As a manual process, this can be complex. However, failover and
failback recovery operations are available to simplify this process and reduce the
risk of error and the time it takes to switch sites and restart I/O operations.
Failover is the process of temporarily switching production to a backup facility
(normally your recovery site) following a scheduled maintenance period or a
disaster at your production (or local) site. A failover operation is always followed
by a failback operation, which is the process of returning production to its original
location. These operations use remote mirror and copy functions to reduce the time
that is required to synchronize volumes after switching sites during planned or
unplanned outages.
The failover and failback operations allow change recording to be enabled on the
target volumes without having to communicate between the target and source
storage units. This method eliminates the need to perform a full volume copy from
your recovery site to the production site, which can reduce the time that is
required to resume operations at your production site.
In a typical remote mirror and copy environment, processing will temporarily
failover to the storage unit at your recovery site if an outage occurs at the
production site. Through use of failover operations, the state of a storage unit in
your target configuration changes. As a result, the storage unit is recognized as the
source storage unit in the pair. Because the failover process puts the volumes into a
suspended state, changes are tracked within a bitmap. Assuming that change
recording is enabled, only change data is sent to the production site to synchronize
the volumes, thereby reducing the time that is required to complete the failback
operation.
When it is safe to return to your production site, assuming that no physical
damage has occurred to the storage unit in the location, you can delete paths and
create new ones from your production site to your recovery site. Then, you can
create a failback recovery request to restore the storage unit as the production
storage unit in the relationship.
The following considerations are for failover and failback operations:
v The failover operation does not reverse the direction of a remote mirror and
v The failback recovery operation can be issued against any remote mirror and
132
DS8000 User's Guide
copy pair. It changes a target device into a suspended source device, while
leaving the source device in its current state.
copy volume that is in a primary suspended state. The operation copies required
data from the source volume to the target volume in order to resume mirroring.
Failback operations are commonly used after a failover operation has been
issued to restart mirroring either in the reverse direction (remote site to local
site) or in the original direction (local site to remote site).