Failover and failback operations

Edit online

If a session has failover and failback capabilities, you can perform a site switch in which you move the operation of Copy Services Manager from one site to another, and change the direction of the copy without completing a full copy.

Failover is the process of temporarily switching production to a backup facility after a scheduled maintenance period or a disaster at a production site. A failover operation can be followed by a failback operation, which is the process of returning production to its original location and resuming the copy between the two sites.

The failover and failback operations use remote mirror and copy functions for these tasks:

To reduce the time that is required to synchronize volumes after sites are switched during planned or unplanned outages.
To eliminate the need to perform a full volume copy from the recovery site to the production site, which can reduce the time that is required to resume operations at the production site.

Failover operations that are managed by other applications

Applications such as the IBM® Series i Toolkit, VMware Site Recovery Manager, and Veritas Cluster Server can also manage failover operations for the session types that are listed in the following table.

Session type	Storage systems
Metro Mirror Single Direction	IBM DS8000 Systems running IBM Spectrum® Virtualize
Metro Mirror Failover/Failback	IBM DS8000 Systems running IBM Storage Virtualize Systems running IBM Spectrum Accelerate, including: XIV® FlashSystem A9000 FlashSystem A9000R
Metro Mirror Failover/Failback with Practice	IBM DS8000 Systems running IBM Storage Virtualize
Metro Mirror Failover/Failback with Change Volumes	Systems running IBM Storage Virtualize
Multi-target session types: Metro Mirror - Metro Mirror Metro Mirror - Global Mirror Metro Mirror - Global Mirror with Practice Metro Mirror - Global Mirror with Site 3 Global Mirror	IBM DS8000 Note: Multi-target session types are available only for IBM DS8000 storage systems with a microcode level that supports single source to multi-target relationships. To determine whether you can use this session type, see the IBM DS8000 documentation for the microcode level that you are using. IBM FlashSystem® A9000 and A9000R Note: Only the Metro Mirror - Global Mirror multi-target session type is available for the FlashSystem A9000 and A9000R storage systems.
Global Mirror Single Direction	IBM DS8000 Systems running IBM Storage Virtualize
Global Mirror Either Direction with Two-Site Practice	IBM DS8000
Global Mirror Failover/Failback	IBM DS8000 Systems running IBM Storage Virtualize Systems running IBM Spectrum Accelerate, including: XIV FlashSystem A9000 FlashSystem A9000R
Global Mirror Failover/Failback with Practice	IBM DS8000 Systems running IBM Storage Virtualize
Global Mirror Failover/Failback with Change Volumes	Systems running IBM Storage Virtualize

If an application completes a failover operation for a session, the Severe status is displayed for the session on the Session Details page in Copy Services Manager. An error message is also generated for the role pairs for which the failover occurred. To view the error message:

On the Session Details page, click the link for each role pair for which the failover occurred.
On the Role Pair Details page, click the Show link for each role pair. This link is in the Details column.

To change the session status to Normal, use the application to confirm that the role pairs that are in the session are consistent and then restart the session.