Cascading Failover in a Continentalclusters, December 2005

3. Start the application on the primary site in the primary cluster.
Procedure 2: Replicating from the Secondary to the Recovery Disk Array
This procedure is illustrated in Figure 3.
Figure 3 - Replicating from the Secondary to the Recovery Disk Array
Execute the following steps:
1. Freeze the application I/O to prevent the application from writing data to the primary
devices in the primary disk array (in case of adding the recovery cluster to the existing
Metrocluster). Note, that the method of freezing the I/O is application dependent
2. Split the local mirror devices (device B’) in the secondary disk array from the primary
replication group
3. Resume the application I/O to the primary devices in the primary disk array if needed. Also,
note that the method of resuming the I/O is application dependent.
4. Establish the data replication between secondary disk array (device B’) and the recovery disk
array (device C). The data is copied from local mirror devices on the secondary disk array to
the destination devices on the recovery disk array.
5. Check the data synchronization state of the replication. Wait until the data replication
completes.
6. Once the copy completes, split the data replication link between the secondary disk array
and the recovery disk array (that is the recovery replication group).
7. Incrementally establish the local mirror devices in (device C’) the recovery disk array as
mirrors of the standard devices if they are not already established. These devices were fully
established when the volume group were created.
8. Re-establish the local mirror devices (device B’) in the secondary disk array as mirrors of the
standard devices.
Refreshing data on the Recovery Site
Once the application starts writing data to the primary disk array devices, the data on the recovery
disk array is out of sync with the primary data; the data is not current but consistent. The data from
the primary cluster needs to periodically replicate to the recovery disk array, so its data is more up to
date. As long as the application continues writing new data to the primary disk array, the data on
the recovery disk array will always be behind. The level of data currency on the recovery disk array is
6