Disaster recovery rehearsal in Continentalclusters

4. Edit the package configuration file saved in the rehearsal package directory and change

PACKAGE_NAME, RUN_SCRIPT and HALT_SCRIPT as follows:

PACKAGE_NAME ”sales_rac9i_rhpkg_2”

RUN_SCRIPT /etc/cmcluster/

sales_rac9i_rhpkg_2/sales_rac9i_rhpkg_2.cntl

HALT_SCRIPT /etc/cmcluster/

sales_rac9i_rhpkg_2/sales_rac9i_rhpkg_2.cntl

Also modify the SERVICE_NAME parameter.

5. Edit the package control file saved in the rehearsal package directory, and ensure that the

package directory paths are changed to rehearsal package directory paths. Modify the

SERVICE_NAME parameter. Also modify the IP[] parameter, to use a different IP address.

6. Apply the rehearsal package configuration on the recovery cluster as follows.

#cmapplyconf –P /etc/cmcluster/sales_rac9i_rhpkg_2/sales_rac9i_rhpkg_2.conf

Note: Ensure that the AUTO_RUN is set to false for the rehearsal packages.

Caution: While creating the rehearsal packages, ensure that the Metrocluster environment file is not copied to the

rehearsal package directory. The rehearsal startup using cmrecovercl –r with the presence of Metrocluster environment

file can potentially prepare the secondary mirror copy as primary mirror copy. Hence it can result in the rehearsal

package being started on the primary production copy, resulting in data corruption.

Primary package MC environment file

In case of Continentalclusters using EMC SRDF data replication, before starting rehearsal, the

Metrocluster environment variable AUTOSPLITR1 in the primary package, has to be set to 1. This

ensures that, during rehearsal, the primary packages will be highly available at the primary cluster.

In this example, it is assumed that the recovery group sales_rac9i_recgp_1 and sales_rac9i_recgp_2

are configured with an SRDF device group. If the variable is not set, follow the steps shown on

next page.