HP StorageWorks XP24000 Continuous Access Journal Software User and Reference Guide, v01 (T5278-96001, June 2007)

9 Performing Disaster Recovery
Operations
Preparing for Disaster Recovery
The t ype of disaster and the status of the Continuous Access Journal volume pairs will determine the best
approach for disaster recovery. Unfortunately, some disasters are not so orderly and involve intermittent
or
gradual failures occurring over a longer period of time. The user should anticipate and plan for all
typesoffailuresand disasters.
The major steps in preparing for disaster recovery are:
1. Identify the journal groups and data volumes that contain important files and data for disaster
recovery .
2. Install the Remote Web Console computer and Continuous Access Journal hardware and
software, and establish Continuous Access Journal operations for the journal groups and data
volumes identified in step (1).
3. Establish file and database recovery p rocedures. These procedures should already be
established for recovering data volumes that become inaccessible due to some failure.
4. Install and configure host failover software between the primary and secondary sites.
File and Database Recovery Procedures
When the primary or secondar y storage system suspends a Continuous Access Journal pair due to
adisast
er, the secondary data volume may contain in-process data. A data set could be open o r
transactions may not have completed. Therefore, you need to establish fi le recovery procedures. These
procedures should be the sam e as those used for recovering a data volume that becomes inaccessible
due
to a control unit failure.
Continu
ous Access Journal does not provide any procedure for detecting and retrieving lost updates. To
detect
and
recreate lost updates, you must check other current information (for example, database log
file) th
at
was active at the primary site when the disaster occurred. Since this detection/retrieval process
can
take a while, your disaster recovery scenario should be designed so that detection/retrieval of lost
updates is per formed after the application h as been started at the secondary site.
You should prepare for file and da tabase recovery by using:
• Files
for file recovery (for example, database log files which have been verified as current).
Important: Remote copy and disaster recovery procedures are inherently complex. Consult your HP
service representative on recovery procedures.
Note:
See “Recovering a Pinned Track” on page 140 for information on recovering a pinned track on
aContinuousAccessJournal volume.
Switching Operations to the Secondary Site
If a disasterorfailure occurs at theprimary site, the first disaster recovery activity is to switch your
operations to the secondary site. Executing the horctakeover command of R AID Manager on Continuous
Access Journal pairs enables your business tasks to be taken over to the secondary site, using secondary
data volumes in the secondary site. The horctakeover command checks the pair status of secondary
data volumes and splits journal groups that will become a Continuous Access Journal pair. This ensures
consistency of secondary data volumes and makes the secondary data volumes usable. If possible, the
horctakeover command attempts to restore pairs to reverse primary data volumes and secondary data
volumes. If the horctakeover comm and executes successfully, your b usiness tasks will be taken over to
XP24000 Continuous Access Journal Software User a nd Reference Guide
145