RDF System Management Manual for H-Series RVUs (RDF 1.8)

where system is the name of the backup system, ddmmmyyyy is today’s date (such as
17DEC2004), and hh:mm is an appropriate timestamp prior to the current time (see the
description of the INITIALIZE RDF command in Chapter 8 (page 173)).
NOTE: If you have multiple RDF environments, be sure to specify an appropriate SUFFIX
parameter in the above INITIALIZE RDF command to keep this RDF configuration separate
from the other RDF configurations on the primary system. Also, if this configuration is to
protect different data from the other RDF configurations, you might want to consider using
the INCLUDE and EXCLUDE options for your updaters.
3. Configure RDF and then issue a START RDF, UPDATE OFF command on the primary
system.
4. Make a copy of all tables and files in the database:
Create an empty set of duplicate key-sequenced and relative tables and files on either the
primary or backup system. The duplicate tables and files must not be audited.
You can create these files and tables on either the primary or backup system, but you should
only create them on the backup system if you have sufficient Expand bandwidth between
the two systems to handle both the audit sent by the extractor and the data sent by the load
operations.
5. Using load commands, populate the empty tables and files.
For NonStop SQL/MP tables, use SQLCI LOAD commands with the SHARE option.
For Enscribe files, use FUP LOAD commands with the SHARE option.
6. When the load operations are done, issue the RDFCOM STOP SYNCH command on the
primary system. This command issues a message to the extractor. The purpose of this
command is to enable RDF to determine when the synchronization operation has completed
and the backup database is synchronized with the primary database.
When the extractor completes its role in the online synchronization operation, it generates
the RDF Event 782 and then resumes normal operations. For more detailed information, see
“Phases of Online Database Synchronization” (page 169).
7. If the duplicate tables and files were created on the primary system in step 4, use
BACKUP/RESTORE or FUP DUP operations to copy them to the backup system.
For ENSCRIBE files with alternate key files, after restoring the files to the backup system,
if the name of the alternate key file is in network form, then you must manually alter the
system name of the alternate key file in the file label, replacing the name of the primary
system with that of the backup.
For example, suppose that on the primary system (\PRIMARY) you have a file named
$DATA.TEST.PART0100 with an alternate key file named ALTF0100.
After restoring both files to the backup system (\BACKUP), you must then use a FUP ALTER
command to alter the file label of PART0100 to point to the alternate key file on the backup
system.
FUP ALTER $DATA.TEST.PART0100,
ALTFILE ( 0, \BACKUP.$DATA.TEST.ALTF0100 )
This command does not pertain to NonStop SQL/MP indexes because their labels are
automatically corrected by the MAP NAMES option of the RESTORE utility.
The same issue pertains to Enscribe partitioned files. If the primary partition references
secondary partitions that include the primary system name, you must alter the primary
system name to that of the backup system.
If you have an RDF network for replicating network transactions, then you will need to alter
the partition names to reference the correct names of the backup systems where the partitions
are located.
156 Online Database Synchronization