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

Creating the Fuzzy Copy on the Backup System
The advantage of this method is that it eliminates the use of temporary objects as well as tape
handling because you create your backup objects directly on the backup system. The disadvantage
is that it requires you to load the data from your primary objects to your backup objects over
Expand lines, which might take longer than the alternate method given above if the data is great
in size.
To create the fuzzy copy on the backup system, perform these steps.
1. Create the backup catalog on your backup system. This operation is identical to operation
in “Creating a NonStop SQL/MX Backup Database From an Existing Primary Database”
(page 299).
2. Create the schema on the backup system. This operation is identical to that outlined in
“Creating a NonStop SQL/MX Backup Database From an Existing Primary Database”
(page 299).
3. So each catalog can be seen from both systems, you must register your primary and backup
catalogs as described in “Creating NonStop SQL/MX Primary and Backup Databases From
Scratch” (page 296).
4. Obtain the fully qualified Guardian filenames for all objects on the primary system that you
want replicated.
5. Create each object on the backup system. See “Creating NonStop SQL/MX Primary and
Backup Databases From Scratch” (page 296).
6. Determine where you will run the command to load the data from the primary objects to
the backup objects. If you run the command on the primary system, NonStop SQL/MX
selects the data locally and inserts over the network into the backup object. Alternatively,
you can run the operation on the backup system, which selects the data remotely and inserts
the data locally.
7. Populate the backup objects by running an INSERT statement on whichever system you
prefer.
INSERT INTO backup-table SELECT * FROM primary-table;
where backup-table and primary-table are the 3-part ANSI names of the two tables.
The backup database is now ready for RDF replication.
Offline Synchronization for a Single Partition
You must first determine the key ranges for each of your partitions.
For the partition that you want to synchronize, find out whether a partition already exists on the
backup system. If a partition exists, then there are two methods to synchronize the backup
partition to the primary: directly and indirectly.
Directly From the Primary to the Backup
1. Delete all rows in the backup partition. If the partition whose rows you want to delete has
the key range of "F"-"J" and the next partition starts with "K", delete rows from the F-J
partition:
DELETE FROM name
WHERE key-column >= 'F' and key-column < 'K';
2. Load the rows from the primary partition into the backup partition. This requires each
catalog to be registered on the other node as described in “Creating NonStop SQL/MX
Primary and Backup Databases From Scratch” (page 296).
INSERT INTO backup-table SELECT * FROM primary-table
WHERE key-column >= 'F' AND key-column < 'K';
Offline Synchronization for a Single Partition 303