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

To load the secondary partition only, issue the following command:
FUP LOAD $DATA2.TEST.PART0100, $DATA2.TEMP.PART0100,
PARTONLY,SHARE
When the load operations are finished, use BACKUP and RESTORE (or FUP DUP) with the
PARTONLY option to copy the partition you need to the backup system.
Relative Files
First create a non-audited duplicate file on the primary system. You must create the entire file
with all its partitions. Unlike key-sequenced files, you must load the entire file. For example,
assume the file has two partitions: $DATA1.TEST. PART0100 (primary) and
$DATA2.TEST.PART0100 (secondary). Issue the following command:
FUP CREATE $DATA1.TEMP.PART0100,
LIKE $DATA1.TEST.PART0100, NO AUDIT
That command creates the two files $DATA1.TEMP.PART0100 (primary partition) and
$DATA2.TEMP.PART0100 (secondary partition)
You must load all the partitions of a relative file. Therefore, only one command is possible.
FUP LOAD $DATA1.TEST.PART0100, $DATA1.TEMP.PART0100,
SHARE
You can then use BACKUP and RESTORE (or FUP DUP) with the PARTONLY option to copy
the loaded partition you need to the backup system.
Entry-Sequenced and Unstructured Files
Currently there is no reliable mechanism to synchronize these file types online. Entry-sequenced
and unstructured files should be synchronized offline.
NonStop SQL/MP Tables Without Partitions
Key-Sequenced Tables With SYSKEY
Because the SQLCI LOAD operation generates new SYSKEY values in the backup table,
key-sequenced tables containing SYSKEYs cannot be synchronized online.
Key-Sequenced Tables Without SYSKEY and Relative Tables
First create a non-audited duplicate table on the primary system and then load it. Use BACKUP
and RESTORE (or SQLCI DUP) to move the duplicate table to the backup system.
Alternatively, you can use the create and load method to put the duplicate tables directly onto
the backup system, provided you have enough Expand capacity to handle both the data being
loaded and the audit being shipped to the backup system by the extractor.
There are no special considerations for key-sequenced tables with indexes, but see below for
issues regarding the synchronization of indexes.
NonStop SQL/MP Tables With Partitions
The utilities associated with and related to the NonStop SQL/MP product have limitations that
make synchronization of individual partitions complicated and difficult. The following represent
methods that enable you to circumvent these limitations.
• SQLCI DUP does not have a PARTONLY option. Therefore, you cannot duplicate only an
individual partition.
• While BACKUP and RESTORE have the PARTONLY option, if you have backed up tables
from the primary system, you must use the MAP NAMES option when restoring them on
the backup system in order to specify the correct system name. You cannot, however, include
both the MAP NAMES and PARTONLY options in the RESTORE operation. Therefore,
because you must use MAP NAMES, you cannot restore only a single partition.
Synchronizing Selected Database Portions Online 165