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

labels of the duplicate files on the backup system so they reflect the correct volume mapping
of the various partitions on the backup system.
For example, suppose you have a partitioned Enscribe file on the primary system whose
primary partition is on $DATA1 and secondary partition is on $DATA2. If, on the backup
system, the primary partition is on $DATA1 but the secondary partition is on $DATA3, you
must change the volume name in the file label of the duplicate secondary partition from
$DATA2 to $DATA3:
FUP ALTER $DATA1.subvol.file, PART (1,$DATA3)
• If you move duplicate Enscribe alternate key files, you must alter the system name in the
file label of the duplicate file or table to specify the backup system.
For example, if you moved a duplicate Enscribe alternate key file named ALTF0100 associated
with the file PART0100, you must change the system name in the file label of the duplicate
alternate key file to that of the backup system:
FUP ALTER $DATA1.TEST.PART0100
ALTFILE (0,\backup.$DATA.TEST.ALTF0100)
• If you use the SQLCI DUP command to move duplicate partitioned NonStop SQL/MP tables,
you must use the MAP NAMES option to specify the backup system name.
• If you use the SQLCI DUP command to move NonStop SQL/MP tables with index tables,
you must use the MAP NAMES option to specify the backup system name.
Example of Synchronizing An Entire Database Online
Following is a summary of the steps necessary to perform an online synchronization of an entire
database.
1. Issue the following RDFCOM command:
STOP RDF
2. Purge the RDF control subvolume and then issue the following RDFCOM command:
INITIALIZE RDF, BACKUPSYSTEM \RDFB, SYNCHDBTIME 17JUN2004 17:05 !
NOTE: If you have multiple RDF environments, you can do online synchronization within
one RDF environment without disturbing any of the other RDF environments. Just be sure
to include the appropriate SUFFIX parameter in the above INITIALIZE RDF command.
3. Configure RDF and then issue the following RDFCOM command:
START RDF, UPDATE OFF
4. Create a set of empty nonaudited Enscribe files on the primary system with the same file
structure as the database files being synchronized. Although you can use the create filey
like filex method, the complete sequence of FUP SET and CREATE commands are included
below to show the file structure of the partitioned files.
volume $data2.test
reset
set type k, keyoff 2, keylen 4, no audit
set buffered, ext(10,10)
set rec 300, block 4096
set maxextents 16
set code 4700
set part (1, $data3, 2, 2, [0,0,0,195] )
set altkey (1, file 0, keyoff 6, keylen 2 )
set altkey (2, file 0, keyoff 6, keylen 2 )
set altkey (3, file 0, keyoff 6, keylen 2, no update )
set altkey (4, file 0, keyoff 6, keylen 2, no update )
set altkey (5, file 0, keyoff 6, keylen 2 )
set altfile (0, $data2.test.altf0100 )
create $data2.test.part0100
Synchronizing Entire Databases Online 161