RDF System Management Manual

Managing RDF
HP NonStop RDF System Management Manual—524388-003
5-31
Performing Shared Access DDL Operations
Database administrators with a clear understanding of the underlying TMF auditing
issues might elect to skip some of these steps as long as the DDL operations and
other audited operations are performed in the correct sequence on the primary and
backup systems. For example, it is not absolutely necessary to stop TMF (and thus
RDF), but it is safest to do so. As long as application processing is stopped and the
display from a STATUS RDF command shows that the RTD time for every updater
process is zero, the DDL operations can be safely applied.
Performing Shared Access DDL Operations
DDL operations that include the WITH SHARED ACCESS option, and which are
performed on the primary system, generate a special Stop-RDF-Updater audit record
in the MAT. As each updater on the backup system encounters that record in its image
trail file, that updater logs an RDF message 733 and then shuts down. When all of the
updaters have done so, RDF logs a message 908 indicating that it is now safe to
perform the same DDL operations on the backup system. Do that, and then issue a
START UPDATE command.
If RDF aborts while the updaters are in the process of shutting down, check the RDF
log to see if RDF generated a message 908. If it did, then do as follows:
1. Issue a START RDF, UPDATE OFF command on the primary system.
2. Perform the DDL operation(s) on the backup system.
3. Issue a START UPDATE command on the primary system.
Whether or not RDF aborted while the updaters were shutting down, if one or more
updaters did not generate a message 733, RDF logs a message 905 indicating that
you must not perform the DDL operations on the backup system. If that happens, issue
a START RDF, UPDATE ON command (only those updaters that did not generate a
message 733 are started). Check the RDF log again to see if RDF generated another
message 905. If it did, issue another START RDF, UPDATE ON command, and so
forth, until RDF logs a message 908. When RDF logs the message 908, do the Steps 2
and 3 listed above.
RDF does not support shared access NonStop SQL/MP DDL operations where the
target file is located on a different node than the primary system. For example, if you
have two RDF subsystems, one going from system \A to system \B and the other going
from system \X to system \Y, you cannot move a partition boundary for a table on
Note. If you do not stop TMF on the primary system and wait for RDF to stop itself in response
to the TMFCOM STOP TMF command, make sure the RDF time delay (RTD) time for every
updater process is zero and that the updater processes have closed all database tables before
you start any DDL operation. To check the RTD times for the processes, issue a STATUS RDF
command. For this method to work, all applications protected by RDF must be shut down and
only the data base administrator manipulates the primary database.