RDF System Management Manual for J-series and H-series RVUs (RDF 1.10)

Please note that in each of these configurations, there is no possible overlap of data being replicated
by the two RDF subsystems in a reciprocal configuration. See Reciprocal and Chain Replication
in Chapter 1 for a discussion of the potential problem that can occur if you do not use one of the
two methods described above.
The steps for performing a planned switchover from \A to \B in such a configuration are:
1. On system \B, stop RDF subsystem # 2. Note the local system time; you will need it later.
2. On system \A, stop the business applications that access the primary database (Applications
#1).
3. On system \A, stop TMF(or if you do not want to stop TMF, use the STOP RDF, DRAIN
command).
4. Wait for RDF subsystem #1 on \A to shut down.
5. On system \B, restart Applications #1.
At this point, the RDF subsystem is down on both systems, the business applications from system
\A are now running on system \B, the business applications that were running on system \B
are still running on system \B, and all audit records are being queued in TMF audit trails on
system \B.
6. When system \A is ready to resume its normal operations, restart TMF.
7. On system \B, restart RDF subsystem #2 (the RDF subsystem that replicates data from \B to
\A for the business applications that normally run on \B). The subsystem resumes its processing
exactly where it was when you stopped it in step 1.
8. On system \B, initialize RDF subsystem #1. In the INITIALIZE RDF command, include the
INITTIME option and specify the timestamp you noted in step 1. Configure the subsystem to
replicate data from \B to \A for Applications #1 (the business applications that were moved
from system \A to system \B).
When the extractor for RDF subsystem #1 reports an RTD time of 0:00, then you know that extractor
has caught up and you can then prepare for another switchover operation to move your application
processing back to \A, as follows:
1. On system \B, create an audited Enscribe file on each data volume in the RDF subsystem #1
configuration.
2. Wait until all of those files are created on system \A.
3. On system \B, stop RDF subsystem #1.
4. Purge the Enscribe files on both systems.
5. On system \A, initialize RDF subsystem #1 using the INITTIME option and specifying the
current (for \A) local time.
6. On system \A, restart Applications #1.
NOTE: There are variety of variations you can do to achieve the above operations. The method
provided above is just one means to achieve this.
Takeover Operations
If the primary system fails and you want to switch application processing to the backup system,
you need to issue the TAKEOVER command on the backup system. The TAKEOVER command
causes RDF to shut down after bringing the backup database to a consistent state.
The RDF Takeover Operation
When updating is enabled, updaters apply audit as soon as it is safe-stored in the image trails on
the backup system. In this respect, they apply audit without waiting to determine if the associated
transactions committed or aborted. At the moment when you lose your primary system due to some
unplanned outage, the updaters might have applied audit for transactions whose outcomes were
not resolved (committed or aborted) on the primary system at the time the primary system failed.
Alternatively, the transactions might have been resolved on the primary system, but the extractor
130 Critical Operations, Special Situations, and Error Conditions