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

• Issues a logical REDO request to the disk process (during the normal forward pass over the
image trail) for each update associated with its volume.
• Issues logical UNDO requests to the disk process when backing out changes associated with
transactions that need to be undone during RDF takeover or stop-update-to-timestamp operations.
• Bundles the REDO and UNDO requests into batch TMF transactions, the duration of which is
specified by the UPDATERTXTIME configuration parameter.
• For Enscribe files only, performs the following DDL operations:
CREATE, PURGE (if REPLICATEPURGE is enabled), PURGEDATA, ALTER MAXEXTENTS (used
only for increasing MAXEXTENTS).
• For NonStop SQL files only, performs the following DDL operation: PURGEDATA
An updater cannot always respond immediately to the STOP UPDATE and STOP RDF commands.
If an updater has audit records queued for the disk process, the updater must wait until all of that
information is processed before it can shut down.
You specify the primary and backup CPUs for each updater. If the original backup process has to
take over because the primary CPU failed, this backup process runs by itself. When it determines
that the primary CPU has come back up, it creates a new backup process in that CPU. When it
has to take over, the original backup process becomes the primary process, and remains so even
after it creates a new backup process; that is, the updater does not switch back to the original
CPU configuration after the new backup process is created. If you stop the updaters by way of a
STOP RDF or STOP UPDATE command, however, when you restart the updaters, your original
configuration is once again used.
The updaters will shut down if any of the following occurs:
• You issue a STOP RDF or STOP UPDATE command on the primary system.
• You issue a STOP RDF command on the backup system when the communications lines between
the two systems are down.
• You issue a STOP TMF command on the primary system.
• The monitor detects the unexpected termination of any RDF process and sends out abort RDF
messages.
• You perform a NonStop SQL DDL operation on the primary system that includes the WITH
SHARED ACCESS option for an RDF-protected file. For more information, see “Performing
Shared Access DDL Operations” (page 143).
If you perform a NonStop DDL operation WITH SHARED ACCESS on a table or index that is
not configured for RDF protection by your current RDF subsystem, then this current RDF subsystem
does not shut down.
• A takeover operation completes on the RDF backup system.
Audited Database Files
All database files on the backup system are audited files.
Each updater maintains a file status table to keep track of the files it has open. An updater closes
any database file that has not been updated recently. Updaters also close database files when a
STOP RDF or STOP UPDATE command is issued, or when the updater restarts because of error
conditions. Additionally, if you alter the updater's OPENMODE while UPDATE is ON, then the
updater closes all its file and then reopens them with the new OPENMODE.
An updater process can have up to 3000 files open simultaneously. When it has the maximum
number of files open and needs to open another file, it first determines if there are any files that
have not been accessed recently and closes just them; if all of the open files have been accessed
recently, then the updater closes all of them before it continues processing. For the SMF ramifications
of this file limit, see the note in “Using SMF With RDF” (page 60).
42 Introducing RDF