Processing for the Dcf Commands - Documentation for Database Director for Ims 8.2
Bmc Database Director for Ims Issues: for Full Function Databases, Dcf Chkp and Dcf Logdump Commands for Fast Path Databases (Dedbs), Dcf Fpchkp and Dcf...
BMC Database Director for IMS issues:
-
For full function databases, DCF CHKP and DCF LOGDUMP commands
-
For Fast Path databases (DEDBs), DCF FPCHKP and DCF FPLGDUMP commands
Database Director issues a DCF CHKP or DCF FPCHKP command to obtain a transactionally consistent synchronization point for the IMS databases being processed by a command (such as the REORG command).
The technology eliminates the need to issue /DBR or /DBD and /STA commands in order to create a transactionally consistent set of database data sets.
During this time, DCF:
-
Quiesces online updating of the databases
-
Establishes a log record intercept
-
Releases online transactional activity for the databases
To obtain a transactionally consistent synchronization point, all updaters of the databases being processed must reach a syncpoint before the log record intercept can be established. In order to establish a syncpoint, an online application program must:
-
Issue a ROLB, SYNC or CHKP call
-
Issue a GU to their I/O PCB
-
Terminate (normally or abnormally)
There is more than 1 method to quiesce online updating of the database.
| Condition at Your Site | Action to Take | Action Database Director Takes |
|---|---|---|
| The syncpoint interval for BMPs is regular and frequent | No action required | Database Director waits for 5 seconds for all the targeted BMPs and MPPs to reach syncpoint. (In most cases it takes less than 1 second.) If the BMPs and MPPs do not reach a syncpoint in 5 seconds, Database Director:
|
| The syncpoint interval for BMPs is irregular or long (for example, 1 or 2 minutes) | Either: | Database Director waits for all targeted BMPs to reach their syncpoint for twice the time you specify for the: For example, if you specify a 300 second (5 minute) timeout value, Database Director waits for 600 seconds (10 minutes). If the BMPs do not reach a syncpoint in the time specified, Database Director release online transactional updating and issues the RETRY or CANCEL command. |
After the logical copy phase is complete for all database data sets, Database Director will issue a DCF LOGDUMP or DCF FPLGDUMP command. The command:
The captured log records are then sorted and used to back out any online updates which may have been applied to the shadow database data sets. This will bring all the shadow databases involved in the reorganization or clone process back to the same transactionally consistent point-in-time.
When Database Director is processing Fast Path databases, the captured log records are passed to BMC iRecover for IMS. iRecover performs a forward recovery to bring all of the areas to a transactionally consistent point in time.
David Miller brings 15 years of experience in global economics, personal finance strategy, and market dynamics. He specializes in turning complex economic trends into actionable insights for everyday readers.
