Copyright © TIBCO Software Inc. All Rights Reserved
Copyright © TIBCO Software Inc. All Rights Reserved


Chapter 5 Backing Up Your System : Sample Continuous Backup Implementation

Sample Continuous Backup Implementation
Overview of Continuous Backup Data Flow
The following diagram illustrates one possible continuous backup strategy. To develop a customized plan for your installation, consult your TIBCO Support representative.
Steps in the Continuous Backup Illustration
This diagram illustrates the following steps:
1.
2.
Images of all updated (Pagestore) pages are copied— journaled—to the active journal data set by the Data Object Broker. If you need to, set up a similar process for batch utility journaling.
3.
When the active journal fills up, TIBCO Object Service Broker starts a spin job—named SPINnn—to copy the contents of the journal data sets into the backup process.
4.
Periodically, when the user-defined limit is reached on the number of JOURNAL.SPINOUT data sets, SPINMRG is automatically submitted to consolidate information into a new generation of the JOURNAL.SPINMRG data set. All generations of JOURNAL.SPINOUT are deleted.
When a specific page appears more than once in the input SPINOUT data sets, a TIBCO Object Service Broker exit in this job’s sort phase causes only the most recent version of the page to be written out. (If you are using point-in-time recovery, another exit, which retains all page images, is substituted.)
5.
Having created a consolidated SPINMRG data set, consisting of the most recent images of updated pages, the previous SPINMRG job automatically submits BKUPCON. This updates the current Pagestore backup, creating a new generation of the BACKUP data set.
The BKUPCON job takes the most recent JOURNAL.SPINMRG data set as input. This means that new SPINOUT generations or page images not spun from the current JOURNAL data sets—beyond those read by the latest BKUPCON— go into another backup generation at a later time. The output of BKUPCON is a new generation of the BACKUP generation data set. This data set contains a complete set of page images that can be used to recover a Pagestore using the TIBCO Object Service Broker restore utilities. You should always run the S6BBRPTR (Batch Pointer Check) utility against newly created backup data sets to ensure their integrity.
See Also
TIBCO Object Service Broker for z/OS Utilities for information on the backup utilities and on batch journaling.

Copyright © TIBCO Software Inc. All Rights Reserved
Copyright © TIBCO Software Inc. All Rights Reserved