The SPILL function transfers data from the Log Area (LXX) to the Recovery File (RXX). All spilling is controlled and done through the Multi-User Facility, which must be enabled to do a spill. The SPILL function is allowed for both active and inactive recovery. The SPILLREW function also transfers data, but provides the REWIND= parameter for you to specify rewinding instructions (see SPILLREW Function).
If blocks are spillable, the utility opens the Recovery File, copies the appropriate log blocks (from Multi-User) to the Recovery File, and closes the file. The space occupied by the spilled data is made available when the function is complete.
With an Active Recovery File, the spill is automatically done when the percentage is equal or above the LOGSPILL percentage and log blocks are spillable.
See the CA Datacom/DB Database and System Administration Guide for details about the LOGSPILL, LOGRCV, and MESSAGE Multi-User startup options.
Note: When using inactive recovery (that is, NO specified for the Multi-User startup option LOGRCV) and checking for spillable data based upon the LOGSPILL percentage, CA Datacom/DB presumes that the portion of the Log Area (LXX) that contains pipelined index and/or data need not be spilled when determining the number of blocks spillable. This allows a spill option to reduce index/data I/O when preparing for a spill.
|
Copyright © 2015 CA Technologies.
All rights reserved.
|
|