DBAs noticed that since the upgrade to version 21, the database RT table is growing at an exponential rate. Automic maintenance executes daily (Archive, Reorg and Remove/Unload).
Why is the RT table not staying at a manageable size? DBAs are having to increase the filesystem to compensate but that is only a band-aid solution.
The DBAs just reported to me that the filesystem has grown by 5 gb in the past 2 days.
Release : 21.0
Component :
Take a look at your older ini files for the reorg an unload (ucybdbre.ini and ucybdbun.ini respectively). Do they include no_archive_check=1?
Some question to ask:
Do you have a requirement to records in a flat file (archive) for auditing?
Or is okay to simply delete the records from the database?
If it's okay to delete them, then updating the no_archive_check to 1 in both ini files will allow for deletion without archiving.
It may be that after the upgrade, since new ini files are being used, no_archive_check is set to 0 which means that the reorg will not flag rows for deletion and the unload will not delete them unless the archive is run first.
when no_archive_check is set to 0 in EITHER the reorg or unload ini files:
With no_archive_check set to 1 in both the reorg and unload ini files:
DB table information:
AH – Statistics
RH – header
RT - Reports
MELD –Messages
OH – all object informations including Version Management and Deleted Objects
XAO – Auditing Records
Info on no_archive_check can be found in the documentation for the ucybdbre.ini and ucybdbun.ini files