We keep track of tablespaces by a weekly process. It started with one or 2
databases, then moved into all production dbs (about 20). Eventually it turned
into tables (over 1GB in size), then indexes. Now we have a running tab of
tablespace, table and index growth that goes back 4+ years. It is great for
looking at growth trends.
Jeremy
From: oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On
Behalf Of Upendra nerilla
Sent: Wednesday, November 11, 2015 5:14 PM
To: Oracle-L
Subject: Segment growth monitoring
This is an EXTERNAL email. Exercise caution. DO NOT open attachments or click
links from unknown senders or unexpected email.
________________________________
Hello all -
I am trying to track the growth of a few problem tables.. if there is a runaway
process, they might continue to write without dying.. I looked at
dba_hist_seg_stat.space_used_delta, it doesn't seem to be consistent in
tracking the growth. Especially for situations like, if the table gets
truncated or if there are massive deletes, this data doesn't accurately
represent reality..
Any suggestions on how it could be tracked?
Worst case I am thinking of dumping the dba_segment data into a table nightly
to track it. Trying to see if there are better ways to handle this.
Thanks
-Upendra
