using flashback in streams env

From: Venkat Krish <venkat.lear_at_gmail.com>
Date: Tue, 23 Aug 2011 10:22:17 -0400
Message-ID: <CAOV8h7tYTSPQyLMTY8265aSaBd2B6jSXCnQ5sji98TM8k3ufjw_at_mail.gmail.com>

Hello listers -
 Just wondering if anyone have experience on dealing/implementing with Streams+Flashback combination.

Here is my situation..

• 30+ tables changes are to be captured in source db (say db_source) and to be replicated to diff database (say db_target)
• db version: 11.2.0.2 (both database) running on Solaris

this is a testing environment & there is a pre-req to run some process from application to setup some mandatory data into database before the start of each test. (no access to application code and the option of getting it done from db-scripts is not feasible)

I know flashback would work great , but I am not sure about the same on an env with streams configured.

Here is my idea

(-- assume all db configurations are in place to store flashback logs and
database is enabled with Flash-logging )

t1) stop capture process at source db, run the application to set up data
(needed before every test)

t2) create a restore point
t3) start capture
t4) run the test

now if I need to do a different test, I am hoping to get back to t2
(flashback database + db bounce).

Am I missing/overlooking anything to consider? The target database (where apply is running) is not critical to be in-sync with source or to maintain data integrity. (if needed, I can stop apply process and truncate tables on target..)

Any clues/pointers would be much appreciated,

Thanks in advance,
Venkat

PS: Only a handful tables ( less than 50 tables ) are being replicated. So I am also open to run flashback table (instead of flashback database).

 -

--
http://www.freelists.org/webpage/oracle-l

Received on Tue Aug 23 2011 - 09:22:17 CDT