Thanks Riyaj, that is very good info.
Couple of questions:
- Is there a way to identify connections that come over using database links?
- Once all remote DB Links connecting into our DB are identified, is there a
way to find which ones may be causing the most SCN jump/increments ?
On Thursday, May 26, 2016 10:51 AM, Riyaj Shamsudeen
<riyaj.shamsudeen@xxxxxxxxx> wrote:
Hello Deepak, Your assertion is correct, database link sessions will
synchronize the SCN between the databases, thereby, increasing the SCN of one
or more databases sharply.
So, if any one of your database is encountering from the bug, then, all
databases interconnected through database links can suffer the consequences.
You may be using RMAN backup, however, some database connecting to your
database over database link(s) might be using database level 'begin backup'
command, infecting your database SCN value too.
This will be an hard issue to debug: You need to identify if any of the
database connecting to your database, directly or indirectly, is using 'alter
database begin backup' command and if so, does that database have the bug fix
for the issue ?
Yeah, 55 days seems serious, if the database is affected by the bug, you can
consume 55 days in a matter of days (if not hours).
Cheers
Riyaj Shamsudeen
Principal DBA,
Ora!nternals - http://www.orainternals.com - Specialists in Performance, RAC ;
and EBS
Blog: http://orainternals.wordpress.com/
Oracle ACE Director and OakTable member
Co-author of the books: Expert Oracle Practices, Pro Oracle SQL, Expert RAC
Practices 12c. Expert PL/SQL practices
On Wed, May 25, 2016 at 9:15 PM, Deepak Sharma <dmarc-noreply@xxxxxxxxxxxxx>
wrote:
Just recently (last week) we came across an alert from OEM for one of our
primary databases, related to SCN Jumping or increasing very rapidly.
We already have an Oracle SR in place, but wanted to check with the community
about their thoughts in case someone has had similar experience in the past
(and any solution).
This is an Oracle 11.2.0.4 database (1) using RMAN backup, and (2) has several
connections coming over using database links to it (research shows the issue
could be db link related).
Some of the other DBs that have DB Links to ours, see messages related to
sudden increase of SCNs for their own DBs, in alertlog (fyi).
The OEM alert seem pretty serious, and says we have about 55 days before the DB
would freeze temporarily or throw ORA-600 errors.
Could be related to this :
https://community.oracle.com/thread/2336601?tstart=0
