Re: ** replication delay analysis

Mohammed : Thanks for your help.
We do not make direct updates to one of the site and it has the push job broken. So effectively it is one-way. I am looking at the replication administration topology map (which it takes 6-7 minutes to refresh) screen of OEM and see the number of transactions going from primary to secondary. This is same number as defcall/deftran on primary. So I am still trying to find out a way to know if the push job is the bottle neck or it is the execution on secondary or some other problem. Any help is appreciated. I do not know if the 6-7 minutes to get info is due to replication tables being in 'SYSTEM' tablespace with small extents. I would like to move them to another tablespace. Has anyone done that? Any issues or precautions to take? Thanks  

Thanks

mkb <mkb125_at_yahoo.com> wrote:
--- A Joshi wrote:
> I am trying to analyze replication delay. It is
> advanced two way replication implemented one-way
> only so data goes from source to target.

Wait a minute. How can it be advanced two way replication if it is only implemented one-way?

> Is there a way to know how much deftran has reached
> the target and waiting to be executed. Currently the
> deftran at source is increasing and is that
> indication that source is the problem?

Not the best tool to use, but try using OEM to see if the trxns have been pushed to your target. You may also see them on the replication map in OEM.

Could also be a huge trxn that took place at the source. I've seen replication get stuck when you try mass deletes/updates (millions of rows). Check to see if defcall is increasing.

--
mohammed




__________________________________ 
Celebrate Yahoo!'s 10th Birthday! 
Yahoo! Netrospective: 100 Moments of the Web 
http://birthday.yahoo.com/netrospective/

		
---------------------------------
Celebrate Yahoo!'s 10th Birthday! 
 Yahoo! Netrospective: 100 Moments of the Web 

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