Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Oracle multi-master replicaiton: index generation necessary?

Re: Oracle multi-master replicaiton: index generation necessary?

From: Nuno Souto <nsouto_at_optushome.com.au.nospam>
Date: Mon, 26 Aug 2002 21:10:48 +1000
Message-ID: <3d6a0eb0$0$28864$afc38c87@news.optusnet.com.au>


In article <3tka9.15283$g9.48096_at_newsfeeds.bigpond.com>, you said (and I quote):
>
> I don't believe you *must* replicate indexes (although you have the option
> to specify which indexes you might wish to add to a replication group). The
> exception to this is the PK which all replicated tables must have else
> Oracle will turn on you angrily.

AFAIK, indexes are not replicated. What is replicated (sent back and forwards) is the contents of a row (columns) with an instruction that says "delete, insert or update this" depending on the original op and the row's PK. It's all done by triggers that fire off packaged procedures. The column values are passed around, not the data blocks.

So, it doesn't matter what indexes you have anywhere, the INSERT/UPDATE/DELETE statement fired off by the local replication code will do whatever needs to be done to the local indexes just like any other statement. Transparently. Like you say, it's good practice to have a PK for these tables, but only because Oracle will chuck a wobblie if it can't find the row during an update or delete.

HTH

-- 
Cheers
Nuno Souto
nsouto_at_optushome.com.au.nospam
Received on Mon Aug 26 2002 - 06:10:48 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US