May 15, 2009 at 11:20 pm
I want something like this....
1. Live DB working busy on processing incoming data
2. Replication DB which replicate data from Live DB
3. Replication DB need to process data and then pass back to Live DB so that both have the same data
So how i can do it... Please help and need it urgently..
thanks
jason
May 18, 2009 at 8:03 am
Go for merge replication.
Only thing you need to check for application as it will add one extra unique row so that it can sync both live and replicated DB.
May 18, 2009 at 11:39 am
Or peer-to-peer
* Noel
May 18, 2009 at 1:04 pm
perhaps immediately updating subscribers or queued updating
http://msdn.microsoft.com/en-us/library/ms151718(SQL.90).aspx
---------------------------------------------------------------------
May 19, 2009 at 6:35 am
The problem with setting up replication is that it often requires you to make changes to your database design. Transactional replication requires a primary key on all tables. Also peer-to-peer replication is only in the Enterprise Edition available. I woudn't use transactional replication with updating subscribers if you plan to make updates on a regular basis at the subscriber database.
Merge replication on the other hand will add a rowguid column and triggers to the tables. You have to check if that will break your application. Compared to peer-to-peer replication it has the advantage of automatic conflict resolution.
P2P has conflict detection in SQL 2008, but by default after a conflict is detected the agent will stop synchronizing.
See also this article:
http://qa.sqlservercentral.com/redirect/articles/66970/
[font="Verdana"]Markus Bohse[/font]
May 19, 2009 at 8:25 am
So what method you suggested ?? and is the best for my situation
May 19, 2009 at 10:24 am
jason (5/19/2009)
So what method you suggested ?? and is the best for my situation
My first thought would be merge replication, but to be honest you didn't provide a lot of detail and I won't know if your application can handle the extra column or not.
Can you make changes to the application or is it a thrid-party product which you can't change.
I suggest you read some more articles about replication and see what fits best for your situationm before making the final decision.
[font="Verdana"]Markus Bohse[/font]
May 19, 2009 at 6:17 pm
Ok let me give some more detail and hope you guys can give me a good suggestion....
Live DB A => accept data from Web , handphone and pda
Live DB B => replicate Live DB A and process some data inside some of the table and then it need to replicate back to Live DB A
At the mean time, data inside Live DB A may be will be updated or deleted while the data is being process in Live DB B..
So do you think of the solution that suitable for this situation....
thanks in advance for the reply....
May 24, 2009 at 9:48 pm
At the mean time, data inside Live DB A may be will be updated or deleted while the data is being process in Live DB B..
The changes will be reflected depending on who is the conflict owner.
From your scenario DB B will process the data and replicate it to the DB A meantime if the same data in the same row have been modified then there will be a conflict and the depending on the conflict resolver type which you had setup during the replication will be the owner.
I think first you go through some links and read on replication types and How to resolve conflict.
For more information on merge replication & How it resolve conflict please read the below links.
http://technet.microsoft.com/en-us/library/ms151215.aspx
http://technet.microsoft.com/en-us/library/ms151257.aspx
"More Green More Oxygen !! Plant a tree today"
Viewing 9 posts - 1 through 8 (of 8 total)
You must be logged in to reply to this topic. Login to reply