SQL Server 2005 Replication

Question

SQL Server 2005 Replication

ravivarman.m

SSC Rookie

Points: 43
More actions
December 29, 2008 at 5:00 am

#383648

While using Pull subscription, How to synchronize without applying the Snapshot Scripts at very first time?

Viewing 8 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic. Login to reply

Manish Khattar Old Hand Points: 399 More actions · Answer 1

If I have not misunderstood you do not want to transfer initial data new Subscriber(s). There are two ways to achieve this:

1. SQL Server Management Studio:

You can use New Subscription Wizard and on the Initialize Subscription screen uncheck the Initialize checkbox. This is possible in both Transactional and Merge Replication.

2. Replication Stored procedure:

Merge Replication:

exec sp_addmergesubscription @publication = N' ', @subscription_type = N'pull', @subscriber_type = N'local', @subscription_priority = 0, @sync_type = N'None'

@sync_type should be 'none' as mentioned by default it is 'automatic'

If automatic, the schema and initial data for published tables are transferred to the Subscriber first. If none, it is assumed the Subscriber already has the schema and initial data for published tables. System tables and data are always transferred.

Microsoft do not recommend specifying a value of none.

Transactional Replication:

exec sp_addsubscription @publication = N' ', @sync_type = N'replication support only', @subscription_type = N'pull', @update_mode = N'read only'

@sync_type = N'replication support only'.

I hope this serves your purpose.

thanks

manish khattar

ravivarman.m SSC Rookie Points: 43 More actions · Answer 2

Thanks for Your Reply.

I have done the following steps. but still its applying the entire schemas from the publisher snapshots.

step 1. Published the DataBase in Publisher and backup that database

step 2. Restore that in the subscriber

step 3. Creates the Subscription with do not initialize option.

step 4. When synchronizing its applying the entire schemas from the publisher.

Thanks,

Ravivarman

Manish Khattar Old Hand Points: 399 More actions · Answer 3

It must have generated the snapshot but it will only apply system objects and metadata required by replication.

r, manish

ravivarman.m SSC Rookie Points: 43 More actions · Answer 4

Thanks for spending time and Reply my queries.

In this Reply i have attached the Screen shots for your Reference,

1) I restored the database using the backup of published database

2) i created the pull subscription with do not initialize option.

3) While synchronizing its applying the all schemas of user created tables.

Please see the attachment...

Thanks,

Ravivarman

chileu17 Hall of Fame Points: 3093 More actions · Answer 5

have you tried using in the precreation cmd 'none'?????

but you should have in mind that, you would have first to synchronize your metadata on both servers and then you wouldn't need the initial synchronization 😉

ravivarman.m SSC Rookie Points: 43 More actions · Answer 6

Thanks for your response.

i have tried with that precreation cmd 'none'.

But i am not sure about that.

please guide me what i have to do...

Thanks.

Ravivarman

chileu17 Hall of Fame Points: 3093 More actions · Answer 7

ravivarman.m (12/31/2008)
Thanks for your response.
i have tried with that precreation cmd 'none'.
But i am not sure about that.
please guide me what i have to do...
Thanks.
Ravivarman

If you are configuring this from the GUI then you have to select the publication, then properties, and on the window that appers you should select the articles page, highlight an article (table), then click on the button article properties and there is a property that says "Action if name is in use" under Destination Object properties. There you can select the value "Keep existing object unchanged" that is actually the value none that appears when you are using scripts.

The thing I told you about that you should have the same metadata in both servers, I meant that your tables must have the same structure. You know that when decide to do a merge replication, all your publicated tables get added a ROWGUID column in your publisher server, if you don't wanna use initial synchronization, then you have to create such columns in your subscribing server as well manually. If not you will get some nasty errors which I don't wish them to anybody not even to my worse enemy lol

Good luck. Hope I could have helped you.