Merge Replication of large database

Hi,
I am setting up a merge replication of a large database across internet
between two servers which are located in two geographically different
location and can connect to each other with a fast T1 Connection.
The purpose of it is to have a hot backup site available for the DB.
The problem i am facing is that the DB size before setting up replication is
2GB.
When I set up merge replication and selected snapshot to create , it grow
the DB size to 6GB. Most of the increase in Trasaction log.
Why?
I will apply the backup of these Snapshot generated DB to the subscriber
manually first and then set up the pull subscription on the subscriber.
Is this the right way to handle this situation?
How can I reduce the size of db after activating the replication?
I tried it with shrinkdatabase command but it does not shing the log file
much. What transactions are open that it does not shrink?
Please advice any better solution.
Sam
Backup the transaction log and then issue a shrink. This should correct the
problem if you are using the full recover model.
Also issue a dbcc opentran to see if there are any open transactions. Use
sp_who2 to evaluate whether the spid is orphaned and if you can kill it.
Hilary Cotter
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
"sam" <sam@.discussions.microsoft.com> wrote in message
news:8D35DB00-6FF6-4B75-AE4B-3A6E4EF8322B@.microsoft.com...
> Hi,
> I am setting up a merge replication of a large database across internet
> between two servers which are located in two geographically different
> location and can connect to each other with a fast T1 Connection.
> The purpose of it is to have a hot backup site available for the DB.
> The problem i am facing is that the DB size before setting up replication
is
> 2GB.
> When I set up merge replication and selected snapshot to create , it grow
> the DB size to 6GB. Most of the increase in Trasaction log.
> Why?
> I will apply the backup of these Snapshot generated DB to the subscriber
> manually first and then set up the pull subscription on the subscriber.
> Is this the right way to handle this situation?
> How can I reduce the size of db after activating the replication?
> I tried it with shrinkdatabase command but it does not shing the log file
> much. What transactions are open that it does not shrink?
> Please advice any better solution.
> Sam
>