June 6, 2005 at 2:07 pm
I have a database I'm trying to get replication to function properly on.
The destination database does get the full database snapshot but the segregation across various filegroups isn't acknowledged and all data is punched into a single file that is expanded as required.
Options doesn't have anything that resembles relative to filegroups.
I spindleize (technical term) the database across various RAID 10 and RAID 1 units. Database isn't too large (50 - 100GB) depending on whether you've rebuilt indexes recently.
June 7, 2005 at 8:54 am
One option is to create the database on your subscriber yourself with the tables/indexes spread across filegroups as you wish. Then create your publication with options that state the subscriber already has the proper schema.
Alternativley, you can add each article with your own custom script that includes the filegroup clause .
look at the options for sp_addarticle or sp_addmergearticle
@pre_creation_cmd
@creation_script
@schema_option
June 8, 2005 at 9:14 am
I am currently reviewing your recommendations.
My initial plan was to create a one time Snapshot publication that was basically a throw away after it was created on the rep server and that's when the trouble started. The database is part of an application (client-server) at its best where none of the tables have Primary Keys... the majority have clustered indexes and that's as close as it gets. Someone told me all of the tables had to have Primary Keys but I do not believe that. The same scenario occurred using DTS to copy the database from one server to another; on the destination the import was defaulted to the PRIMARY filegroup...
After the snapshot I planned to set up the Transaction publication to push or pull subscriptions hourly or daily (whichever impacted the production system less). I see problems using the EM with this as it doesn't allow me to mark the articles (approx 150) to include in the publication.... it allows Stored Procedures but not tables... don't understand why but never gave it much time as I moved on to other options.
I have currently grabbed a backup that I'm restoring on the replication server so I will have 2 identical databases (point in time mirror). This is fortunately on a test system... I want to get this right before I move it into production.
If someone has an EASY button, now's the time to share.
Thanks
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply