Merge Replication of table with auto indexed field

Last Post 03 Jun 2005 10:42 AM by jcm01. 3 Replies.
AddThis - Bookmarking and Sharing Button Printer Friendly
  •  
  •  
  •  
  •  
  •  
Sort:
PrevPrev NextNext
You are not authorized to post a reply.
Author Messages
jcm01
New Member
New Member

--
03 Jun 2005 07:42 AM
I want to merge replicate a db that contains a table that has an auto incremented "index" field. I want inserts in each of the tables to be able to occur asynchronusly and independent of each other. However, due to the auto incremented "index" field, will each db be forced to retrieve the max index value from the other instance of the db before it can determine the value of its own next "index"? The auto incremented "index" field is NOT the primary key. Will the auto incremented "index" field effectively destroy the asychronus-ness of the merge replication?
jcm01
New Member
New Member

--
03 Jun 2005 09:07 AM
Yes, I guess I am refering to an identity column. Column Name is "[index]", Data Type is int, Length is 4, Identity is "yes", Identity Seed is 1, and Identity Increment is 1.
jcm01
New Member
New Member

--
03 Jun 2005 10:36 AM
That's an interesting idea. Thanks.
I assume then that if I change nothing I will at best have performance issues and at worst the scheme may not work at all. Yes?
jcm01
New Member
New Member

--
03 Jun 2005 10:42 AM
But what about this: Even if one db is incrementing even and one is incrementing odd, how would SQL Server know not to bother checking one db when incrementing the other?

The Identity value for a field can be set to "Yes (Not For Rreplication)". What's that all about?
You are not authorized to post a reply.

Acceptable Use Policy