Please start any new threads on our new site at https://forums.sqlteam.com. We've got lots of great SQL Server experts to answer whatever question you can come up with.

All Forums  SQL Server 2008 Forums  Replication (2008)  Merge Replication with Identity Columns

Author  Topic  Hotsdogs Starting Member 1 Post Posted - 2012-03-01 : 16:53:39 I am using merge replication with SQL2008 Standard as the publisher and replicating to 9 servers using Sql Express. One for the tables has an identity column that is no a primary key but was added to give our customers an integer number for each transaction we perform for them. They did not like the GUIDs. When I am setting up the publisher in a test scenario with two subscribers I set the properties of the identity column Range management properties in the publication articles to have a publisher range size of 1000000 and a subscriber range size of 1000000 and a Range Threshold percentage of 80. When I synchronize the subscribers and start to update them I an noticing the identity values in each of the subscribers is started at the last value in the publisher + 5000000. So if the current identity value in the publisher is 250000 when the snapshot is created, the next row in subscriber A is 5250000 and in subscriber B is 10250000. I would have expected 1250000 or 1000000 in subscriber A and 2250000 or 2000000 in subscriber B. Where are these values coming from? It causes no problem so far but the magic going on is a little concerning. We will never get near the 2 billion limit with our 9 Production machines but I guess I an not understanding how the ranges are working.