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.
Author |
Topic |
sameerchachad
Starting Member
1 Post |
Posted - 2010-08-06 : 04:54:16
|
We have a scenario for Merge replication - MS SQL Server 2005 and client application for Sync up on .NET 3.5 Framework, where the system table MSmerge_current_partition_mappings is getting duplicate records for each client after a Sync is performed from a .NET Application. Thus whenever there is a new client sync-up, rows are getting appended along with the existing rows already present in the table. Currently there are more than 2000 clients doing a sync-up everyday. Due to this there are more than million records existing in this table. This can lead to a performance downgrade. The User tables of the application are not facing any duplicate records problem. Its only the system tables where the records are duplicated. Insertion of records in this table is a process handled internally by the Publication mechanism of MS SQL Server. There is hardly any coding involved as its handled internally by MS SQL Server. The findings are - When we did a Sync for Client-2, duplicate records for the first client were also present. This should be avoided. We were thinking of writing some Delete Scripts of deleting data from this system table, but there are other System Tables also involved when a Replication takes place. Some of them are - MSmerge_contents MSmerge_tombstone MSmerge_genhistory, etc.Kindly look into the same.Thanks in advance for the solution.Looking forward to it.Thanks and warm regards,Sameer Chachad.sameer.chachad@patni.comsameerchachad@gmail.com |
|
|
|
|