February 11, 2010 at 3:59 am
We are replicating a table from publisher to subscriber.
The table has a unique index, and i have confirmed this in the scheema file.
Over the weekend a number of updates were done to the source table, and the article was reinitialized.
Article was dropped, subscription was dropped, updates were done, article and subscription were recreated and snapshot agent was kicked off.
The data is now duplicated in the destination subscriber table. As there is a unique index on the table i cannot see how dups got into it.
Does replication do something strange like not apply indexes until after population via snapshot? this is the only way i can see dups getting in there, that or someone modifying the scheema files.
February 15, 2010 at 4:44 pm
what do you mean by a unique index? do you mean a unique key? show us the table schema and an example of a row that is duplicated. It could be that one of the publication settings is set to ignore the key.
February 16, 2010 at 9:20 am
Your index only exists in your source Table it will not get propagated by the replication. when you initialise your subscription check that the Destination table get droped then recreated.
The only way you will get duplicated data is if the destination table is not being dropped or someone is adding data to the destination table not the source.
An other option could be if you are merging 2 tables as a source to go into one Table in the destionation then check what the replication SP are doing.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply