I run one publisher on the main server and I have 4 subscribers over the WAN. I am using merge replication because I need the data replicated to and from the subscribers real time.
I also run the distributor on the same server as my publisher.
My Database backup file is 10GB as of now.
My snapshot always contains all tables.
1. When I am to run replication, I normally take a backup of the main database where the publisher seats and then restore it at the subscriber and then run replication. Is this the right way?
2. When a run replication it takes over 48 hrs to finish. Kindly advise if there is a better way to reduce on the time.
In my view I think that it might be attributed to slow internet or a big logfile.
3. At times the snapshots at some subscribes becomes invalid which call for re-initialization or recreating the replication. Users at...