DXRE Use of Event Numbers and Workgroups in Replications
If you are running NCR Counterpoint SQL, then you probably know about using workgroups. Today, I want to talk about workgroups in a multi-site environment, in particular.
In a multi-site environment, where you have multiple stores, and are synchronizing using DXRE replication, the proper use of workgroups is very important. In this case, each location needs to have a separate workgroup, and all posting at the location must be done using that workgroup. Otherwise, replication may begin failing.
For example, let’s look at the case where your hub is operating under the workgroup "MAIN", and you have one store that operates under the workgroup "1". In the workgroup, one thing that is tracked is the "next event number". All posting operations have an event number associated with them, and this is where that number comes from.
Let’s say the "MAIN" workgroup has a "next event number" of 123, and you post anything (tickets, purchase orders, transfers, etc.), the event number 123 will be assigned to that posting. For our purposes, let’s say that at the hub, an inventory transfer is posted, so it will get event number 123 assigned to it. Now, if the Store posts tickets, for example, using the "MAIN’ workgroup instead of workgroup "1" as it should before the data is replicated, that posting will also be assigned event number 123.
This is where the problem arises. When the next replication is run, one side will try to replicate event number 123 as an inventory posting, and the other will try to replicate it as a ticket posting. In that case replication will fail, and will continue to fail until one of the posting events is moved to a new, and unique, event number. Cleaning this up requires directly changing the data in the SQL database, and can be a very time consuming process, depending on what type of posting is involved.
For more information on how to plan and use your Workgroups correctly to ensure trouble-free replication, contact CCS Retail Systems Support.