I am currently planning out how to best handle separating out client data. The most obvious answer is to pay for several databases, but I was curious if there was some way to have one database and somehow have different tables to separate data. The database is only open to the AWS Gateway so in theory with proper precautions I control the queries and show table data to only those authorized to see their tables. It would seem I could change the name of the edge gateway, which would trigger a new table, but is there any other way to accomplish this or is this the only way to create new tables?
Another scenario I can imagine is if say we wanted to separate data for a location change. This would be a good reason for separating a table.
If your UI has suitable code to filter data by client, there is really no need. (That is, not relying on any browser input to control filtering.)
If you are trying to do this so different clients can access different projects in the designer, then you will be disappointed. The designer's ability to push code into gateway scope means any designer access on any project can get to anything else in the gateway. Fine-grained rules for designer privileges are guard rails, not cages.
Aren't the Schemas for Tag Historian locked down? If I could add my own columns that would be great, but I was under the impression that this wasn't possible.
The cost of a Amazon database is so low we agreed that we would get separate databases for clients. I'm in the process of designing my own tables to organize everything. I wanted some kind of initial landing page that had a table listing all of the client's different sites (edge clients) then they could drill down into the data. But I also wanted to be able for my company to have a master list to see every edge client so we could quickly locate the data we needed. I realized, this morning, I need another master database that contains all of these sites for every client. Otherwise we would have to query every database for their site list.
I think with this much separation of data, my original need for separating tables isn't necessary, but I am still curious if there is someway to have Ignition end a sqlt_data table other than renaming an edge device.
If you look at the documentation for these tables:
you see that the driver ID represents a unique combination of source gateway name and source gateway's source tag provider name. You won't get a new table without changing one of those components.