Database Redundancy

I’d be interested to know how people have handled redundancy with respect to a database (specifically MS SQL Server).

We’re looking at a project where the customer has asked for “fully redundant servers so no data is lost”.

Now obviously there is a redundancy option for Ignition itself but what that doesn’t really cover is what to do in terms of the database.

As well as SQLtag history I could quite feasibly have transaction groups plus inserts/updates/deletes and stored procedures within scripting as well.

What’s the best way to cope with the database failing?

database replication is outside of Ignition. I have tried using MySQL replication, but it is flaky at best and needs constant babysitting. Sort of hard to do when you set up an application in a remote plant with no IT support.
I believe MS SQL Server may be more robust in replication, but I have not tried it yet.