Going off on a tangent here, but the HA software probably sees the trigger
but turns it off. The trigger is still there on the backup machine but is
set to not execute. There's good reason for that. If a trigger on FileA
updates FileB but both FileA and FileB are replicated to the backup
machine you're basically confusing the heck out of FileB on the backup
machine. Or, if a trigger on FileA sends an email off to your customer do
you want that same trigger to execute on FileA on the backup machine, so
the customer get's two emails? Most HA software has an option to do this.
And their switch process turns these back on.