Data health checks
No due date
50% complete
we're holding significant amounts of data.
it's not feasible to manually review data sanity
find a way to:
- get known good states of databases and save them somewhere (github?)
- check for data equivalency between the latest known good and running DBs periodically