We were profiling our log files today and I noticed we have a lot of limit 1, sorted by _id

I think this is because we started using code similar to:

u = User.where(email: '...').first

When we really just wanted to say "we're not sure if the user exists with that email, so don't raise a not found error". This is also why we don't use find_by, because it's important for us to have raise_not_found_error left at the default value of true, so we can use that where it's needed. We have a unique index on email, so it's not like the sort is doing crazy tons of extra work, but I believe it is causing Mongo to use a different index than the default one (the index on email).

Not a huge deal, but I was just curious if profiling had ever been done to see if those extra _id sorts where causing any extra work, or if there was a better way to accomplish what we were after. Thanks!