What is the performance impact of doing pg:pull on a Heroku app?

Question

I'm interested in pulling down a 15MB database from an app that is in production but am concerned about the impact it may have on end users. Is this something that can be done without fear or best scheduled at 4am?

Answer 1

A better idea would be to do a Postgres snapshot on Heroku and then download that separately. There is great documentation on the PG Backups add-on .

If you're really worried about load on the DB, I'd suggest creating a follower database and then download a snapshot of that, as the replication will pose less risk to your master production database.

This will cost some money, by provisioning another production-level database, but is only required for the time needed to do the backup. Unless you have a massive amount of data, we're likely talking only a few pennies to provision, have the new production database follow your master, do the backup, then delete the follower database.