Skip to content

Comment on Globally Distributed Postgresparent

Comments

Yeah, if you are doing global replication you are much more likely to fall behind so you should have aggressive monitoring on the replica to remove it if replica lag gets to high, that will prevent the primary from retaining WAL files for too long.

It also means you will need to be able to resync the replica from a backup once write volume calms down.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.