Adds a section to the source-postgres docs discussing replication slot confirmed_flush_lsn vs restart_lsn, what happens if the slot stops advancing due to capture downtime or long-running transactions, how the max_slot_wal_keep_size setting can be used to protect a production database from running out of disk space, and what the potential downsides are of doing that.
Description:
Adds a section to the
source-postgres
docs discussing replication slotconfirmed_flush_lsn
vsrestart_lsn
, what happens if the slot stops advancing due to capture downtime or long-running transactions, how themax_slot_wal_keep_size
setting can be used to protect a production database from running out of disk space, and what the potential downsides are of doing that.This change isβ![Reviewable](https://reviewable.io/review_button.svg)