Replication

Cascaded standby

Also called: downstream standby, tier-2 standby

A cascaded standby is a replica that streams its WAL from another standby rather than from the primary. Its primary_conninfo points at an upstream standby (which must have hot_standby = on and spare WAL sender slots), and it is read-only like any standby.

What this means

A replica that gets its WAL from another standby rather than from the primary. Its connection points at an upstream standby instead, and it's still read-only like any standby. It's the downstream node in a cascading setup.

Why it matters operationally

Cascaded standbys are how you build replication tiers, but they carry one hard limitation: they can never be synchronous standbys, because the primary only tracks and waits on its directly connected children. Plan synchronous replicas as direct children of the primary.

Related & next

Cascading replication and delayed standbys explainedCascading replication, the topology it formsReplication lag, which can accumulate per tier

Mentioned from

← All glossary terms · GUC reference · Error catalog