In some cases it is possible for PostgreSQL replication slots [1] to exist without having the changes they track consumed.

This can happen if a HA standby server fails and is not brought back online or if a global region database is not responding.

This is fairly simple to check for by inspecting the pg_replication_slots view's "active" column [2]. I'm suggesting that ManageIQ/CFME implement some monitoring for inactive replication slots and raise an event/notification if they persist for some set amount of time.

[1] https://www.postgresql.org/docs/9.5/static/warm-standby.html#STREAMING-REPLICATION-SLOTS
[2] https://www.postgresql.org/docs/9.5/static/view-pg-replication-slots.html