Another thing I noticed in v34 is that, due to the change in show_effective_wal_level(), a standby now reports effective_wal_level = 'logical' when wal_level = 'logical' is set on the standby, even if the primary has effective_wal_level='replica'. Is this change in behavior intentional?
It seems to contradict the documented behavior as well: + On standby servers, <varname>effective_wal_level</varname> matches + the value of <varname>effective_wal_level</varname> from the most + upstream server in the replication chain. ~~ I see that XLogLogicalInfoXactCache is removed, but it is still referenced at few places: +extern PGDLLEXPORT int XLogLogicalInfoXactCache; +/* + * A process local cache of LogicalDecodingCtl->xlog_logical_info. This is + * initialized at process startup time, and could be updated when absorbing + * the process barrier signal in ProcessBarrierUpdateXLogLogicalInfo(). + * See the comments of XLogLogicalInfoXactCache too for details of how this + * cache value is used within a transaction. + */ thanks Shveta
