"ERROR: recovery is in progress" from Standby node

Hello,
We have a pgpool 3 nodes cluster (1x Primary, 2x Standby) running PG13. We recently deployed PMM2 (2.34), and everything seems to be working fine. However on one of the standby node (and only this one!), we are getting the following error repeatedly:

time=[2023-02-13 14:00:35 UTC], pid=[2610745]  db=[postgres], usr=[monitoring], client=[xxx.yyy.zzz] , app=[[unknown]], line=[1] ERROR:  recovery is in progress
time=[2023-02-13 14:00:35 UTC], pid=[2610745]  db=[postgres], usr=[monitoring], client=[xxx.yyy.zzz] , app=[[unknown]], line=[2] HINT:  WAL control functions cannot be executed during recovery.
time=[2023-02-13 14:00:35 UTC], pid=[2610745]  db=[postgres], usr=[monitoring], client=[xxx.yyy.zzz] , app=[[unknown]], line=[3] STATEMENT:
                                SELECT slot_name, database, active, pg_wal_lsn_diff(pg_current_wal_lsn(), restart_lsn)
                                FROM pg_replication_slots

NB: It is possible that this node was a Primary when it was configured and registered with the PMM server, and that a failover occurred afterwards, but unfortunately I have no artefacts of this event.

My question is: is there any way to turn off this feature? (Ie the replication delay check)

As a bonus question: Is PMM “cluster aware” , especially when a postgresql instance role changes from Standby to Primary, or vice-versa?

Cheers,
Steeve