Confusing log message when the wrong connection holds on to the master server lock

In cooperative_monitoring, if another connection holds the master-lock on the current primary server, the log message states:

Cannot acquire lock ‘maxscale_mariadbmonitor_master’ on <monitor> as it’s claimed by another connection (id X). Secondary MaxScales may select the wrong primary.

This is misleading, as the lock is still held on the correct server. A similar message should instead be printed if the lock is held on the wrong server.