The Clustrix monitor autonomously detects the nodes in a cluster and populates services referring to it with the corresponding server objects. Sometimes you want more direct control over what nodes (and ports) should be used. The following should be possible:

[Node-1]

type=server

address=...

port=...

protocol=mariadbbackend

[Node-2]

...

[Node-3]

...

[TheMonitor]

type=monitor

module=clustrixmon

servers=Node-1, Node-2, Node-3

...

dynamic_node_detection=false

[TheService]

type=service

router=readconnroute

...

cluster=TheMonitor

The dynamic_node_detection=false in the monitor section would have the effect that the monitor does not dynamically figure out the structure of the Cluster, but simply uses the servers it has been configured with as such.

That is, the servers the Clustrix monitor has been configured with are not used as bootstrap server from which the Cluster configuration is queried, but as the very nodes the monitor should make available to services.