this is the logic in the cluster.sh - we decided it would be better to restart because everything is changed from the point of view of the slave node - new configuration received from the master - and is not currently possible to see if only part that allows reloading has changed.
this is the logic in the cluster.sh - we decided it would be better to restart because everything is changed from the point of view of the slave node - new configuration received from the master - and is not currently possible to see if only part that allows reloading has changed.
Not sure what can be done in this case :(