Changing transport configuration results in restarting wrong cluster

Steps to reproduce:
1) Use the attached configuration files for the server (this will start two server groups with port offset)
2) Go to the first cache container -> configuration -> Transport tab
3) change "channel" to empty string
4) Confirm that you want to restart now
5) It restarts second-server-group, not main-server-group as requested. The state of invidivudla servers in main-server-group remains "reload-required"