Triggers are driving me crazy

Vagrant
1

Surely I’m not a Ruby expert, but it seems triggers are driving me crazy. Apparently they seem to be a simple concept, still I can’t make them work.

I have a Vagrantfile for configuring a multi-machine setup, like this:

Vagrant.configure(2) do |config| 

        # Config parameters for the environments
	    number_of_nodes = 6
	    cpus_per_node = 4
	    mem_per_node = 4096
        master_node_name = "portainer"
        worker_node_name = "pworker"

        # Provision the nodes.
	    (1..number_of_nodes).each do |i|
               # First node is the master
		       if i == 1 then
			        config.vm.define "#{master_node_name}", primary: true do |master|
				        master.vm.box = "almalinux/9"
				        master.vm.hostname = 'master'
				        master.vm.provider "hyperv" do |hv|
					        hv.enable_virtualization_extensions = true
    				        hv.linked_clone = true
					        hv.vmname = "Portainer"	
					        hv.cpus = cpus_per_node
					        hv.memory = mem_per_node
				        end
                
				        # Define a reload trigger.
				        master.trigger.before :reload do |trigger|
					        trigger.info = "Setting Hyper-V switch to 'LabSwitch' to allow for static IP..."
					        trigger.run = {
						        privileged: "true", powershell_elevated_interactive: "true", path: "./scripts/powershell/set-hyperv-switch.ps1", args: "Portainer"
					        } 
				        end

				        # Provision some stuff.
				
				        # Reload the machine.
				        master.vm.provision :reload
                    end
                else
			            config.vm.define "#{worker_node_name}#{i - 1}" do |worker|
				            worker.vm.box = "almalinux/9"
				            worker.vm.hostname = "worker#{i - 1}"
				            worker.vm.provider "hyperv" do |hv|
					            hv.enable_virtualization_extensions = true
    				                    hv.linked_clone = true
					            hv.vmname = "Portainer Worker #{i - 1}"
					            hv.cpus = cpus_per_node
					            hv.memory = mem_per_node
				            end

				            # Define a reload trigger.
				            worker.trigger.before :reload do |trigger|
					            trigger.info = "Setting Hyper-V switch to 'LabSwitch' to allow for static IP..."
					            trigger.run = {
						            privileged: "true", powershell_elevated_interactive: "true", path: "./scripts/powershell/set-hyperv-switch.ps1", args: "Portainer Worker #{i - 1}"
					            }
				            end
                            # Reload the machine.
			                worker.vm.provision :reload
                        end
                end
        end
end

As you can see, there are two “guest scoped” triggers: thery’re the same, the only thing that changes is the parameter that the trigger takes. In the first machine, the trigger works fine.
In the second machine (the else) when the directive worker.vm.provision :reload is fired, the executed trigger seems to be the first one, and not the second.

Any help / idea on why this is happening?