Hi there,
You could check the settings of the distributed and auto-discovery attributes in the properties.xml file.
If those are both set to true then it looks like the multicast UDP traffic is being blocked somehow. Some routers (eg Cisco) disable it by default. I've migrated a couple of customers to the Amazon Cloud (AWS) where multicast isn't allowed at all. Instead we use JDBCPing to do much the same thing using the database to discover which nodes are present, and it uses the Beacon/Client port.
JDBC Ping relies on having ephemeral ports open. The following commands in a powershell window opened as administrator will enable things. They should be executed on all servers in the cluster.
New-NetFirewallRule -DisplayName 'Clarity (TCP Ephemeral)' -Enabled True -Direction Inbound -Action Allow -LocalPort 49152-65535 -Protocol TCP
New-NetFirewallRule -DisplayName 'Clarity (UDP Ephemeral)' -Enabled True -Direction Inbound -Action Allow -LocalPort 49152-65535 -Protocol UDP
New-NetFirewallRule -DisplayName 'Clarity (TCP)' -Enabled True -Direction Inbound -Action Allow -LocalPort 9090-9091 -Protocol TCP
New-NetFirewallRule -DisplayName 'Clarity (WEB)' -Enabled True -Direction Inbound -Action Allow -LocalPort 80,8080,8090,8093,443 -Protocol TCP
After this it’s just a matter of ensuring that the distributed attribute is set to true, adding in the useJDBCPing=”true” field into the properties file in the <nsa> section, and setting autoDiscovery=”true”.
Good luck.
Paul