NetLD assumes a fairly vanilla install of CentOS/RedHat in which IPv6 is enabled by default, even if IPv6 addresses are not explicitly used.  During discovery NetLD binds the UDP socket to "::", which is the IPv6 equivalent of the IPv4 "", also known as anylocal.  In dual-stack systems like Linux and Windows, IPv4 traffic is typically also received on IPv6 bound ports, so this allows NetLD to simultaneously support both IPv4 and IPv6 discovery.


However, on Linux when IPv6 is disabled at the network configuration level, as opposed to being disabled at the kernel module level, then Linux allows the bind IPv6 bind (without error) but silently dumps the SNMP packets.  This may be the cause of the SNMP queries not returning.


It seems likely that IPv6 might be disabled on that server.  Luckily, there is a safety hatch.  Can you edit the /usr/share/netld/config/config.ini file and add the line anywhere (at the top is most convenient):




After that, restart the service:


  root$ service restart netld


 and then try the discovery again