Re: Using Ethernet to communicate with equipment


g4wjs
 

On 17/07/2021 13:51, Dave / NR1DX wrote:
All

I am struggling to move to the "next level" of passing information from DXLab to various pieces of hardware via an Ethernet connection.

My problem is not with DXLab per-se but rather that much of the documentation (from  both  DXLab and the hardware manufacturers) assumes the user knows might actually know something about Ethernet communications to begin with.... which I dont . I dont even know what half the terms mean let alone the intricacies of how to use them ....help

Can someone direct me to a site that is HAMRADIO ETHERNET FOR DUMMIES. What is a UDP?  what is an IP address and when and why does it need to be static and when does in need to be dynamic. What is a NET MASK and and why do I care? etc etc etc.

Dave
NR1DX
Hi Dave,

I'll have a go with some analogies that may help.

UDP is like sending a letter, fire and forget. You put a message (datagram) into an envelope (IP packet) and send it to a already known address. It may or may not arrive, but usually does. There is another protocol called TCP/IP which is more like a telephone call, a stream of information is passed back and forth between two parties as a conversation. As with UDP the data is sent in envelopes (IP packets).

IP addresses are like premises addresses, you have to know them before you can send something to them. A static IP address is one way a destination gets a unique IP address and it is useful if you want others to know where to contact you (also know as well know addresses). Dynamic IP addresses are allocated by a service running on the local network from a pool, they are efficient in that they maximize the use of available addresses but are not useful as well known addresses.

When data is sent from A to B with protocols like UDP and TCP/IP there is a return address passed, that allows receivers of messages to send back replies without having prior knowledge of the sender address. This is how holders of dynamic IP addresses can communicate with others.

A typical exchange is a client with a dynamic IP address sends a request to a server with a well known static IP address (actually another layer called DNS is also used which allows (domain) names to be mapped to IP addresses), the server fulfills the request by replying to the client's return address. E.g. fetching a web page from a web server somewhere on the Internet.

The net mask determines the scope of the local network, network local scopes can be of various sizes depending on the needs of the owner. To send data outside of the scope defined by the net mask requires a router to act as a gateway. You don't care about net masks, they are implicit to the class of network you are on.

You don't ask about service port numbers, they are another fundamental part of Internet protocols. Service port numbers differentiate services offered on a particular host machine (actually on each network interface of a host machine). They are part of the addressing mechanism, e.g. port 443 is the well know port where secure webpages are served from (old insecure web servers respond on port 80). Like IP addresses there are well known port numbers and there are ephemeral ports granted to clients on a need to use basis, also like IP addresses a return port number is included with protocols like UDP and TCP/IP which complete the return address. Service port numbers allow multiple servers and clients to reside on a single host, for example you can browse the web, send and receive emails without messages from these different services getting tangled up with each other. You might think of service port numbers as different stores on a high street, although the analogy is pushing it a bit.



--
73

Bill

G4WJS.

Join DXLab@groups.io to automatically receive all group messages.