Re: [Ntp] Symmetric mode

[Danny Mayer <[e-mail]>]

On 10/1/22 2:28 AM, Hal Murray wrote:
>> Peer modes (1/active and 2/passive) are for exchanging time between peers.
> I can get the same results with a pair of client/server connections.  Right?
Yes but you are only talking about the on-wire protocol. The 1/2 modes 
indicates that they are peers and the system can take different actions 
based on that knowledge. It's not just about whether you are the client 
or the server.
>
>
>> Local policy choices include "do we want to use peer mode on any machines?"
> Why would anybody choose peer mode over a pair of back-to-back client/server
> connections?
>
Harlan has provided some additional information about that in a 
different reply.
> Is there any good reason that we should:
>    include symmetric modes in NTPv5
>    work on NTS for symmetric modes
Yes.
> The main reason not to is that it adds unnecessary complexity.
Not at all. The complexity is not on-wire, but how the system deal with it.
> Symmetric mode uses half the bandwidth.  That doesn't seem important with
> modern network technology.  It might be important in some cases.  I wonder how
> many of those cases don't have a better solution.
>
That is an untrue statement. Symmetric mode uses EXACTLY the same bandwidth.

Danny