> On 6 Jan 2019, at 19:54, Guy Harris <guy@xxxxxxxxxxxx> wrote:
>
> On Jan 6, 2019, at 10:30 AM, Jaap Keuter <jaap.keuter@xxxxxxxxx> wrote:
>
>> Rather than simplistic endpoint ID’s I think we need an ID tuple per endpoint,
>
> How is a tuple not itself an ID?
It is, just not limited to the address/port combinations most of them are right now. It incorporates the whole list of IDs that uniquely identify an endpoint in a conversation among all other endpoints in other conversations in the same capture.
>
> And not all conversations necessarily have specific endpoints.
You mean, where the wildcards come into play?
>
>> which may be combined with one (or more) other tuples representing single (and multipoint) connections.
>> Examples are an aggregating tap/monitor port which monitors various VLANs, or an MPLS link. Or even closer to home, a multi port capture in a pcapng file, lets say of two ports of a switch or router. The conversations therein would need to be identified from the capture interface on up.
>
> The intent here is to have a general concept of a "conversation", with no specification, at that layer, as to how a "conversation" is identified - think of it as an abstract base class - with subclasses that use different ways of identifying whether a packet belongs to a given conversation or not. Multiple subclasses can share code for identifying that; TCP and UDP might share the "IP address and port" identification code.
>
Sure, so the tuple I’ve talked about would be a derived class consisting of all elements within the tuple. That makes sense.
> (I"m not sure I like the name "conversation", but I'm not sure I like "flow" as that strikes me as half of a conversation going in only one direction, and I'm not sure what other name would be good for that broad a concept.)
Yes, well, “conversation” itself is not loaded in some technical context with a specific meaning (AFAIK), so allows us to make the definition for ourselves, with the flexibility it needs.
Thanks,
Jaap