Thanks for the answers, I'm begining to throw together a small netlib
dissector (since it is such a simple protocol). I'm running into one
problem that I can't seem to find the answer to. Many of the protocols
depend on there being a well known port to which data is being sent.
While MS SQL server has a default port of 1433, but it's not fixed.
Looking though the code it seems I have to add some heuristics to detect
the protocol and point it at my dissector. Is there any documentation on
this? If not, I'd be happy to write it as I figure out exactly what I'm
doing here. Unfortunately, not much of it is making sense at the moment.
Cheers,
Brian
On Tue, 20 Nov 2001, Guy Harris wrote:
> > My question is, does it make sense to make netlib its own dissector?
>
> It sounds as if netlib and TDS bear somewhat the same relationship to
> one another as does the NetBIOS Session Service protocol and SMB; as
> such, I might be inclined to make netlib a separate protocol from TDS.
>
> > And, how do other protocols deal with a trace that may start somewhere in the
> > middle, missing important data from the start of the stream?
>
> By giving up if they can't handle it, or by offering, say, a protocol
> preference item to allow a user to supply information that can't be
> inferred from the capture. (We may want to have some way of specifying
> per-conversation information, separate from global preferences, for
> this. Such a mechanism doesn't currently exist.)
>
> _______________________________________________
> Ethereal-dev mailing list
> Ethereal-dev@xxxxxxxxxxxx
> http://www.ethereal.com/mailman/listinfo/ethereal-dev
>