Hi,
I guess this is right way to go.
FTAM/CMIP/MSS dissectors could register itself in the both PRES and ACSE
dissectors.
Anders Broman (AL/EAB) wrote:
Hi,
Looking at the trace I have http://www.ethereal.com/lists/ethereal-users/200506/msg00263.html
it seems like the context defenition is set up with two items one to ACSE
and the second one pointing out the CMIP protocol. Further looking into the code of ACSE
it does keep track of the Presentation context and uses that for certain SES PDU types(Data).
It would be no problem to register the CMIP dissector with two OID's if thet's the right thing to do and
let the PRES dissector keep track of the presentation context's and call the apropriate OID dissector.
I'm not that familiar with the ITU-T X/ISO protocols so I don't know if ther'd be any pitfalls with this method.
Best regards
Anders
-----Original Message-----
From: ethereal-dev-bounces@xxxxxxxxxxxx
[mailto:ethereal-dev-bounces@xxxxxxxxxxxx]On Behalf Of Graeme Lunt
Sent: den 30 augusti 2005 11:32
To: Ethereal development
Subject: Re: [Ethereal-dev] ISO 8823 OSI Presentation Protocol
Hi,
Nice timing.
I have checked in an asn2eth generated PRES dissector. Please test it.
Points to check:
I'm unsure about the selecton of which PDU type to dissect - works with my
limited captures.
Probably ASCE should not be called directly but rather a table of context's
be built and the next dissector be called based on the information received,
but this was how it was made in the previous dissector.
I shall try this out as I have found problems with the existing PRES dissector.
It wasn't maintaining the presentation context list at all - which
made it impossible to dispatch on the indirect references. My local
patched version dissected based on the presentation context (abstract
syntax) rather than the association context. This bypassed the ACSE
dissector altogether (when not necessary of course).
Note that my change means you need to register the dissector on the
presentation context, not the association context. I'm not sure, but I
guess that the existing protocol dissectors only use a single
presentation context so there is a one to one mapping with the
association context. X.400 generally uses multiple PCs.
I assume this asn2eth PRES dissector is the preferred way to go now?
I have asn2eth-based RTSE, ROS, X.411 (P1), X.420 (P22) dissectors
which I will submit when I have a verified they work with this new
PRES dissector and the ROS has been verified to work independently of
RTSE (I'm implementing an asn2eth-based X.511 (DAP) for this).
Graeme
_______________________________________________
Ethereal-dev mailing list
Ethereal-dev@xxxxxxxxxxxx
http://www.ethereal.com/mailman/listinfo/ethereal-dev
_______________________________________________
Ethereal-dev mailing list
Ethereal-dev@xxxxxxxxxxxx
http://www.ethereal.com/mailman/listinfo/ethereal-dev
.............................................................
Программа создания веб-сайтов за два дня: http://web.sema.ru/
