Ethereal-dev: Re: [Ethereal-dev] Problems with H.323 dissector(s)

Note: This archive is from the project's previous web site, ethereal.com. This list is no longer active.

From: "Martin Regner" <martin.regner@xxxxxxxxx>
Date: Sat, 30 Aug 2003 07:10:07 +0200
Andreas Sikkema:
>I just checked some H.323 captures I have in my archive and I found a
>couple of problems.
>
>1. InformationRequestResponses are not decoded properly (I have not
>   yet seen one decoded properly) (H.225.0 RAS)
>2. There is a problem with H.225.0 CS messages (particularly with
>   SETUP messages for some reason.
>3. I have seen problems with H.245
>
>The last two are probably related to TPKT handling. The first two
>problems are evident in the attached capture.


Hi,

I have also noticed problems with IRR messages. I haven't looked so much closer on that issue, but I sent a sample capture to
Ronnie yesterday.

I checked the Setup messages in your capture and I don't see any problem. Can you send a printout of the Setup message
and also indicate more in detail what is faulty. Are you using TCP desegmentation?
Are you using the latest version of the H.225/H.245 dissectors?

I am going to test the latest version of the H.245/H.225 dissectors together with a lot of captures during the weekend.
The tests I did Friday went very well then (but then I only tested a subset of captures). The only problems I found then was:

  1) IRR messages:
================

                perCallInfo
                    Sequence-Of Length: 6
                    Item 0
                        perCallInfo_item
                            CallReferenceValue: 256
[Malformed Packet: H225]

Same problem as appeared in your capture.

  2) productID/versionID without null-termination resulting in rubish last in the productID/versionID:
==========================================================================

e.g.
                        h221NonStandard
                            t35CountryCode: 11
                            t35Extension: 11
                            manufacturerCode: 11
                        Octet String Length: 28
                        productID: InnoMedia InfoView Broadband\00101\001@
                        Octet String Length: 2
                        versionID: 01\001@
instead of:
                        h221NonStandard
                            t35CountryCode: 11
                            t35Extension: 11
                            manufacturerCode: 11
                        Octet String Length: 28
                        productID: InnoMedia InfoView Broadband
                        Octet String Length: 2
                        versionID: 01