[Pascal Quantin <pascal.quantin@xxxxxxxxx>]

2015-11-03 16:13 GMT+01:00 POZUELO Gloria (BCS/PSD) <gloria.pozuelo@xxxxxxxx>:

Hi all,

 

I’m developing an extension for the GTPv1 dissector and while I was debugging I’ve encountered something that it seems to me a bit confusing. In the dissect_gtp_common function from the packet-gtp.c source, there is a section that makes the matching between requests and responses. The thing is that I have a pcap in which I have a create PDP context request with sequence number equal to 0, then when in the code asks if the message has sequence number, it never enters in that part because, I guess, that 0 is equal than NULL.

 

If (seq_no) {

                /* matching is done */

}

 

I don’t know If this is a little bug or maybe the sequence number can’t be equal to 0. Can someone help me?

 

Thanks!

 

Regards.

Hi Gloria,

at first glance this seems to be a bug. Per 3GPP 29.060 chapter 9.3.1.1:

9.3.1.1            Usage of Sequence Number

The sending GGSN and SRNC shall use 0 for the value of the Sequence Number of the first G-PDU in a tunnel, only during the PDP context activation, and shall increment the Sequence Number for each following G-PDU. The value shall wrap to zero after 65535.

The receiving GGSN and SRNC shall set the content of a counter to zero, only during the PDP context activation. When the receiving GGSN and SRNC receives a valid G-PDU, it shall increment this counter by one. This counter shall wrap to zero after 65535. It defines the "Expected Sequence Number".

Could you please fill a bug on https://bugs.wireshark.org and even better, upload a fix on Gerrit (see https://www.wireshark.org/docs/wsdg_html_chunked/ChSrcContribute.html for details) ? Presumably we should have a booleab telling whether a sequence number was retrieved or not and test it (instead of seq_no being different from 0).

Regards,

Pascal.