On Mon, Sep 30, 2019 at 7:57 PM Pascal Quantin <pas...@wireshark.org> wrote:

> Hi Achuthan,
>
> Le lun. 30 sept. 2019 à 19:53, Achuthan Paramanathan <a...@kamstrup.com> a
> écrit :
>
>> Dear wireshark,
>>
>>
>>
>> I recently noticed that decryption of downlink traffic in wireshark  -
>> Version 3.0.3 is not done properly.  Even though it is marked as correct
>> decoded in wireshark view. The uplink traffic is decrypted correctly.
>>
>>
>>
>> I then had a look in the code (*master branch*)
>> “../dissectors/packet-lorawan.c” line 795
>>
>> Here, the function decrypt_lorawan_frame_payload() which takes in an
>> argument dir, in this case it is statically set to 0 indicating uplink
>> traffic. This will work as long as the traffic is uplink, however in
>> downlink this will result in an wrong decryption.
>>
>>
>>
>> My suggestion is to use the flag “uplink” which is set to FALSE in line
>> 753 when mac_mtype is of type downlink, as dir argument in function call
>> decrypt_lorawan_frame_payload()
>>
>
> Could you please fill a bug on https://bugs.wireshark.org with a pcap
> containing a bit of UL and DL traffic? It would allow to verify your fix
> proposal.
>
> Thanks for the help,
> Pascal.
>
>>
>>
> ___________________________________________________________________________
> Sent via:    Wireshark-dev mailing list <wireshark-dev@wireshark.org>
> Archives:    https://www.wireshark.org/lists/wireshark-dev
> Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev
>              mailto:wireshark-dev-requ...@wireshark.org
> ?subject=unsubscribe


It would be great to have a pcap, please don't forget to include the keys
to decrypt it.
I will be happy to update the dissector.

Regards,
Erik
___________________________________________________________________________
Sent via:    Wireshark-dev mailing list <wireshark-dev@wireshark.org>
Archives:    https://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev
             mailto:wireshark-dev-requ...@wireshark.org?subject=unsubscribe

Reply via email to