[Xastir] Position data from Xastir not accepted by Yaesu FTM-400 or Direwolf

[Tom Henderson]

Thanks Tom, I have updated to the new version. Glad to know there was a 
reason for that strange behavior and that it has been fixed.

Tom Henderson

On 11/13/18 12:17 AM, Tom Russo wrote:
> I just looked at my station's raw packets at aprs.fi, and every single
> packet my home station has transmitted ever since I did the git pull of
> the problematic commit has been getting rejected as an unsupported packet.
> I had never looked.
>
> If you are running a version of Xastir from a git checkout after August 1st,
> you definitely want to do another pull tonight.
>
> If you were running more than one interface, including both NET and TNC,
> the behavior of your station has been exactly what Tom Henderson describes:
> if your NET interface appears in the list of interfaces prior to your TNC,
> your packets going out over the air have been broken.
>
> On Mon, Nov 12, 2018 at 06:40:16PM -0700, we recorded a bogon-computron collision of the <russo at bogodyn.org> flavor, containing:
>> I just encountered this very issue.  I have not been using APRS mobile for
>> quite some time, but today we had a snow day and I got into my truck with
>> my D700 for the first time in ages.  I noticed that all of my home Xastir
>> packets were showing up as invalid, with exactly the issue of these extra
>> TCPIP* headers being inserted.  I have had exactly the same set of interfaces
>> defined for *years* in exactly the same order, and never noticed this before.
>>
>> I backed up my git checkout to a state only a few commits back, and the
>> behavior went away.
>>
>> This bug was introduced in commit  3abfd059, titled "Bugfixes" as part of pull
>> request #41 by Mike Nix, which was intended to correct a possible
>> uninitialized variable issue, and which Curt merged on August 1st.
>>
>> As part of that set of "bug fixes" an "output_net" string array was set to a
>> null string (by writing '\0' into its first character) as an initialization
>> step, and several places where it was explictly done when needed were deleted.
>>
>> Removing that "fix" restores my Xastir to correct function, where my
>> transmitted data (which is done via an AGWPE link) no longer has the
>> inappropriate second header.
>>
>> The problem here is that the initialization of output_net was placed
>> *OUTSIDE* the loop over ports, and all the specific initializations of that
>> array are inside case statements inside the loop.  All of the interface-
>> specific clearing of that array were removed from inside the loop, and
>> so any interface that put something *OTHER* than a blank string into
>> output_net caused all subsequent interfaces (which used to explicitly
>> clear the array before the "fix") reused the non-blank version.
>>
>> This explains why it suddenly seemed to matter what order interfaces were
>> defined in, because information from low-numbered net interfaces was not
>> being properly cleared by higher-numbered TNC interfaces.
>>
>> I have undone this mistake in commit 8d403fe61.
>>
>> It is once again irrelevant what order you define your net and tnc interfaces.
>>
>> The other bug fixes in commit 3abfd059 don't look wrong, so I left them in
>> place.
>>
>> On Wed, Oct 24, 2018 at 07:14:39AM -0500, we recorded a bogon-computron collision of the <Tom at Henderson4.us> flavor, containing:
>>> Aha! I fixed it! Since the 2nd header was showing TCPIP*, I started
>>> wondering if it was an artifact of the prior interface 0, which was my
>>> APRS-IS interface. So I shut down Xastir and re-arranged the interfaces
>>> so that the serial KISS TNC was the first interface.
>>>
>>> Viola! now the transmitted packet doesn't have the second TCPIP header
>>> in it. And my FT1XD reads the packet as valid and shows distance and PHG
>>> info.
>>>
>>> Tom Henderson
>>>
>>> On 10/23/18 7:38 PM, Tom Henderson wrote:
>>>> Radio 1 is KW4TOM
>>>>
>>>> Radio 2 has no call sign. it is only receiving APRS packets that
>>>> direwolf is decoding.
>>>>
>>>> Radio 3 is KW4TOM-7, but that has nothing to do witht he malformed
>>>> packet it receives ffrom xastir as far as i can tell.
>>>>
>>>> Tom Henderson
>>>>
>>>> On 10/23/18 7:19 PM, Bill Vodall wrote:
>>>>> What are the callsigns for all 3 stations?
>>>>>
>>>>>> Xastir is connected to one radio via a Mobilnkd TNC2 and receives/xmits
>>>>>> through radio 1, and Icom IC-V8000
>>>>>>
>>>>>> Radio 2 is a Yaesu FT-991 with a built-in USB sound card. I'm using
>>>>>> direwolf to ready the APRS packets from this radio
>>>>>>
>>>>>> Radio 3 is a Yaesu FT1XD with built in APRS.
>>>>> _______________________________________________
>>>>> Xastir mailing list
>>>>> Xastir at lists.xastir.org
>>>>> http://xastir.org/mailman/listinfo/xastir
>>>>>
>>>> _______________________________________________
>>>> Xastir mailing list
>>>> Xastir at lists.xastir.org
>>>> http://xastir.org/mailman/listinfo/xastir
>>>>
>>> _______________________________________________
>>> Xastir mailing list
>>> Xastir at lists.xastir.org
>>> http://xastir.org/mailman/listinfo/xastir
>> -- 
>> Tom Russo    KM5VY
>> Tijeras, NM
>>
>>   echo "prpv_a'rfg_cnf_har_cvcr" | sed -e 's/_/ /g' | tr [a-m][n-z] [n-z][a-m]
>>
>> _______________________________________________
>> Xastir mailing list
>> Xastir at lists.xastir.org
>> http://xastir.org/mailman/listinfo/xastir