[Xastir] Bug#876626: Xastir loose TCP/IP data afer 12 hours of use
MLHPUB
mlhpub at free.fr
Sun Oct 29 11:47:05 PDT 2017
Test is done at 1727z.
We can see on joined screenshots :
- first message and immediate ACK on telnet session but not seen by Xastir
- if a second message is sent, ACK is also immediate
- sent message counter growing in message window without any effect on
APRS-IS telnet session (only first message visible). A second message is
really sent when counter is #4.
- offset about 2 minutes between an APRS-IS message not taken into
account by Xastir and its display in messages window (example with group
removal)
I have tried to stop and restart APRS-IS interface without any
improvement. The only way to retrieve a normal result is to shutdown and
restart Xastir.
Config here :
- HP4310s laptop
- RAM 3 Gb
- CPU Intel(R) Core(TM)2 Duo CPU T6670 @ 2.20GHz
- Debian 9 amd64
- KDE
Any ideas ?
73 de F4ACU
Matthieu
Le 28/10/2017 à 18:11, Lynn W. Deffenbaugh (Mr) a écrit :
> telnet rotate.aprs2.net 14580 (or whatever server you have configured
> in xastir to keep things equal)
>
> After the greeting, type:
>
> user <YOURCALL> pass -1 (And you must do this without errors or
> backspaces)
>
> This will get you periodic heartbeats from the APRS-IS server. If you
> want to see packets, you cannot use the m/ filter because you'll never
> provide a posit packet. I usually do a buddy filter for some callsign I
> know to be active such as my own to pick up my other stations' packets.
>
> #filter b/KJ4ERJ* (this will get you some traffic, possibly more than
> you're wanting)
>
> Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
>
> PS. This assumes that you're not using your straight call in xastir.
> Otherwise, modify <YOURCALL> to include a unique (doesn't have to be
> numeric) -SSID like YOURCALL-TS or somesuch. APRS-IS servers will
> routinely disconnect multiple connections from a single callsign-SSID.
>
>
>
>
> On 10/28/2017 11:03 AM, Jason KG4WSV wrote:
>> You can use another tool like telnet to establish an APRS-is
>> connection at the same time / on the same system as xastir.
>>
>> Open a terminal window and type
>>
>> telnet APRS.server port
>>
>> Where APRS.server is the name of the server you connect to and port is
>> the port number.
>>
>> Once connected you may have to send an authentication string. I can’t
>> recall the details and can’t look them up at the moment, but maybe
>> someone else can chime in.
>>
>> You should see a stream on APRS packets in that window.
>>
>> Then start xastir in another window. When xastir stops, see if the
>> data is still flowing in the telnet session.
>>
>> If telnet stops when xastir does, you know it is not an xastir problem
>> and likely is a network problem.
>>
>> -Jason
>>
>>
>>> On Oct 28, 2017, at 2:43 AM, MLHPUB <mlhpub at free.fr> wrote:
>>>
>>> Hello all of you,
>>>
>>> Sorry, no more news from me (a bit busy now).
>>>
>>> To check any network issue, can you tell me what kind of test I
>>> should do ?
>>>
>>> 73 de F4ACU
>>>
>>> Matthieu
>>>
>>> F4ACU TCP/IP
>>> F4ACU-1 Robust-Packet
>>> F4ACU-5 Android
>>> F4ACU-7 Yaesu VX8-GE
>>>
>>>
>>>> Le 27/10/2017 à 03:10, David A Aitcheson a écrit :
>>>> Gerry/Jason,
>>>> I confirmed with Mattthieu F4ACU that the problem starts about the
>>>> 12 _HOUR_ mark so it very well could be a home network or a home
>>>> router issue.
>>>> Mattthieu,
>>>> Any updates for us?
>>>> Dave
>>>> KB3EFS
>>>>> On 10/26/2017 09:00 PM, Jason KG4WSV wrote:
>>>>>> On Thu, Oct 26, 2017 at 2:14 PM, Gerry Creager - NOAA Affiliate
>>>>>> <gerry.creager at noaa.gov> wrote:
>>>>>> A tcpdump would be helpful here, to assure it's not an
>>>>>> ADSL/provider issue.
>>>>> Now that Gerry makes me think about the network, it could be a home
>>>>> network issue - e.g. some sort of PAT forwarding table timeout,
>>>>> error, etc.. There are probably few days/weeks long TCP connections
>>>>> to compare it with, though. I guess you could use telnet to connect
>>>>> to your favorite APRS-IS server and see if it dies at the same time
>>>>> xastir does. If so, that points at the network and not xastir.
>>>>> -Jason kg4wsv _______________________________________________
>>>>> 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
More information about the Xastir
mailing list