I tried the config that they showed on the page for kiss tcp connections, with both computers setup on the same wlan. I used 127.0.0.1 as the ip address in computer 1's config and computer 1's ip address on the network as the ip address in computer 2's config. Both computers were set to port 8100 (which was not previously occupied.) It didn't work, but I was able to get it to work VERY well in AGW mode with Paracon as the client. I was able to install PE1RRR's web browser and load 70+ line web pages with it, so it's working great now.

?

Did it sound like I set it up right for the kiss tcp connection? I have attached the client's config file. That said, I think I recently read on here somewhere that Linbpq nodes don't work very well as clients.

?

Does upnp do anything major in bpq? Is it safe to exclude from builds in the future or will that impact functionality at all?

Starting with Ubuntu 24.10 this will impact anyone trying to build, and Fedora's lagging behind a little bit but they'll catch up soon I'm sure!

Cheers,

--?

? Hibby

? Debian Developer

? Packet Radioist

? MM0RFN

On Sun, 18 Aug 2024, at 2:06 AM, Dave Hibberd wrote:

Hi there,

?

libminiupnpc has been updated and now UPNP_GetValidIGD() requires another argument - you can see the change and new definition on Github here:

?

?

This is what's being shipped in debian unstable/testing, so it's going to filter down to other systems soon and is preventing me from building packages for the repo.

?

This has impact on upnp.c only (so far).

?

It builds for bookworm fine still, as that version won't change for the life of the distro!

Hi there,

?

libminiupnpc has been updated and now UPNP_GetValidIGD() requires another argument - you can see the change and new definition on Github here:

?

?

This is what's being shipped in debian unstable/testing, so it's going to filter down to other systems soon and is preventing me from building packages for the repo.

?

This has impact on upnp.c only (so far).

?

It builds for bookworm fine still, as that version won't change for the life of the distro!

On 8/24/24 05:28, Jerry Kutche wrote:

Do you have individual MAP statements to each desired link and do those desired stations have your MAP info ?

AUTOADDMAP Can be dangerous.. or at least yield undesirable links. 

Hi, Jerry:

AFAIK: No, 1 is an individual MAP line. Others are via a network broadcast.

AFAIK:
Each member of the AXUDP network has a credential and a key.
The network is managed by Matt KB8UVN.
I trust Matt.
Ergo, I trust AUTOADDMAP on this network.

Perhaps I should have an additional PORT for the 1 AXUDP neighbor that is not 'in-network'.

:-|

73, Chuck

That being the case, we have a serious security problem. With AUTOADDMAP active, unknown non ham entities can? gain access to what's supposed to be a closed network. It's like opening the border. Anyone and everyone will be able to waltz in unhindered. That was recently experienced here when I made the mistake of activating AUTOADDMAP. What was reportedly some sort of European CB type node gained direct access to my system, and I to theirs. That is a non starter and will not end well if left uncontrolled.? It's advisable to NOT enable AUTOADDMAP.
My system seems to (mostly)? work just fine without it.
However, in consideration of the above, I wonder if not running AUTOADDMAP could be the reason for some AXIP/BBS partners using DDNS.NET not being accessible? I've lost at least two AXIP/BBS partners who utilize DDNS.NET in that they are simply not reachable. They may resolve to an IP address but just don't respond. There's no route.? It's been reported that they see me on their end, but there is no connectivity. A route to them never or rarely appears on my end. If a route appears, it is dead/unresponsive. They have my correct MAP information entered on their end, and I theirs. My solution was to dump DDNS.NET and replace it with DYNU/FREEDDNS.ORG. That works well for me. DDNS.NET must be eliminated on both ends just to be sure.
Best regards,
Gary, K7EK
Get BlueMail for Android <>
On Aug 24, 2024, at 05:22, Chuck Gelm <nc8q-aredn@... <mailto:nc8q-aredn@...>> wrote:
On 8/23/24 16:41, John G8BPQ wrote:
If you are talking about AUTOADDMAP that is not needed.
73, John
On 23/08/2024 20:52, Chuck Gelm wrote:
If I understand, this is necessary for AXUDP links to
function as
intended.
73, Chuck
Hi, John:
Thanks.
I want to get you and Matt KB8UVN chatting about this.
When I commented my AUTOADDMAP, I lost all links.
When I re-enabled AUTOADDMAP I got my links back.
I will send you a copy of Matt's recent email to me
containing an explanation of the protocol in use.
vy73, Chuck

On 8/23/24 16:41, John G8BPQ wrote:
 If you are talking about AUTOADDMAP that is not needed.

 73, John

 On 23/08/2024 20:52, Chuck Gelm wrote:
 If I understand, this is necessary for AXUDP links to function as 
 intended.
 73, Chuck

Hi, John:

Thanks.
I want to get you and Matt KB8UVN chatting about this.
When I commented my AUTOADDMAP, I lost all links.
When I re-enabled AUTOADDMAP I got my links back.
I will send you a copy of Matt's recent email to me
containing an explanation of the protocol in use.

vy73, Chuck

If you are talking about AUTOADDMAP that is not needed.

73, John

On 23/08/2024 20:52, Chuck Gelm wrote:

If I understand, this is necessary for AXUDP links to function as
intended.
73, Chuck

If I understand, this is necessary for AXUDP links to function as intended.

?

73, Chuck

?

That is what I thought. I had checked IDLETIME setting before I took the system to the tower and thought it was working as you described. I'll double check on what I think happened.

?

I may be confussed. Is the timer reset on any activity from any user?

?

Dave Menges

I have a LinBPQ system running with just Direwolf and a KISS connection on the same Pi. I thought that by using SIMPLE in the config file I would invoke a 15 minute session timer with ?IDLETIME. But, I see that IDLETIME is just for ax.25 Level 2 Parameters. Is there a way to set a session time for a packet port?

?

Dave Menges

WB9TEN

?