Re: user-creation with axspawn [again]

[Bob Meyer]

Robert Schelander wrote:

> Hi to all!
>
> I've configured axspawn to create users automatically,
> but creating the password doesn't seem to work.
> It inserts a + into the passwd file, but if this is supposed
> to be the password it does not work.
> Anyway why doesn't write axspawn anything into the
> shadow file where the passwords normally are?
>
> oe8rsq:+:400:200:oe8rsq:/home/ax25/oe8rsq:/bin/bash
>
> Do I have to modify axspawn.c to do this correctly?

yes

>
> If you don't know how to do this in C even some hints
> how you would do that on the cmdline would help.
> Maybe it's possible to use passwd / crypt or something
> else out of a script to realize correct standard-password
> creation.
>

You can create an entry in /etc/shadow without a password.

You could run passwd and use the callsign for a password

You could do a callsign lookup and use the name or some other part of
the reply as a password.

Bob

Re: Soundmodem questions

[Hamish Moffatt]

On Sun, Oct 17, 1999 at 02:40:26PM -0500, Joe Martine, N5USR wrote:
> I was able to get the audio portion working immediately, and proceeded to
> try to get the PTT going.  I first tried to use a serial port, but have so
> far been totally unable to get the RTS or TxD to toggle.  I reconfigured to 
> use the parallel port, and it immediately worked.  Is there something
> else I need to do to get the serial port working?  The ports are properly
> configured for ttyS0 and ttyS1, standard I/O and IRQs.  Nothing else is
> using them, no gettys or anything.  If I fire up minicom, using hardware
> flow control RTS toggles nicely so I know the hardware works.  (These

Perhaps that's the problem -- the soundmodem driver has its own driver
for the serial port. You must tell the serial driver not to use that port,
so that soundmodem can use it. Either remove the serial module (if you don't
need it for other ports), or use "setserial /dev/ttySn uart none".



cheers

Hamish VK3SB

Re: Routing problem [solved]

[Robert Schelander]

Thanks Tomi! 
Creating the directory solved the problem.

Automatic creation should be added, I think.

>There was a posting about this on this list some time ago I think...
>Yes, see .
>You need to create /var/ax25/ax25rtd/.

Thanks again
Robert

Re: ham web

[Steve Fraser]

Harold Hartley wrote:
> > It is a configurable server designed for limited web pages from the
> > internet to include callbook servers, dxcluster, or any other service
you
> > wish to provide and will interface to http on the 'net.
> >
> > According to the QST article (written by John Hansen, W2FS) the server
> > uses UI frames exclusively, so essentially it runs in broadcast mode all
> > the time. So what I gather is that depending on the request and baud
rate
> > of the requesting station, different information replies to the
requester
> > are sent (it is up to the requesting stationor server to determine
> > efficiency of his/her connection - baud rate, etc.).
> >
then Bob Meyer wrote:

> Okay it's a file server using ax25 in disconnected mode.  Kind of an aprs
> style file server.
>
> Wouldn't this work a lot better?
> http://gw.ko6ri.ampr.org/bbs/cgi-bin/callbook.cgi?kb6fst


I think both these miss the REAL point of ham web - that it's a broadcast
protocol. It's conceptually similar to that used by the pacsats. It's real
benefit, and one that makes it very suitable to web serving over packet
radio, is that ANY file can be transferred  to MANY stations simultaneously.
IIRC it doesnt have the "hole request" capability of the pacsat protocol,
but sends the file(s) repeatedly. Hence bulletins, web pages, mail, or
anything else that can be bundled up as a file, transferred, then unbundled
and processed can make use of it. The existing version already has some
hooks into existing software packages. Web is just one application for it.

It's also worth pointing out that the specifications are documented, so a
Linux version could be implemented.

73

Steve

user-creation with axspawn [again]

[Robert Schelander]

Hi to all!

I've configured axspawn to create users automatically,
but creating the password doesn't seem to work.
It inserts a + into the passwd file, but if this is supposed
to be the password it does not work. 
Anyway why doesn't write axspawn anything into the 
shadow file where the passwords normally are?

oe8rsq:+:400:200:oe8rsq:/home/ax25/oe8rsq:/bin/bash

Do I have to modify axspawn.c to do this correctly?
If you don't know how to do this in C even some hints 
how you would do that on the cmdline would help. 
Maybe it's possible to use passwd / crypt or something
else out of a script to realize correct standard-password 
creation.

PS: I do not want to use axspawn for telnet access, but
only for user creation and then use /bin/ftponly as standard
shell.

Thanks
Robert

Re: Routing problem [was: ARP problem]

[James S. Kaplan KG7FU]

I'm also having trouble with ax25rtd. The maintainer has a patch for
ax25utils, but I'm using rpm's from
SuSE 6.2Anyone have binaries or a tarball they would share for 2.2.9
libc6?

Similarly, rspfd-0.08 doesn't compile with libc6, depends on
/usr/include/sys/socketio.h from an old libc.
Anyone have solutions for this one?

nrparms seems to be broken in the SuSE distribution also. nrparms -routes
works as advertised. However, nrparms -nodes will not take input in any
form, even from files saved with nodesave.

-
James S. Kaplan KG7FU
Eugene Oregon USA
[EMAIL PROTECTED]
http://www.rio.com/~kg7fu
ICQ # 1227639
Have YOU tried Linux today?
-


- Original Message -
From: Robert Schelander <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Sunday, October 17, 1999 11:43 AM
Subject: Re: Routing problem [was: ARP problem]


> Very good idea. Thank you.
> After looking at the docs of ax25rtd, I knew that this is exactly what I
> need.
>
> >Here I think ax25rtd might help. I understand it is supposed to do just
> >that hack automatically. I'm not sure however as I haven't used it.
>
> My "only" problem now is, that ax25rtd doesn't run.  :(
>
> oe8rsq:/etc/ax25 # startproc /usr/sbin/ax25rtd
> bind Control socket: No such file or directory
>
> By the way, I've found in the manpage of axparms, that
> axparms -route add   [] [-ipmode V|D]
> is able to set VC/Datagram from each route individually.
> ax25rtd goes a step further and is able to set it automatically to
> the mode the calling station uses. (unfortunately after the kernel
> sends the answer - not good if the other station uses this autoset
> too)
>
> Thanks
> Robert

Re: Mobile Packet Radio

[Jeff Grammer]

Thanks Paul!  So your suggestion is to use a laptop with a Baycom modem and
software; rather than a TNC in KISS mode right?  My main use would be
probably for message transmission.

Thanks again;

Jeff KF4WYM

(BTW, does anyone know if there are any know packet networks in the
Kentucky area?  2m/70cm what frequency?)

At 08:08 AM 10/17/99 +1000, you wrote:
>On Sat, 16 Oct 1999, Jeff Grammer wrote:
>
>> 
>> I know this may sound a bit odd, but here's what I want to do.  I have an
>> 
>> Connect up a Laptop computer to it for use as a mobile packet radio system.
>> 
>
>Hi Jeff,
>
>Depends upon what you exactly want to do while packet mobile. I have run
>packet radio mobile for a while now and found the best combination to be a
>small laptop with a Baycom modem and Baycom software. Of course this might
>not be what you want.
>A good antenna and some power is a must as there is a great deal of
>flutter and multi-pathing while moving. I found 2M to give better results
>than 70cm but that could be due to the location of the digipeaters and
>surrounding terrain.
>Experiment and see what suits you best.
>
>regards,
>
>Paul VK3ALE
>[EMAIL PROTECTED]
>
>
>


... The free UNIX operating system
  :::'  ...  ..
  :::  *  ::.::'
  ::: .::  .:.::.  .:: .::  ::. :'
  :::  ::   ::  ::  ::  ::   :::.
  ::: .::. .::  ::.  . .:'  ::.
..:::.::'   ..

Re: Any hams in the Netherlands?

[Regnerus Dantuma]

Hi Fotis

For sure there are people interrested in doing that - and also
in Amsterdam. I may be spending December and January in that area.

vy 73 

Regnerus Dantuma wp2b <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>

On Sat, 16 Oct 1999, Fotis Georgatos wrote:

> 
> Hi,
> 
> is somebody from the list in NL, especially Amsterdam,
> interested for packet radio experiments?
> 
> I've relocated and I'm settling electronically now...
> 
> Please, reply to this email address.
> 
> cheers
> 
> --
> "Optimists think this is the best world we could live in,
> pessimists are afraid that optimists are right"
> 
> fotis
> 

Soundmodem questions

[Joe Martine, N5USR]

I finally decided I was sick of 1200 baud, and wanted to use the
soundmodem driver to speed things up a bit.  So far, so good, I spent the
morning getting the afsk1200 talking to my existing station, but do have a
couple of problems that arose.  I keep thinking I've seen these come up
before, but I can't find them in the archives anywhere, so here goes...

The system is a P-166 w/ Red Hat 6.1 and the latest three ax25 packages
(lib, tools, apps).  Everything compiled perfectly, and I configured the
files in /etc/ax25.  (I compiled the three packages to use the old
locations.)

I was able to get the audio portion working immediately, and proceeded to
try to get the PTT going.  I first tried to use a serial port, but have so
far been totally unable to get the RTS or TxD to toggle.  I reconfigured to 
use the parallel port, and it immediately worked.  Is there something
else I need to do to get the serial port working?  The ports are properly
configured for ttyS0 and ttyS1, standard I/O and IRQs.  Nothing else is
using them, no gettys or anything.  If I fire up minicom, using hardware
flow control RTS toggles nicely so I know the hardware works.  (These
are on-motherboard serial ports if it matters...)  The parallel port will
work for now, but I would sure like to free it up!

And what happened to smdiag?  After installing the utils I have smmixer,
sethdlc, so forth but no smdiag...  I was able thru trial and error to get
the audio levels right at 1200, but I expect the higher baud rates will
require more accurate level adjustments...

Thanks,
Joe Martine, N5USR
[EMAIL PROTECTED]

Re: Routing problem [was: ARP problem]

[Tomi Manninen]

On Sun, 17 Oct 1999, Robert Schelander wrote:

> My "only" problem now is, that ax25rtd doesn't run.  :(
> 
> oe8rsq:/etc/ax25 # startproc /usr/sbin/ax25rtd
> bind Control socket: No such file or directory

There was a posting about this on this list some time ago I think...
Yes, see .
You need to create /var/ax25/ax25rtd/.

> By the way, I've found in the manpage of axparms, that
> axparms -route add   [] [-ipmode V|D]
> is able to set VC/Datagram from each route individually.
> ax25rtd goes a step further and is able to set it automatically to
> the mode the calling station uses. (unfortunately after the kernel
> sends the answer - not good if the other station uses this autoset 
> too)

Hmm. Didn't know that. I have never needed AX.25 routes for anything...

-- 
Tomi Manninen   Internet:  [EMAIL PROTECTED]
OH2BNS  AX.25: [EMAIL PROTECTED]
KP20ME04Amprnet:   [EMAIL PROTECTED]

Re: Datagram <-> Virtual Circuit - UI-frames <-> I-frames

[Dominik Schaefer]

Hi Robert!

Am Sat, 16 Oct 1999 schrieb Robert Schelander:
> Do you know how I can tell the linux system to use UI's only for
> local stations on the local port and use I's for digipeated stations
> which log in from the port which is connected to a link.
You can tell Linux which mode to use when you define a route with axparms:
e.g.: "axparms -route add yam0 db0bri-10 db0afs -ipmode V" for a route which
uses Virtual Circuit and "-ipmode D" for Datagrams.
The default behavior for a port can be set in
"/proc/sys/net/ax25//ip_default_route" (0==Datagram, 1==VC).

--
Bye...
Dominik
PGP-Fingerprint: DD 96 28 DE E6 30 4E 98  95 45 07 61 D3 6A 32 9E 94 21 EF CC

Re: Routing problem [was: ARP problem]

[Robert Schelander]

Very good idea. Thank you.
After looking at the docs of ax25rtd, I knew that this is exactly what I
need.

>Here I think ax25rtd might help. I understand it is supposed to do just
>that hack automatically. I'm not sure however as I haven't used it.

My "only" problem now is, that ax25rtd doesn't run.  :(

oe8rsq:/etc/ax25 # startproc /usr/sbin/ax25rtd
bind Control socket: No such file or directory

By the way, I've found in the manpage of axparms, that
axparms -route add   [] [-ipmode V|D]
is able to set VC/Datagram from each route individually.
ax25rtd goes a step further and is able to set it automatically to
the mode the calling station uses. (unfortunately after the kernel
sends the answer - not good if the other station uses this autoset 
too)

Thanks
Robert

Re: ham web

[Bob Meyer]

John Cusick wrote:

> On Sat, 16 Oct 1999, Bob Meyer wrote:
> > Harold Hartley wrote:
> >
> > > you can find the web site at http://www.tapr.org/~wa0ptv and it tells
> > > you the concept of it..
> >
> 
> > My question is;  what does it have to do with web pages?
> >
> > Bob
>
> Bob, et. al.
>
> This is sort of a synopsys of the article for those who do not reveive
> QST:
>
> It is a configurable server designed for limited web pages from the
> internet to include callbook servers, dxcluster, or any other service you
> wish to provide and will interface to http on the 'net.
>
> According to the QST article (written by John Hansen, W2FS) the server
> uses UI frames exclusively, so essentially it runs in broadcast mode all
> the time. So what I gather is that depending on the request and baud rate
> of the requesting station, different information replies to the requester
> are sent (it is up to the requesting stationor server to determine
> efficiency of his/her connection - baud rate, etc.).
>
> In other words, the server understands a request qrz://kb6fst and does a
> callsign lookup. Or html://kb6fst.ampr.org and recieves html markup
> homepage - obviously ineficient at 1200 baud, but graphics would be
> possible at higher speeds. Or http://kb6fst.ampr.org and gets a text only
> reply of the homepage with no markup - for slow connections.
>
> The server is setup to allow/deny types of requests/pages available based
> on the chosen setup and appears to be very flexible, particularly for
> mobile sites. There is a client program for "browsing" available, although
> its possible to use a terminal if you ignore the headers.
>
> There are more examples in the article in QST. It looks very interesting.
> I may have to accelerate my reading of "Learning Perl"
>

Okay it's a file server using ax25 in disconnected mode.  Kind of an aprs
style file server.

Wouldn't this work a lot better?
http://gw.ko6ri.ampr.org/bbs/cgi-bin/callbook.cgi?kb6fst  All the end user
need do is install sv2agw's win98 stuff and point and click there way.  This
really does work at 1200 baud.  The interesting thing about this unconnected
file server is that the image files that reside on the client system could be
sent in that way.  The server could transmit after not hearing anything for
some period of time and pause whenever it heard anything that was not related
to its function.


Bob

Re: ham web

[John Cusick]

On Sat, 16 Oct 1999, Bob Meyer wrote:
> Harold Hartley wrote:
> 
> > you can find the web site at http://www.tapr.org/~wa0ptv and it tells
> > you the concept of it..
> 
 
> My question is;  what does it have to do with web pages?
> 
> Bob

Bob, et. al.

This is sort of a synopsys of the article for those who do not reveive
QST:

It is a configurable server designed for limited web pages from the
internet to include callbook servers, dxcluster, or any other service you
wish to provide and will interface to http on the 'net.

According to the QST article (written by John Hansen, W2FS) the server
uses UI frames exclusively, so essentially it runs in broadcast mode all
the time. So what I gather is that depending on the request and baud rate
of the requesting station, different information replies to the requester
are sent (it is up to the requesting stationor server to determine 
efficiency of his/her connection - baud rate, etc.).

In other words, the server understands a request qrz://kb6fst and does a
callsign lookup. Or html://kb6fst.ampr.org and recieves html markup
homepage - obviously ineficient at 1200 baud, but graphics would be
possible at higher speeds. Or http://kb6fst.ampr.org and gets a text only
reply of the homepage with no markup - for slow connections. 

The server is setup to allow/deny types of requests/pages available based
on the chosen setup and appears to be very flexible, particularly for 
mobile sites. There is a client program for "browsing" available, although
its possible to use a terminal if you ignore the headers.

There are more examples in the article in QST. It looks very interesting.
I may have to accelerate my reading of "Learning Perl"

John C.
KB6FST

Re: ARP & Datagram vs. Virtual Circuit

[Alexandre Fornieles]

Hello,


Robert Schelander wrote:
> 
> I've got a strange ARP problem since I've installed the new
> apps/tools/libax, new kernel everything from the scratch

I helped a friend to do the same and start from scratch with a SuSE 6.2
and the delivered 2.2.10 kernel.

I don't personnally use the AX.25 from 2.2 kernels yet even if i've
already tried it. I'm quite used now to 2.0 AX.25 setups.

I wrote the basic scripts for my friend which set up a KISS interface
with basic IP routing.

I've noticed that ARP won't work. His ARP requests look like this :

[Sat Oct 16 17:16:27 1999]
Port users: AX25: F5OZP->QST  pid=ARP
ARP: len 30 hwtype AX25 prot 0x800 prlen 4 op REQUEST
sender hwaddr F5OZP
target

So it seems that IP related info is corrupted : prot 0x800 and no target
nor sender IP adresses.
This requests never get answered from my 2.0.38 server.
The problem remains if my 2.0.38 (F5KOW) box initiates the ARP request
as even if they're correctly formatted :

[Sun Oct 17 18:11:28 1999]
Port users: AX25: F5KOW-10->QST  pid=ARP
ARP: len 30 hwtype AX25 prot IP op REQUEST
sender IPaddr 44.151.40.40 hwaddr F5KOW-10
target IPaddr 44.151.40.2

My friend's box won't answer by giving its HWAddress.

The only trick to get the machines talk to each other is to manually set
the ARP entry on both machines. When doing that, my friend's box still
won't show a nice "ham" HWAdress but something more similar to an
Ethernet MAC (no Copy/paste available for 'arp -v' on its box).


When this ARP behaviour is "solved" by manually manipulating the ARP
table, i found a problem concerning DG vs. VC, i included the usual
command in the AX.25 script :

echo 1 > /proc/sys/net/ax25/ax0/ip_default_mode

Even if it gets set (checked the /proc files values after entering the
previous command), my friend's Linux box won't talk VC but always DG !
I haven't yet tried to reverse the value form 1 to 0 just in case it
changed from 2.0 or early 2.2 i've personnally tried, but i'm not very
optimistic on the effect it could have...


This is what i had to say/report and i would be glad to here from
someone that had the same problems and manage to really solve them.

Tkanks for taking time to read.
Alex.

-- 
Alexandre Fornieles   |   F-40150 Hossegor
UIN : 1519880 |   AX.25 : f1cay @ f5kow.faqi.fra.eu
-
Software is like sex : It's better when it's Free !

Re: Routing problem [was: ARP problem]

[Tomi Manninen]

On Sun, 17 Oct 1999, Robert Schelander wrote:

> You're right what you've said about ARP. Although the hardwareaddress is
> already in the packet  and it would be a very good idea to add this
> thanksfully to the arp table (like flexnet does) it is vaild to check with
> an ARP request.

It is actually not only valid but necessary because as I just said that
info from an incoming packet should not be used as there is NO guarantee
it's the right info. The fact that Flexnet uses it and thus breaks the
protocol specs doesn't make it the right behaviour.

> The problem is that my system does not even learn the route
> where the packet has came from. So if a new station sends an echo request my
> system is completely blind and disorientated where to get the
> hardwareaddress to reply. (really stupid behaviour with respect to the fact
> that it could get all the information from the received request...)

Blind is exactly as it is supposed to be. It should not make any
assumptions based on that incoming packet. At this point your system does
not have any valid information about the hardware address of the said
other system.

> Of
> course the station mentioned below can answer ARP, but my system sends it to
> QST instead of the real route where it came from. And of course the simple
> digipeater (not an IP router) isn't impressed by a frame addressed to QST
> which isn't even a valid callsign.

QST happens to be the standardized broadcast address that is used for ARP.
So nothing wrong here. Your system doesn't know where to reach the other
system so the only possibility it has is to send a broadcast asking for
the info.

But here I now see the source of your problem, a detail you left
unmentioned earlier. The other station is not directly accessible but
behind a digipeater (ewww...). That changes things. Read below.

> So since it doesn't seem to be possible to tell linux to do arp learning by
> listening, my new question is how to tell it to learn routes.
> 
> Another question: How to setup a default route for a given port.

Ok so now we are talking about AX.25 routes. As a digipeater AX.25 network
is not ARP capable (it has no true broadcast mechanism) we have to depend
on using hacks... That is we need to add an AX.25 (digipeater) route for
the other station and the manually add an ARP entry to use that AX.25
route. 

Here I think ax25rtd might help. I understand it is supposed to do just
that hack automatically. I'm not sure however as I haven't used it.

> By the way, I've tested this VC and it seems to work fine. Thanks again.
> But this insistance to do ARP in a VirtualCircuit is even more strange I
> think. Now it does the connection, even sends an RR to the AX25 including
> the echo request but then it is not able to reply within the already
> establised VC connection!! It would be so easy...

How could it answer there if it doesn't have any trustworthy information
saying it should? IP over AX.25 just doesn't work like that.

-- 
Tomi Manninen   Internet:  [EMAIL PROTECTED]
OH2BNS  AX.25: [EMAIL PROTECTED]
KP20ME04Amprnet:   [EMAIL PROTECTED]

AX.25 kernel and libc dependencies?

[John Ackermann]

Sorry if this repeats earlier discussions on the list, but I'm trying
to tie down some loose ends.

What are the minimum kernel and libc versions required for the new 3-part
AX.25 distribution?  Will the new packages work with a pre-glibc system,
and/or with a 2.0.35 (or patched earlier) kernel, or are they only for
glibc 2.0/kernel 2.2.x?

I'm hoping to derive a table for my linux-ax25 web page, and for my
 TCP/IP book, that shows the appropriate tools to use based on the 
underlying system versions.

Thanks,

John N8UR
[EMAIL PROTECTED]