According to Thorsten Kranzkowski: While burning my CPU.
>
> > kissattach needs it.
I would imagen its needed because thats the way things were written when
ax25/netrom was first introduced into the kernel, many things have changed,
but the ax25 system is still based on the way it was then.
We now have /proc/sys/net/ax_interfaces which was not presant at the start.
Many other things have changed, things are still changing and it seems that
we are going to see a complete new ax25-utils set very shortly, according
to messages to this list a while back.
>
> again: use ifconfig
>
> The tty argument will typically be that of a serial port
> with a KISS TNC attached, although it could be a psuedo
> tty or a KISS port emulator such as an SCC card.
>
> tty argument _is_ obviously required - but one already passes it as a
> commandline parameter and not as an axports entry.
When using ifconfig to up my SCC interface the port and callsign has to be
in axports.
>
> > The example above means, assign the callsign df0pt to the user account "tk"
>
> That's exactly the point: the 'user' essentially makes an outgoing call on an
> interface, not on a 'port bound to a callsign' - he doesn't even use that
> callsign!
He could, its up to you, if you as system operator want to comply with your
countries requirments ie, its your system transmitting so realy your
callsign should be used, you just assign your callsign+ssid to his user
account, or his callsign+15 but that is another story.
>
> > Only root can do that so where's the problem.?.
> The example wasn't well thought.
> Of course, usually "axparms -assoc" will happen somewhere in /etc/rc.d/...
>
> >
> > >
> > > I think all the information in axports is superfluous:
Node is another program which obtains its port information from axports.
> > >
> > > Disadvantages of axports usage:
> > > - axutils stop working when you assign the same call to two or more
> > > interfaces
> >
> > Not true or at least you can have the same callsign on every port, however
>
> again - I do know how to handle that in the present design - I do
> complain about the design!
If you would like to rewrite all the ax25 code then change it, no one can
stop you and no one will complain, as to changes being accepted by the AX25
maintainer is another matter.
There is always room for "improvement", but if it works well as it is, why
break it.
>
> > NOT in axports as such, use the VIA option in ax25d.conf when using ax25d.
>
> Imagine:
> /sbin/ifconfig bcsf0 hw ax25 dl8bcu up
> /sbin/ifconfig bcsf0 inet 44.130.8.19 mtu 1500
> /sbin/ifconfig bcsf1 hw ax25 dl8bcu up
> /sbin/ifconfig bcsf1 inet 44.130.8.19 mtu 1500
>
> Now you have two interfaces with identical Hardwere addresses - which
> is perfectly ok for tcpip. But anything that uses axports (e.g. 'call') now
> gets confused because it cannot decide which interface to use. So it
> stops with an error message.
You can achive the above via ax25d.conf but every interface will need
another ssid when attaching them.
> > Same here use ax25d to have ONE AX25 call + ssid the same one listning on
> ---------- I disagree.
I dont think a disscussion along these lines will get very far.
>
> > O no, if you do that, you will have many more problems that you think,
> > espesaly with 2.2.x kernels.
>
> ok - what problems?
> and what has that to do with 2.2.x kernels?
Take a look at the linux-hams archive,
http://hes.iki.fi/archive/linux-hams thier you will find a number of
messages explaning why you need to use -i at least.
>
> > > What do you think about this?
> > > What did I miss? Where am I wrong?
> >
> > Possably ax25d and the use thereof.
>
> You told me several times _how_ to configure this. That was not my question.
> I wanted to know, _why_ are we doing so, as I don't see the point!
Thorsten, i do see some of your points, however it all comes down to the
axports file being needed by most apps and programs in the ax25 collection.
>
> So please enlighten me ! :-)
>
> vy73 Thorsten
>
> --
> | Thorsten Kranzkowski Snail: Niemannsweg 30, 49201 Dissen, Germany |
> | Mobile: ++49 161 7210230 Inet: [EMAIL PROTECTED] |
> | Ampr: dl8bcu@db0nei.#nrw.deu.eu, [EMAIL PROTECTED] [44.130.8.19] |
>
--
Regards Richard.
[EMAIL PROTECTED]
