[nos-bbs] Interesting (only to me?) issues about JNOS

[George (Skip) VerDuin]

Greetings to the "how it works" group.

I notice a couple things while watching jnos do it's thing...
I offer what follows as a couple questions that might have obvious
answers if I had been around longer than my year with nos.

Today at my k8rra node:
I show BBS=2 k8rra and wa8rsa signed in and idle on the console header
and similar for the BBS "mbox" command.
At the same time "ax25 status" shows no active sessions (k8rra is a
telnet login) - wa8rsa is a netrom login.
At the same time "netrom status" shows wa8rsa "connected" with Snd-W=3
Snd-Q=0 Rcv-Q=0.
Now this shows unfinished business with wa8rsa here at my node.
A number of circumstances may have led to this situation, but is is
proper for my node to be passive?
Have I got a configuration problem to fix?
Should I see my node later attempt to re-connect and pass on the SND-W=3
packets (top-down) or has wa8rsa had disconnect message showing he
should reconnect to get missing information (bottom-up)?

Consistently day to day:
I see TCP/IP traffic on the RF link to wa8rsa,  it does not matter what
the traffic content is.
My trace does show an ARP to establish a AX.25 link from k8rra to
wa8rsa, and TCP/IP content encapsulated into valid AX.25 packets.
While this is active my console "ax25 status" command does not list
wa8rsa as any status as all.
My read of everything technical says to me that it is proper to consider
IP encapsulated into AX.25 for this traffic.
My read of this reflector is more ambiguous - some hint encapsulation,
some hint separate.
If encapsulation is correct, then I suggest that the status command
should show the wa8rsa-k8rra link established.
It might be useful to designate the ax is driven by IP (or also
netrom?), but is it accurate to find it missing?

Tying the two observations together:
I see a relationship between NET/ROM and AX.25 where net/rom "rides on
top of" ax.25 to carry content.
I see less relationship between TCP/IP and AX.25 until I look at the
packet traces, then pid=IP is similar to pid=netrom.
So why do we treat these two forms of traffic on AX differently?

One more loosely related subject:
Without doing an audit to compute statistics, my intuitive read on
nos-bbs subject matter suggests application examples resolve many
questions.

      * This is easily true for newbies.  To have a "go-by" example to
        pick from a list and tweek into their own needs might speed
        deployment?
      * Displayed by feature maybe.  To have a net/rom specific example
        with ax.25 prerequisite settings would speed troubleshooting?
      * Inclusive of related files.  To have sample popuser file with
        user example commands speeds both configuration and education?

My question becomes: If there are shining examples of this concept of
"an example is worth 1000 words?", where might they be?
Is there support for the idea that more examples of successful installs
would benefit our community even more than command detail?
I'm 1/2 tempted to boil down nos-bbs reflector threads into FAQ examples
for publication...

The above may go outside the box but seems (to me) relevant to building
a broader user base for jnos.
Have I suggested that nos needs a little de-geeking?  I do hope no
offense is taken.

73
de Skip k8rra k



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.tapr.org/pipermail/nos-bbs_lists.tapr.org/attachments/20060828/abf1592c/attachment.html>