[nos-bbs] JNOS on Debian Etch

[George [Skip] VerDuin]

Hi all.

Recently I fell on to a couple details where two Linux distros act 
differently.
Fedora 10 is the OS for my station and my neighbors.
Debian Etch is one of several distros I'm using for a project.

The first is seen on the establishment of tun bridge.  In the statement:
    ...ifconfig tun0 192.168.1.32 pointtopoint 44.102.132.20 mtu 1500 up
it matters if the address carries a preceding zero.  This means that
  a) 44.102.132.20           is OK
  b) 44.102.132.020        fails -- resulting in an IP of 44.102.132.39
as seen later in both "ifconfig" and "route".  When it fails, the 
session acts badly.

The second is seen in telnet to JNOS.  In the statement:
    ...$ telnet 44.102.132.20
it matters if the user session originates at the host [as a "tty"] or as 
a remote login via ssh [as a "pts/0"].  In this case, if the user is 
logged on locally from the host console using one of the ALT-CTL-Fn 
keys, the user may run as many sessions of JNOS as he wants.  However if 
the user logs in from a remote location using SSH then the "telnet" 
command responds:
    ... Connection refused.
After this refusal happens, subsequent logins at the host console [just 
exactly as before] fail with the same message.  The only solution I 
found is to restart JNOS.

Both of the examples are seen on the Etch OS and not seen on F10 OS.  
This is important [to me] because it *seems* to disadvantage Etch as a 
platform for JNOS.

Anyone else share this experience?
73
Skip