[nos-bbs] Background JNOS operation

[Mark Phillips]

Screen geometry is your problem here.

Make sure you resize your terminal window to 80x24. 


If you look carefully you'll find JNOS has respawned too.

I run my JNOS on my headless WRAP machine in the same manner as you
describe. It works flawlessly until I log in with the wrong terminal
size.

Mark

 
On Fri, 2009-02-20 at 13:58 -0500, George [Skip] VerDuin wrote:
> Greetings.
> 
> The next step at K8RRA is to encourage JNOS to live peacefully in the 
> background.  A "headless" computer is the further target.  Using 
> "screen" as a detachable/re-attachable link to gain access to the sysop 
> console almost works nicely.  Also, jnos&screen seem to be just happy 
> operating from a background process.
> 
> Here is an interaction between jnos--screen that is misbehaving.  When 
> started as:
>      ...# screen -d -m  jnos ...
> the processes start up nicely and are happy until attaching for the 
> *first time* with:
>      ...# screen -r
> Wherein jnos issues the message "NOS PANIC  Lost keyboard (-1/4)" 
> followed by an automatic restart of jnos.  The second killer is that the 
> restart initiates "tun1" rather than tun0 [for an unknown reason] which 
> kills the link from the new instantiation of jnos and the OS -- Fedora 
> in my case.  The "ifconfig" shows tun0 to be active and tun1 to be 
> non-existant [another mystery].
> 
> The non-detached jnos startup is very normal when started with:
>     ...# screen  jnos ...
> In both this case, and with the "-d -m" startup above, the manual 
> detach-then-reattach sequence  work nicely with no ill effects found 
> here thus far...
> 
> SO -- the question I pose is "why the PANIC?".  The message originates 
> from curses.c and I wonder why curses is bothered by attachment of the 
> keyboard later than initial startup [but before the first character is 
> typed].  Thus far, it seems to me I must revise the jnos code to 
> suppress the panic, and perhaps fix the tun1/tun0 issue upon restart too.
> 
> The more interesting question is "Has anyone found a work-around for 
> this problem to avoid revising jnos source?".
> 
> Cheers to all & 73.
> Skip
> 
> _______________________________________________
> nos-bbs mailing list
> nos-bbs at tapr.org
> https://www.tapr.org/cgi-bin/mailman/listinfo/nos-bbs
-- 



Mark Phillips, G7LTT/NI2O
Randolph, NJ