[nos-bbs] kam induced crash?

[(Skip) K8RRA]

Hi guys.
Here is another "don't do this" for the moment.

I crashed jnos while working out the details of attaching a Kantronics
KAM-98 TNC to the hfdd port for Pactor I communications.  The attached
files:
  a) screen.dump4 is the sysop console record
  b) screen.dump3 is the gdb record with back trace.
What I actually did (hairbrain move) is to duplicate the hfdd commands
already executed by autoexec.nos.  After I stopped re-issuing the start
command the crash stopped.  >> It may be a "bug" anyway?

There is another matter with hfdd - I don't believe it can tell when it
is in fact in HOST MODE.  The third file hfdd-kam.log is an extract from
todays log for examination.  SEE: the string of messages "...NOT in host
mode..."?  I can vouch for the TNC really being in host mode by
inspection with minicom after shutting down jnos.

As a last matter I needed to change the kam.cfg file a little and put
the RESTORE on the human list of actions before starting jnos.  The
changes should be self explanatory and I will continue to attempt to get
the autobaud pain to work reliably over the next days.


73
de [George (Skip) VerDuin] K8RRA k
-------------- next part --------------
16:33 001720 CONV=0 LNKS=1
BBS:
0 Command:
Copyright 1991 by Phil Karn (KA9Q) and contributors.
Callsign logging by source enabled, by destination disabled
Pactor: 1
FullDup: 0
Persist: 31
SlotTime: 15
TxDelay: 77
TxTail: 1
*** file "etc/rf2q.conf", line 35: echo "> vhf & kam RF Ports complete"

*** file "etc/netul.conf", line 46: echo ">TCP/IP network port is open"

SIOCADDRT: File exists
                      *** file "etc/route_local", line 48: echo >Local Routes fo
vhf: input output (ASCII dump) trace file: trace/v1
kam: input output (ASCII dump) trace file: trace/hk
tun0: input output (ASCII dump) trace file: trace/t
input line: echo > Servers are up and running
input line: echo >>>>>>>>>>>> Have a great session starting NOW! <<<<<<<<<<<<
jnos>
jnos> hfdd debug
jnos> hfdd server kam start
-------------- next part --------------
Loaded symbols for /lib/libc.so.6
Reading symbols from /lib/libtinfo.so.5...done.
Loaded symbols for /lib/libtinfo.so.5
Reading symbols from /lib/libdl.so.2...done.
Loaded symbols for /lib/libdl.so.2
Reading symbols from /lib/ld-linux.so.2...done.
Loaded symbols for /lib/ld-linux.so.2
0x00110402 in __kernel_vsyscall ()
(gdb) c
Continuing.

Program received signal SIGSEGV, Segmentation fault.
0x00b4b407 in free () from /lib/libc.so.6
(gdb) bt
#0  0x00b4b407 in free () from /lib/libc.so.6
#1  0x080b6216 in j_free (p=0x80e249f) at unix.c:281
#2  0x08093fa8 in killproc (pp=0x8c07130) at kernel.c:173
#3  0x080940d3 in killer (i=0, v1=0x0, v2=0x0) at kernel.c:215
#4  0x080948d6 in _kicker (func=0x809406b <killer>, iarg=0, parg1=0x0,
    parg2=0x0) at ksubr.c:116
#5  0x00b18c24 in makecontext () from /lib/libc.so.6
#6  0x0809406b in killself () at kernel.c:196
#7  0x0808aee8 in usflush (s=131) at sockuser.c:678
#8  0x0804ae04 in main (argc=2, argv=0xbfefd164) at main.c:674
(gdb)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hfdd-kam.log
Type: text/x-log
Size: 1906 bytes
Desc: not available
URL: <http://lists.tapr.org/pipermail/nos-bbs_lists.tapr.org/attachments/20071013/5d2a3d18/attachment.log>
-------------- next part --------------
#
# Default KAM initialization file for JNOS 2.0e (new KAM code)
# Modified for K8RRA per experience  today: 10/13/07
#
## RESTORE DEFAULT
## *
## *
## *
## *
## K8RRA
INTFACE TERMINAL
AUTOLF OFF
# SHIFT MODEM
# MARK
# SPACE
LFSUP  OFF
LFADD  OFF
CRADD  OFF
CRSUP  OFF
ARQTIME 1
INTFACE HOST
RESET