> Now that I got my compile issues straightened, I've been playing with
> ethereal-0.9.1 on a Solaris 8 box. I accidentally hit print (I don't even
> have lp running on this box yet),
"Print" doesn't necessarily require a printer; you can print to a file
as well.
> and it crashed ethereal.
>
> Ethereal died with the message:
> ** ERROR **: file print.c: line 142 (get_field_data): should not be reached
> aborting...
>
> Here's the backtrace from the core file:
>
> #0 0xfed9bb54 in _libc_kill () from /usr/lib/libc.so.1
> (gdb) backtrace
> #0 0xfed9bb54 in _libc_kill () from /usr/lib/libc.so.1
> #1 0xfed357c4 in abort () from /usr/lib/libc.so.1
> #2 0xff0593d4 in g_logv (log_domain=0x0, log_level=G_LOG_LEVEL_ERROR,
> format=0x0, args1=0x6) at gmessages.c:389
> #3 0xff05946c in g_log (log_domain=0x0, log_level=G_LOG_LEVEL_ERROR,
> format=0x2c82c0 "file %s: line %d (%s): should not be reached")
> at gmessages.c:406
> #4 0x1602c4 in proto_tree_print ()
> #5 0x16039c in proto_tree_print ()
Either the debugger is broken, or some files weren't compiled with "-g",
or something such as that; unfortunately, this means that we probably
can't get any further information from the debugger.
Did you read a capture file, or did you do a live capture? If you did a
live capture, the capture file is probably in "/tmp" or "/var/tmp" with
a name beginning with "etherXXXX". In either case, save a copy of the
capture file, and try running
tethereal -V -r {the capture file name} >/dev/null
and see if it crashes.
We may need to get a copy of the capture file in order to debug this
further.
