Re: minidlna buffer overflow on netbsd-8/amd64

["John D. Baker" <jdbaker%consolidated.net@localhost>]

On Sun, 27 Oct 2019, Martin Husemann wrote:

> Can you build it with -g and start it manually from gdb? That should point
> out the problem immediately. (Since it likely depends on your library, it
> might not be that easy to reproduce for others)

I'll need a bit more guidance here.  Simply running it manually with
the same arguments as it would have when started by its rc script simply
said:

  (gdb) set args -R -f /usr/pkg/etc/minidlna.conf
  (gdb) run
  Starting program: /usr/pkg/sbin/minidlnad -R -f /usr/pkg/etc/minidlna.conf
  [Inferior 1 (LWP 0) exited normally]
  (gdb) bt
  No stack.

but there was still a syslog entry with matching timestamp that reported
the overflow/terminated condition.

The main(?) minidlnad process is still present even after exiting.

So, perhaps I need gdb to follow fork() and look at the process it spawned?
I'm still quite the neophyte with respect to gdb.

> Martin


-- 
|/"\ John D. Baker, KN5UKS               NetBSD     Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net  OpenBSD            FreeBSD
| X  No HTML/proprietary data in email.   BSD just sits there and works!
|/ \ GPGkeyID:  D703 4A7E 479F 63F8 D3F4  BD99 9572 8F23 E4AD 1645