ntpd core dumping on 5.3-p5

[Bill Moran]

John Pettitt <jpp at cloudview.com> wrote:
> 
> Bill Moran wrote:
> 
> >John Pettitt <jpp at cloudview.com> wrote:
> >  
> >>Bill Moran wrote:
> >>
> >>>I'm experiencing a problem similar to the problem described in this thread:
> >>>http://lists.freebsd.org/pipermail/freebsd-questions/2005-February/078139.html
> >>>
> >>>My ntp.conf contains:
> >>>server clock.psu.edu
> >>>server fuzz.psc.edu prefer
> >>>server ntp-1.ece.cmu.edu prefer
> >>>server ntp-2.ece.cmu.edu prefer
> >>>
> >>>If I comment out the first two servers, ntpd works fine.  Otherwise, I get
> >>>coredumps.
> >>>
> >>>It seems a little extreme to me that any server could remotely cause ntpd
> >>>to core.  I'm willing to look in to this, but I'm having trouble figuring
> >>>out how to get ntp built with debugging symbols.
> >>>
> >>>FreeBSD bolivia.potentialtech.com 5.3-RELEASE-p5 FreeBSD 5.3-RELEASE-p5 #0: Thu Feb 24 08:26:07 UTC 2005     root@:/usr/obj/usr/src/sys/WORKING  i386
> >>>

<snip>

> Does it core dump as soon as it starts  or after a while? 

It consistenly takes two seconds to coredump.

> What do ntpq -c rv and ntpq -c peer say?

Not really enough time to issue those commands before ntpd cores.

-- 
Bill Moran
Potential Technologies
http://www.potentialtech.com