What are these messages

[Mark_Andrews at isc.org]

> In article <199907030036.KAA02379 at bsdi.dv.isc.org>,
>  <Mark_Andrews at isc.org> wrote:
> >
> >> 
> >> John Tan <d_name at hotmail.com> wrote in message
> >> news:19990701034621.47617.qmail at hotmail.com...
> >> Jul  3 01:10:48 tornado named[340]: Lame server on 'isc.org' (in 'org'?):
> >> [139.130.4.5].53
> >
> >	This your forwarder being reported as lame.  Note the lack of a
> >	domain name after [139.130.4.5].53.  139.130.4.5 is ns1.telstra.net.
> 
> How can a forwarder be lame?  It's not expected to be authoritative in the
> first place.

	Bug.  RD was not set on the outgoing message.  The answer came in
	was a referral that pointed higher up or to the same level in the
	heirachy which matches the requirement for a lame server.

	With 8.2.1 forwarders are theoretically more reliable as they will
	be retried and not just pulled off the top of a linked list.  They
	are added to the front of the array where the addresses for the 
	NS's are added.  Each entry has a bit field to say if it a forwarder
	or not.  The bit field from the wrong entry gets read on the first
	outgoing message.

	See the patch posted seperately (n->0).

	A forwarder with recursion allowed turned off will still be flaged
	as lame (which it is for the job it is supposed to be performing).
> 
> I assumed the missing domain name was because he didn't paste the entire
> log message into his post.
> 

	Mark
	
--
Mark Andrews, Internet Software Consortium
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: marka at isc.org