Date: 19 Dec 1999 00:52:23 -0000
Message-ID: <19991219005223.16101.qmail@cr.yp.to>
Mail-Followup-To: namedroppers@internic.net
From: "D. J. Bernstein" <djb@cr.yp.to>
To: namedroppers@internic.net
Subject: *.143.38.in-addr.arpa

The servers for 143.38.in-addr.arpa respond to a PTR query for
65.113.143.38.in-addr.arpa with the nonsensical referral

   *.143.38.in-addr.arpa NS pri1.dns.psi.net
   *.143.38.in-addr.arpa NS pri2.dns.psi.net
   *.143.38.in-addr.arpa NS pri3.dns.psi.net

Is this the symptom of a common configuration error or BIND bug? Does
BIND take any special action as a client to work around such problems?

According to RFC 1034, this reply ``is bogus and should be ignored.''
But BIND passes it along to the client---violating the RFC 1034 rule
that an RD+RA response must be an answer, an NXDOMAIN, or a temporary
failure. Why doesn't BIND say SERVFAIL here?

---Dan