RFC Errata
RFC 1035, "Domain names - implementation and specification", November 1987
Note: This RFC has been updated by RFC 1101, RFC 1183, RFC 1348, RFC 1876, RFC 1982, RFC 1995, RFC 1996, RFC 2065, RFC 2136, RFC 2181, RFC 2137, RFC 2308, RFC 2535, RFC 2673, RFC 2845, RFC 3425, RFC 3658, RFC 4033, RFC 4034, RFC 4035, RFC 4343, RFC 5936, RFC 5966, RFC 6604, RFC 7766, RFC 8482, RFC 8490, RFC 8767, RFC 9619
Source of RFC: LegacyArea Assignment: int
Errata ID: 5915
Status: Held for Document Update
Type: Editorial
Publication Format(s) : TEXT
Reported By: Alexander Dupuy
Date Reported: 2019-11-21
Held for Document Update by: Warren Kumari (Ops AD)
Date Held: 2021-01-26
Section 6.2 says:
When a response is so long that truncation is required, the truncation should start at the end of the response and work forward in the datagram. Thus if there is any data for the authority section, the answer section is guaranteed to be unique.
It should say:
When a response is so long that truncation is required, the truncation should start at the end of the response and work forward in the datagram. Thus if there is any data for the authority section, the answer section is guaranteed to be complete.
Notes:
It's not clear what it might mean for an answer section to be unique. However, by following the algorithm described of removing RRs from the back to the front, if any RRs remain in the authority (or additional) section, the answer section is guaranteed to be complete.
[ See thread at: https://mailarchive.ietf.org/arch/msg/dnsop/L_yjf4eyDRlkIOqaWULf1HUK8f0/ ]