Hello,

my understanding of the RFC series is, that if an RFC B obsoletes RFC A, then 
RFC B cannot briefly mention a subject and
refer to RFC A for details.

At  https://tools.ietf.org/html/rfc4033 , https://tools.ietf.org/html/rfc4034 
and https://tools.ietf.org/html/rfc4035 is
written that these RFCs together obsolete RFC 3755 and RFC 3757.  The 
obsoleting text is in the Introduction of RFC
4033.

RFC 4033 Section 2 „Definitions of Important DNSSEC Terms“ says: „Key signing 
keys are discussed in more detail in
[RFC3757].“

In “3.1.  Data Origin Authentication and Data Integrity” it states “See 
[RFC3755] for details.”

RFC 4034 contains in section “2.1.1.  The Flags Field”: “Bit 15 of the Flags 
field is the Secure Entry Point flag,
described in [RFC3757].”

So RFC 4033 obsoletes RFC 3757, mentions briefy the term “Key signing keys” and 
refers to RFC 3757 for details.

This way of describing adds additional complexity for the reader to understand 
DNSSEC, as it is not clear in which order
to read the RFCs to understand the topics and leaves unclear, whether obsoleted 
RFCs shall be read.

--
I have a question about https://tools.ietf.org/html/rfc4035#appendix-A “Signed 
Zone Example” for the zone “example.”:

The zone contains:

  3600 DNSKEY 256 3 5 (
    AQOy1b*Yvh0z2542lzMKR4Dh8uZffQ==
  ) ;{id = 38519 (zsk), size = 1024b}
  3600 DNSKEY 257 3 5 (
    AQOeX7+*k5/j8vfd4jWCWD+E1Sze0Q==
  ) ;{id = 9465 (ksk), size = 1024b}
  3600 RRSIG  DNSKEY 5 1 3600 20040509183619 (
    20040409183619 9465 example.
    ZxgauAuI*1Tp4VKDqxqG7R5tTVM=
   )
  3600 RRSIG  DNSKEY 5 1 3600 20040509183619 (
    20040409183619 38519 example.
    eGL0s90g*aFzHKillDt3HRxHIZM=
  )

For brevity I have replaced some base64-encoded text with *.  The id=(zks/ksk) 
comments are not part of the RFCs.  I 
extracted them with ldns-read-zone.

The DNSKEY 256 is the Zone-Signing-Key with id 38519 and 257 is the 
Key-Signing-Key id 9465.

My understanding is that the root zone (.) has signed one DS record for the 
example. zone and this DS record in practice
refers to the 257 Key-Signing-Key.  Then the resolver validates that the hash 
in the DS record matches a KSK DNSKEY in
the zone appex and that this DNSKEY validates the RRSIG.  The valid RRSIG 
validates and covers all DNSKEYs, meaning the
ZSK DNSKEY is validated this way.

Why does example. zone contain a RRSIG for DNSKEY that are signend by the ZSK?

Compare to DS IANA.ORG, which returns currently six DS records for three 
distinct DNSKEYs with labels 6730, 14351 and
39817 (delv +short DS IANA.ORG):

6730 8 1 15A1EE709E51C0652799D9CD364A37ACB8E67285
6730 8 2 FCA03776FA33AB57369AAE8B089938F4001E208094362734242A7EBE 4DDB5755
14351 8 1 4C08C05754DDA7CCF62B41AC450873D271221CC3
14351 8 2 E1A0C92C0663202240FC3481F747D00901CD14E268B884A1142D7A47 5DB1A810
39817 8 1 4B2634F69EDD8D0741DA8629E2064B05E8138E51
39817 8 2 5FB72AE84C35A35DDF7FA24EA791CD9763C4360979C0CBB6FF2C407A F9AE5802


delv +dnssec DNSKEY IANA.ORG   returns:

DNSKEY  256 3 8 AwEAAbS+H*zJr  ; ZSK; alg = RSASHA256 ; key id = 41470
DNSKEY  256 3 8 AwEAAc0xH*yrP  ; ZSK; alg = RSASHA256 ; key id = 1723
DNSKEY  257 3 8 AwEAAa02O*qU=  ; KSK; alg = RSASHA256 ; key id = 21911
DNSKEY  257 3 8 AwEAAb0i1*Ik=  ; KSK; alg = RSASHA256 ; key id = 39817
RRSIG   DNSKEY 8 2 3600 20191127114538 20191106083616 21911 iana.org. bdC*3A==
RRSIG   DNSKEY 8 2 3600 20191127114538 20191106083616 39817 iana.org. ADP*XEQ==

So the RRSIG are signed only by KSKs (21911 and 39817) and the ZSK are not used 
to generate RRSIG for the DNSKEYs.  I
guess label 39817 is in a roll-over process and will disappear soon.


Why the example from RFC 4035 Appendix A generats RRSIG for DNSKEY signed by 
the ZSK, but for IANA.ORG there is no such
RRSIG?

If this is not the right mailing list, which is the right one?

Greetings
  Дилян

_______________________________________________
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to