One document matched: draft-ietf-dnsind-edns-01.txt
Differences from draft-ietf-dnsind-edns-00.txt
DNSIND Working Group Paul Vixie
INTERNET-DRAFT Vixie Enterprises
<draft-ietf-dnsind-edns-01.txt> March, 1998
Extensions to DNS (EDNS)
Status of this Memo
This document is an Internet-Draft. Internet-Drafts are working
documents of the Internet Engineering Task Force (IETF), its areas,
and its working groups. Note that other groups may also distribute
working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as ``work in progress.''
To learn the current status of any Internet-Draft, please check the
``1id-abstracts.txt'' listing contained in the Internet-Drafts Shadow
Directories on ftp.is.co.za (Africa), nic.nordu.net (Europe),
munnari.oz.au (Pacific Rim), ds.internic.net (US East Coast), or
ftp.isi.edu (US West Coast).
Abstract
The Domain Name System's wire protocol includes a number of fixed
fields whose range has been or soon will be exhausted. This document
describes backward compatible mechanisms for allowing the protocol to
grow.
1 - Rationale and Scope
1.1. DNS (see [RFC1035]) specifies a Message Format and within such
messages there are standard formats for encoding options, errors, and
name compression. The maximum allowable size of a DNS Message is fixed.
Many of DNS's protocol limits are too small for uses which are or which
are desired to become common.
1.2. Existing clients will not know how to interpret the protocol
extensions detailed here. In practice, these clients will be upgraded
when they have need of a new feature, and only new features will make
Expires September 1998 [Page 1]
INTERNET-DRAFT EDNS March 1998
use of the extensions. We must however take account of client behaviour
in the face of extra fields, and design a fallback scheme for
interoperability with these clients.
2 - Affected Protocol Elements
2.1. The DNS Message Header's (see [RFC1035 4.1.1]) second full 16-bit
word is divided into a 4-bit OPCODE, a 4-bit RCODE, and a number of
1-bit flags. The original reserved Z bits have been allocated to
various purposes, and most of the RCODE values are now in use. More
types and more possible RCODEs are needed.
2.2. The first two bits of a wire format domain label are used to denote
the type of the label. [RFC1035 4.1.4] allocates two of the four
possible types and reserves the other two. Proposals for use of the
remaining types far outnumber those available. More label types are
needed.
2.3. Compression pointers are 14 bits in size and are relative to the
start of the DNS Message, which can be 64KB in length. 14 bits restrict
pointers to the first 16KB of the message, which makes labels introduced
in the last 48KB of the message unreachable by compression pointers. A
longer pointer format is needed.
2.4. DNS Messages are limited to 512 octets in size when sent over UDP.
While the minimum maximum reassembly buffer size is still 512 bytes,
most of the hosts now connected to the Internet are able to reassemble
larger datagrams. Some mechanism must be created to allow requestors to
advertise larger buffer sizes to responders.
2.5. DNS Messages are limited to 65535 octets in size when sent over
TCP. This acts as an effective maximum on RRset size and on RR size,
since multiple TCP messages are only possible in the case of zone
transfers. Some mechanism must be created to allow normal DNS responses
(other than zone transfers) to span multiple DNS Messages when TCP is
used.
2.6. Multiple queries in a question section have not been supported in
DNS due the applicability of some DNS Message Header flags (such as AA)
and of the RCODE field only to a single QNAME, QTYPE, and QCLASS.
Multiple questions per request are desirable, and some way of asking
them must be made available.
Expires September 1998 [Page 2]
INTERNET-DRAFT EDNS March 1998
3 - Extended Label Types
3.1. The ``1 0'' label type will now indicate an extended label type,
whose value is encoded in the lower six bits of the first octet of a
label. All subsequently developed label types should be encoded using
an extended label type. The ``0 1'' label type should be permanently
reserved.
3.2. The ``0 0 0 0 0 0'' extended label type will indicate an extended
compression pointer, such that the following two octets comprise a
16-bit compression pointer in network byte order. Like the normal
compression pointer, this pointer is relative to the start of the DNS
Message.
3.3. The ``0 0 0 0 0 1'' extended label type will indicate a counted bit
string label with interior longest-match query matching semantics as
described in [CRAW98].
3.4. XXX Any other former contenders for the unextended label types? XXX
4 - OPT pseudo-RR
4.1. The OPT pseudo-RR can be added to the additional data section of
either a request or a response. An OPT is called a pseudo-RR because it
pertains to a particular transport level message and not to any actual
DNS data. OPT RRs shall never be cached, forwarded, or stored in or
loaded from master files.
4.2. An OPT RR has a fixed part and a variable set of options expressed
as {attribute, value} pairs. The fixed part holds some DNS meta data
and also a small collection of new protocol elements which we expect to
be so popular that it would be a waste of wire space to encode them as
{attribute, value} pairs.
4.3. The fixed part of an OPT RR is structured as follows:
Field Name Field Type Description
-----------------------------------------------------
NAME domain name empty (root domain)
TYPE u_int16_t OPT (XXX code)
CLASS u_int16_t sender's UDP buffer size
TTL u_int32_t extended RCODE and flags
RDLEN u_int16_t describes RDATA
RDATA octet stream {attribute,value} pairs
Expires September 1998 [Page 3]
INTERNET-DRAFT EDNS March 1998
4.4. The variable part of an OPT RR is encoded in its RDATA and is
structured as zero or more of the following:
+0 (MSB) +1 (LSB)
+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
0: | OPTION-LENGTH | OPTION-CODE |
+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
2: | |
/ OPTION-DATA /
/ /
+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
OPTION-LENGTH Size (in octets) of OPTION-DATA.
OPTION-CODE Assigned by the IANA. Value 255 is reserved for future
expansion.
OPTION-DATA Varies per OPTION-CODE.
4.5. The sender's UDP buffer size is the number of octets of the largest
UDP payload that is reassemblable and deliverable in the sender's
network stack. Note that path MTU, with or without fragmentation, may
be smaller than this. Also note that a 512-octet UDP payload requires a
576-octet IP reassembly buffer. Choosing 1436 on an Ethernet connected
requestor would be reasonable.
4.6. The extended RCODE and flags are structured as follows:
+0 (MSB) +1 (LSB)
+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
0: | EXTENDED-RCODE | VERSION |
+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
2: |MD |FM | Z |
+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
EXTENDED-RCODE Forms upper 8 bits of extended 12-bit RCODE.
(Meaningless in requests.)
VERSION Indicates the implementation level of whoever sets it.
Full conformance with the draft standard version of this
specification is version ``0.''
Expires September 1998 [Page 4]
INTERNET-DRAFT EDNS March 1998
Z Set to zero by senders and ignored by receivers, unless
modified in a subsequent specification.
MD ``More data'' flag. Valid only in TCP streams where
message ordering is reliable. This flag indicates that
the current message is not the complete response, and
should be aggregated with the following message(s)
before being considered complete. Such responses are
called ``segmented responses.'' It is an error for the
RCODE (including the EXTENDED-RCODE), AA flag, or DNS
Message ID to differ among messages in a segmented
response. It is an error for TC to be set on any
message in a segmented response.
FM ``First match'' flag. Notable only when multiple
questions are sent. If set, questions will be processed
in wire order and the first question whose answer would
be NOERROR AND ANCOUNT>0 is treated as if it were the
only question in the query message. If not set,
questions can be processed in any order and all possible
answer records will be included in the response.
5 - Multiple Questions
5.1. If QDCOUNT>1, multiple questions are present. All questions must
be for the same QNAME and QCLASS; only the QTYPE is allowed to vary. It
is an error for QDCOUNT>1 and any QTYPE=ANY.
5.2. RCODE and AA apply to all RRs in the answer section having the
QNAME that is shared by all questions in the question section.
5.3. QCLASS=ANY is deprecated, for all queries including those with
QDCOUNT=1.
5.4. QTYPE=IQUERY is deprecated, for all queries including those with
QDCOUNT=1.
6 - Transport Considerations
6.1. The presence of an OPT pseudo-RR or any new label type, or
QDCOUNT>1 in a request should be taken as an indication that the
requestor fully implements this specification and can correctly
understand any response that conforms to this specification. If a new
label type or QDCOUNT>1 is used in a message that does not have an OPT
RR, a VERSION of ``0'' shall be imputed.
Expires September 1998 [Page 5]
INTERNET-DRAFT EDNS March 1998
6.2. Lack of use of these features in a request must be taken as an
indication that the requestor does not implement any part of this
specification and that the responder may make no use of any protocol
extension described here in its response.
6.2. Responders who do not understand these protocol extensions are
expected to send a respose with RCODE=NOTIMPL or RCODE=FORMERR or even
RCODE=SERVFAIL. Therefore use of these extensions should be ``probed''
such that a responder who isn't known to support them be allowed a retry
with no extensions in use if it responds with one of the above mentioned
RCODEs. Responder capability with respect to these extensions ought to
be cached by requestors, but a new probe should be sent periodically to
test for upgrades to responder capability.
7 - Security Considerations
No new problems are added to DNS's security model by this specification.
Correspondingly, no old problems in DNS security are resolved by this
specification.
8 - Acknowledgements
Paul Mockapetris, Mark Andrews, Robert Elz, Don Lewis, Bob Halley, and
Donald Eastlake were each instrumental in creating this specification.
9 - References
[RFC1035] P. Mockapetris, ``Domain Names - Implementation and
Specification,'' RFC 1035, USC/Information Sciences
Institute, November 1987.
[CRAW98] M. Crawford, ``Binary Labels in the Domain Name System,''
Draft draft-ietf-dnsind-binary-labels-XX, IETF DNSIND, March
1998.
10 - Author's Address
Paul Vixie
Vixie Enterprises
950 Charter Street
Redwood City, CA 94063
+1 650 779 7001
<paul@vix.com>
Expires September 1998 [Page 6]
| PAFTECH AB 2003-2026 | 2026-04-23 11:21:18 |