One document matched: draft-ietf-homenet-dncp-01.xml
<?xml version='1.0' ?>
<!--
Created: Mon Nov 18 17:55:22 2013 mstenber
split from draft-ietf-homenet-hncp-03-pre - generic parts
-->
<!DOCTYPE rfc SYSTEM 'rfc2629.dtd'>
<?rfc autobreaks="yes"?>
<?rfc compact="yes"?>
<?rfc strict='yes'?>
<?rfc subcompact="no"?>
<?rfc symrefs="yes"?>
<?rfc toc="yes"?>
<?rfc tocindent="yes"?>
<rfc
ipr='trust200902'
docName='draft-ietf-homenet-dncp-01'
category='std'
>
<front>
<title abbrev="Distributed Node Consensus Protocol">
Distributed Node Consensus Protocol
</title>
<author initials="M" surname="Stenberg" fullname="Markus Stenberg">
<address>
<postal>
<street/>
<city>Helsinki</city>
<code>00930</code>
<country>Finland</country>
</postal>
<email>markus.stenberg@iki.fi</email>
</address>
</author>
<author initials="S" surname="Barth" fullname="Steven Barth">
<address>
<postal>
<street/>
<city>Halle</city>
<code>06114</code>
<country>Germany</country>
</postal>
<email>cyrus@openwrt.org</email>
</address>
</author>
<date month="March" year="2015" />
<area>Internet</area>
<workgroup>Homenet Working Group</workgroup>
<keyword>Homenet</keyword>
<abstract>
<t>This document describes the Distributed Node Consensus Protocol
(DNCP), a generic state synchronization protocol which uses Trickle
and Merkle trees. DNCP is transport agnostic and leaves some of the
details to be specified in profiles, which define actual
implementable DNCP based protocols. </t>
</abstract>
</front>
<middle>
<section title="Introduction">
<t>DNCP is designed to provide a way for nodes to publish data
consisting of an ordered set of TLV (Type-Length-Value) tuples and to
receive the data published by all other reachable DNCP nodes.</t>
<t>DNCP validates the set of data within it by ensuring that it is
reachable via nodes that are currently accounted for; therefore,
unlike Time-To-Live (TTL) based solutions, it does not require
periodic re-publishing of the data by the nodes. On the other hand,
it does require the topology to be visible to every node that wants
to be able to identify unreachable nodes and therefore remove old,
stale data. Another notable feature is the use of Trickle to send
status updates as it makes the DNCP network very thrifty when there
are no updates. DNCP is most suitable for data that changes only
gradually to gain the maximum benefit from using Trickle, and if more
rapid state exchanges are needed, something point-to-point is
recommended and just e.g. publishing of addresses of the services
within DNCP. </t>
<t>DNCP has relatively few requirements for the underlying transport;
it requires some way of transmitting either unicast datagram or
stream data to a DNCP peer and, if used in multicast mode, a way of
sending multicast datagrams. If security is desired and one of the
built-in security methods is to be used, support for some TLS-derived
transport scheme - such as <xref target="RFC5246">TLS</xref> on top of
TCP or <xref target="RFC6347">DTLS</xref> on top of UDP - is also
required. </t>
</section>
<section anchor="kwd" title='Requirements Language'>
<t>The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL
NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and
"OPTIONAL" in this document are to be interpreted as described in
<xref target='RFC2119'>RFC 2119</xref>.</t>
</section>
<section title="Terminology">
<t>A DNCP profile is a definition of a set of rules and
values listed in <xref target="profile-bits"/> specifying the
behavior of a DNCP based protocol, such as the used transport
method. For readability, any DNCP profile specific parameters
with a profile-specific fixed value are prefixed with DNCP_.</t>
<t>A DNCP node is a single node which runs a protocol
based on a DNCP profile.</t>
<t>The DNCP network is a set of DNCP nodes running the same DNCP
profile that can reach each other, either via discovered connectivity
in the underlying network, or using each other's addresses learned
via other means. As DNCP exchanges are bidirectional, DNCP nodes
connected via only unidirectional links are not considered
connected. </t>
<t>The node identifier is an opaque fixed-length identifier consisting of
DNCP_NODE_IDENTIFIER_LENGTH bytes which uniquely identifies a DNCP
node within a DNCP network. </t>
<t>A link indicates a link-layer media over which directly connected
nodes can communicate.</t>
<t>An interface indicates a port of a node that is connected to a
particular link. </t>
<t>A connection denotes a locally configured use of DNCP on a DNCP
node, that is attached either to an interface, to a specific remote
unicast address to be contacted, or to a range of remote unicast
addresses that are allowed to contact.</t>
<t>The connection identifier is a 32-bit opaque value, which identifies a
particular connection of that particular DNCP node. The value 0 is
reserved for DNCP and sub-protocol purposes in the TLVs, and MUST NOT
be used to identify an actual connection. This definition is in sync
with <xref target="RFC3493"/>, as the non-zero small positive
integers should comfortably fit within 32 bits.</t>
<t>A (DNCP) peer refers to another DNCP node with which
a DNCP node communicates directly on a particular connection.</t>
<t>The node data is a set of TLVs published by a node in the DNCP
network. The whole node data is owned by the node that publishes it, and
it MUST be passed along as-is, including TLVs unknown to the
forwarder.</t>
<t>The node state is a set of metadata attributes for node data. It
includes a sequence number for versioning, a hash value for comparing
and a timestamp indicating the time passed since its last
publication. The hash function and the number of bits used are
defined in the DNCP profile. </t>
<t>The network state (hash) is a hash value which represents
the current state of the network. The hash function and the number of
bits used are defined in the DNCP profile.
Whenever any node is added, removed or changes its published node
data this hash value changes as well. It is calculated over the
hash values of each reachable nodes' node data in ascending order
of the respective node identifier.</t>
<t>The effective (trust) verdict for a certificate is defined as the
verdict with the highest priority within the set of verdicts
announced for the certificate in the DNCP network.</t>
<t>The neighbor graph is the undirected graph of DNCP nodes produced
by retaining only bidirectional peer relationships between nodes.</t>
</section>
<section anchor="dm" title="Data Model">
<t>A DNCP node has:
<list style="symbols">
<t>A timestamp indicating the most recent neighbor graph traversal
described in <xref target="purge" />.</t>
</list>
</t>
<t>A DNCP node has for every DNCP node in the DNCP network:
<list style="symbols">
<t>A node identifier, which uniquely identifies the node.</t>
<t>The node data, an ordered set of TLV tuples published by that
particular node. This set of TLVs has a well-defined order based on
ascending binary content (including TLV type and length). This
facilitates linear time state delta processing. </t>
<t>The latest update sequence number, a 32 bit number that is
incremented any time the TLV set is published. For comparison
purposes, a looping comparison should be used to avoid problems in
case of overflow. An example would be: a < b <=> (a - b)
% 2^32 & 2^31 != 0.</t>
<t>The relative time (in milliseconds) since the current TLV data
set with the current update sequence number was published. It is
also a 32 bit number on the wire. If this number is close to
overflow (greater than 2^32-2^16), a node MUST re-publish its TLVs
even if there is no change to avoid overflow of the value. In other
words, absent any other changes, the TLV set MUST be re-published
roughly every 49 days.</t>
<t>A timestamp identifying the time it was last reachable based on
neighbor graph traversal described in <xref target="purge" />.</t>
</list>
</t>
<t>Additionally, a DNCP node has a set of connections for which DNCP
is configured to be used. For each such connection, a node has:
<list style="symbols">
<t>A connection identifier.</t>
<t>An interface, a unicast address of a DNCP peer it should
connect with, or a range of addresses from which DNCP peers are
allowed to connect.</t>
<t>A <xref target="RFC6206">Trickle</xref> instance with parameters
I, T, and c.</t>
</list>
</t>
<t>For each DNCP peer detected on a connection, a DNCP node has:
<list style="symbols">
<t>The node identifier of the DNCP peer.</t>
<t>The connection identifier of the DNCP peer.</t>
<t>The most recent address used by the DNCP peer (in an
authenticated message, if security is enabled).</t>
</list>
</t>
</section>
<section title="Operation">
<t>The DNCP protocol consists of <xref target="RFC6206">Trickle</xref>
driven unicast or multicast status messages which indicate the current
status of shared TLV data and additional unicast message exchanges
which ensure DNCP peer reachability and synchronize the data when
necessary. </t>
<t>If DNCP is to be used on a multicast-capable interface, as opposed
to only point-to-point using unicast, a datagram-based transport
which supports multicast SHOULD be defined in the DNCP profile to be
used for the messages to be sent to the whole link. As this is used
only to identify potential new DNCP nodes and to notify that an
unicast exchange should be triggered, the multicast transport does
not have to be particularly secure.</t>
<section title="Trickle-Driven Status Update Messages"
anchor="trickle-updates">
<t>Each node MUST send either a <xref
target="long-net-message">Long Network State Update message</xref>
or a <xref target="short-net-message">Short Network State Update
message</xref> every time the connection-specific <xref
target="RFC6206">Trickle algorithm</xref> instance indicates that
an update should be sent.
The destination address of the message should be multicast in case
of an interface which is multicast-capable, or the unicast address
of the remote party in case of a point-to-point connection.
By default, Long Network State Update messages SHOULD be used, but
if it is defined as undesirable for some case by the DNCP profile,
Short Network State Update message MUST be sent instead. This may
be useful to avoid fragmenting packets to multicast destinations,
or for security reasons.</t>
<t>A Trickle state MUST be maintained separately for each
connection. The Trickle state for all connections is considered
inconsistent and reset if and only if the locally calculated
network state hash changes. This occurs either due to a change in
the local node's own node data, or due to receipt of more recent
data from another node.</t>
<t>The Trickle algorithm has 3 parameters: Imin, Imax and k. Imin
and Imax represent the minimum and maximum values for I, which is
the time interval during which at least k Trickle updates must be
seen on a connection to prevent local state transmission. The
actual suggested Trickle algorithm parameters are DNCP profile
specific, as described in <xref target="profile-bits"/>.</t>
</section>
<section title="Processing of Received Messages" anchor="reception">
<t>This section describes how received messages are processed. The
DNCP profile may specify criteria based on which received messages
are ignored. Any 'reply' mentioned in the steps below denotes
sending of the specified message via unicast to the originator of
the message being processed. If the reply was caused by a multicast
message and sent to a link with shared bandwidth it SHOULD be
delayed by a random timespan in [0, Imin/2]. Sending of replies
SHOULD be rate-limited by the implementation, and in case of excess
load (or some other reason), a reply MAY be omitted altogether. </t>
<t>Upon receipt of:
<list>
<t><xref target="short-net-message">Short Network State
Update</xref>:
If the network state hash within the message differs from the
locally calculated network state hash, the receiver MUST reply
with a <xref target="req-net-message">Network State Request
message</xref>.
</t>
<t><xref target="long-net-message">Long Network State
Update</xref>:
<list style="symbols">
<t>If the network state hash within the message matches the
locally calculated network state hash, stop processing.</t>
<t>Otherwise the receiver MUST identify all nodes for which local
information is outdated (local update sequence number is lower
than that within the message), potentially incorrect (local
update sequence number matches but the hash of the node data TLV
differs) or missing.</t>
<t>If any such nodes are identified, the receiver MUST reply
with one or more <xref target="req-node-message">Node Data
Request message(s)</xref> containing <xref
target="req-node-data">Request Node Data TLV(s)</xref> for the
corresponding nodes.</t>
</list>
</t>
<t><xref target="req-net-message">Network State Request</xref>:
the receiver MUST reply with a <xref
target="long-net-message">Long Network State Update</xref>. </t>
<t><xref target="req-node-message">Node Data Request</xref>: the
receiver MUST reply with the requested data in a <xref
target="reply-node-message">Node Data Reply
message</xref>. Optionally - if specified by the DNCP profile -
multiple replies MAY be sent in order to e.g. keep size of each
datagram within the PMTU to the destination. However these
replies must be valid stand-alone Node Data Reply messages, with
the full state for the particular nodes.</t>
<t><xref target="reply-node-message">Node Data Reply</xref>: If
the message contains Node State TLVs that are more recent than
the local state (the received TLV has a higher update sequence
number, the node data TLV hash differs from the local one, or
local data is missing altogether) and if the message also
contains corresponding Node Data TLVs, the receiver MUST update
its locally stored state.</t>
</list>
</t>
<t>If a message containing <xref target="node-state">Node State
TLVs</xref> is received with the node identifier matching the local
node identifier and a higher update sequence number than its
current local value, or the same update sequence number and a
different hash, the node SHOULD re-publish its own node data with an
update sequence number 1000 higher than the received one. This may
occur normally once due to the local node restarting and not storing
the most recently used update sequence number. If this occurs more
than once, the DNCP profile should provide guidance on how to
handle these situations as it indicates the existence of another
active node with the same node identifier.</t>
</section>
<section title="Adding and Removing Peers" anchor="neighbor">
<t>When receiving a message on a connection from an unknown peer:
<list>
<t>If it is a unicast message, and the message contains a <xref
target="node-id">Node Connection TLV</xref>, the remote node MUST
be added as a peer on the connection and a <xref
target="neighbor-tlv">Neighbor TLV</xref> MUST be created for it.
</t>
<t>If it is a multicast message, and the message contains a <xref
target="node-id">Node Connection TLV</xref>, the remote node
SHOULD be sent a (possibly rate-limited) unicast <xref
target="req-net-message">Network State Request
Message</xref>.</t>
</list>
</t>
<t>If keep-alives are NOT sent by the peer (either the DNCP profile
does not specify the use of keep-alives or the particular peer chooses
not to send keep-alive messages), some other means MUST be employed
to ensure a DNCP peer is present. When the peer is no longer
present, the Neighbor TLV and the local DNCP peer state MUST be
removed.</t>
</section>
<section anchor="purge" title="Purging Unreachable Nodes">
<t>When a Neighbor TLV or a whole node is added or removed, the
neighbor graph SHOULD be traversed, starting from the local
node. The edges to be traversed are identified by looking for
Neighbor TLVs on both nodes, that have the other node's identifier
in the neighbor node identifier, and local and neighbor connection
identifiers swapped. Each node reached should be marked currently
reachable.</t>
<t>DNCP nodes MUST be either purged immediately when not marked
reachable in a particular graph traversal, or eventually after they
have not been marked reachable within DNCP_GRACE_INTERVAL. During
the grace period, the nodes that were not marked reachable in the
most recent graph traversal MUST NOT be used for calculation of the
network state hash, be provided to any applications that need to
use the whole TLV graph, or be provided to remote nodes. </t>
</section>
</section>
<section anchor="ka" title="Keep-Alive Extension">
<t>The Trickle-driven messages provide a mechanism for handling of
new peer detection (if applicable) on a connection, as well as state
change notifications. Another mechanism may be needed to get rid of
old, no longer valid DNCP peers if the transport or lower layers do
not provide one.</t>
<t>If keep-alives are not specified in the DNCP profile, the rest of
this section MUST be ignored.</t>
<t>A DNCP profile MAY specify either per-connection or per-peer
keep-alive support. </t>
<t>For every connection that a keep-alive is specified for in the
DNCP profile, the connection-specific keep-alive interval MUST be
maintained. By default, it is DNCP_KEEPALIVE_INTERVAL. If there is a
local value that is preferred for that for any reason (configuration,
energy conservation, media type, ..), it should be substituted
instead. If a non-default keep-alive interval is used on any
connection, a DNCP node MUST publish appropriate <xref
target="ka-interval">Keep-Alive Interval TLV(s)</xref>.</t>
<section title="Data Model Additions">
<t>The following additions to the <xref target="dm">Data
Model</xref> are needed to support keep-alive:</t>
<t>Each node MUST have a timestamp which indicates the last time a
<xref target="net-state">Network State TLV</xref> was sent for each
connection, i.e. on an interface or to the point-to-point
peer(s).</t>
<t>Each node MUST have for each peer:
<list style="symbols">
<t>Last consistent state timestamp: a timestamp which indicates
the last time a consistent <xref target="net-state">Network State
TLV</xref> was received from the peer. When adding a new peer, it
should be initialized to the current time.</t>
</list>
</t>
</section>
<section title="Per-Connection Periodic Keep-Alive Messages">
<t>If per-connection keep-alives are enabled on connection with a
multicast-enabled link, and if no traffic containing a <xref
target="net-state">Network State TLV</xref> has been sent to a
particular connection within the connection-specific keep-alive
interval, a <xref target="long-net-message">Long Network State
Update message</xref> or a <xref target="short-net-message">Short
Network State Update message</xref> MUST be sent on that
connection. The type of message should be chosen based on the
considerations in <xref target="trickle-updates" />. The actual
sending time SHOULD be further delayed by a random timespan in [0,
Imin/2]. When such a message is sent, a new Trickle transmission
time 't' in [I/2, I] MUST be randomly chosen.</t>
</section>
<section title="Per-Peer Periodic Keep-Alive Messages">
<t>If per-peer keep-alives are enabled on a unicast-only
connection, and if no traffic containing a <xref
target="net-state">Network State TLV</xref> has been sent to a
particular peer within the connection-specific keep-alive interval,
a <xref target="long-net-message">Long Network State Update
message</xref> or a <xref target="short-net-message">Short Network
State Update message</xref> MUST be sent to the peer. The type of
message should be chosen based on the considerations in <xref
target="trickle-updates" />. When such a message is sent, a new
Trickle transmission time 't' in [I/2, I] MUST be randomly
chosen.</t>
</section>
<section title="Received Message Processing Additions">
<t>If a message is received via unicast from the peer, the Last
consistent state timestamp for the peer MUST be updated.</t>
<t>If the received multicast message contains a <xref
target="net-state">Network State TLV</xref> which is consistent
with the locally calculated network state hash, the Last consistent
state timestamp for the peer MUST be updated. If the TLV is
inconsistent, and would not cause any unicast exchange, <xref
target="req-net-message">Network State Request</xref> SHOULD be
sent via unicast.</t>
</section>
<section title="Neighbor Removal">
<t>For every peer on every connection, the connection-specific
keep-alive interval must be calculated by looking for <xref
target="ka-interval">Keep-Alive Interval TLVs</xref> published by
the node, and if none exist, using the default value of
DNCP_KEEPALIVE_INTERVAL. If the peer's last consistent state
timestamp has not been updated for at least
DNCP_KEEPALIVE_MULTIPLIER times the peer's connection-specific
keep-alive interval, the Neighbor TLV for that peer and the local
DNCP peer state MUST be removed.</t>
</section>
</section>
<section title="Support For Dense Broadcast Links">
<t>The DNCP profile or a user configuration should limit the number of
neighbors that are allowed for a (particular type of) link that a
connection runs on. If that limit is exceeded, nodes without the
highest Node Identifier on the link SHOULD treat the connection as an
unicast connection connected to the node that has the highest Node
Identifier detected on the link. The nodes MUST also keep listening
to multicast traffic to both detect the presence of that node, and to
react to nodes with a higher Node Identifier appearing. If the
highest Node Identifier present on the link changes, the connection
endpoint MUST be changed. If the Node Identifier of the local node is
the highest one, the node MUST keep the connection in normal
multicast mode, and the node MUST allow others to peer with it over
the link.</t>
</section>
<section title="Protocol Messages">
<t>For point-to-point exchanges, DNCP can run across
datagram-based or reliable ordered stream-based transports.
If a stream-based transport is used, a 32-bit length-value
in network byte order is sent before each message to indicate the
number of bytes the following message consists of.</t>
<t>DNCP messages are encoded as a concatenated sequence of <xref
target="tlvs">Type-Length-Value objects</xref>. In order to
facilitate fast comparing of local state with that in a received
message update, all TLVs in every encoding scope (either within the
message itself, or within a container TLV) MUST be placed in
ascending order based on the binary comparison of both TLV header
and value. By design, the TLVs which MUST be present have the
lowest available type values, ensuring they will naturally occur at
the start of the Protocol Message, resembling a fixed format
header.</t>
<t>DNCP profiles MAY add additional TLVs to the message specified
here, or even define additional messages as needed. TLVs not
recognized by the receiver MUST be ignored.</t>
<section title="Short Network State Update Message"
anchor="short-net-message">
<t>The Short Network State Update Message is used to announce the
sender's view of the network state using multicast.</t>
<t>The following TLVs MUST be present:
<list style="symbols">
<t>One <xref target="node-id">Node Connection TLV</xref> identifying
the originating node and connection.</t>
<t>One <xref target="net-state">Network State TLV</xref>
containing the network state hash as calculated by the
sender.</t>
</list>
</t>
<t>The Short Network Status update message MUST NOT contain any
<xref target="node-state">Node State TLV(s)</xref>.</t>
</section>
<section title="Long Network State Update Message"
anchor="long-net-message">
<t>The Long Network State Update Message is used to announce the
sender's view of the network state and all node states using
multicast or unicast.</t>
<t>The following TLVs MUST be present:
<list style="symbols">
<t>One <xref target="node-id">Node Connection TLV</xref> identifying
the originating node and connection.</t>
<t>One <xref target="net-state">Network State TLV</xref>
containing the network state hash as calculated by the
sender.</t>
<t>One or more <xref target="node-state">Node State TLVs</xref>
containing the node state of DNCP nodes as currently known to the
sender.</t>
</list>
</t>
<t>The Long Network State Update message MUST include the
corresponding <xref target="node-state">Node State TLV</xref> for
each Node Data TLV used to calculate the network state hash.</t>
</section>
<section title="Network State Request Message"
anchor="req-net-message">
<t>The Network State Request message is used to request the
recipient's view of the network state and all node states currently
known to it.</t>
<t>The following TLVs MUST be present:
<list style="symbols">
<t>One <xref target="req-net-state">Request Network State
TLV</xref> indicating the type of request.</t>
</list>
</t>
</section>
<section title="Node Data Request Message"
anchor="req-node-message">
<t>The Node Data Request message is used to request the node state
and data of one or more DNCP nodes in the network.</t>
<t>The following TLVs MUST be present:
<list style="symbols">
<t>One or more <xref target="req-node-data">Request Node Data
TLVs</xref> indicating the nodes for which state and data is
requested.</t>
</list>
</t>
</section>
<section title="Node Data Reply Message"
anchor="reply-node-message">
<t>The Node Data Request message is used to provide
the node data of one or more DNCP nodes in the network.</t>
<t>The following TLVs MUST be present:
<list style="symbols">
<t>One <xref target="node-id">Node Connection TLV</xref> identifying
the originating node and connection.</t>
<t>One or more <xref target="node-state">Node State TLV</xref>
and <xref target="node-data">Node Data TLV</xref> pairs with
matching node identifiers for each node previously requested in a
<xref target="req-node-message">Node Data Request
message</xref>.</t>
</list>
</t>
</section>
</section>
<section anchor="tlvs" title="Type-Length-Value Objects">
<t>
Each TLV is encoded as a 2 byte type field, followed by a 2 byte
length field (of the value, excluding header; 0 means no value)
followed by the value itself (if any). Both type and length fields
in the header as well as all integer fields inside the value
- unless explicitly stated otherwise - are
represented in network byte order. Zero padding bytes MUST be added
up to the next 4 byte boundary if the length is not divisible by 4.
These padding bytes MUST NOT be included in the length field.
</t>
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Value |
| (variable # of bytes) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
</artwork>
</figure>
<t>
For example, type=123 (0x7b) TLV with value 'x' (120 =
0x78) is encoded as: 007B 0001 7800 0000.
</t>
<t>Notation:
<list>
<t>.. = octet string concatenation operation.</t>
<t>H(x) = non-cryptographic hash function specified by DNCP
profile. </t>
</list>
</t>
<section title="Request TLVs">
<section anchor="req-net-state" title="Request Network State TLV">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: REQ-NETWORK-STATE (2) | Length: 0 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
</artwork>
</figure>
<t>This TLV is used to identify a <xref
target="req-net-message">Network State Request
message</xref>.</t>
</section>
<section anchor="req-node-data" title="Request Node Data TLV">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: REQ-NODE-DATA (3) | Length: >0 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Node Identifier |
| (length fixed in DNCP profile) |
...
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
</artwork>
</figure>
<t>This TLV is used within a <xref target="req-node-message">Node
Data Request message</xref> to request node state and node data
for the node with matching node identifier, if any, to be
included in a subsequent <xref target="reply-node-message">Node
Data Reply message</xref>.</t>
</section>
</section>
<section title="Data TLVs">
<section anchor="node-id" title="Node Connection TLV">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: NODE-CONNECTION (1) | Length: > 4 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Node Identifier |
| (length fixed in DNCP profile) |
...
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Connection Identifier |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
</artwork>
</figure>
<t>This TLV identifies both the local node's node identifier, as
well as the particular connection identifier. It MUST be sent in
all messages if bidirectional peer relationship is desired with
remote nodes. Bidirectional peer relationship is not necessary
for read-only access to the DNCP state, but it is required to be
able to publish something.</t>
</section>
<section anchor="net-state" title="Network State TLV">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: NETWORK-STATE (10) | Length: > 0 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| H(H(node data TLV 1) .. [...] .. H(node data TLV N)) |
| (length fixed in DNCP profile) |
...
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
</artwork>
</figure>
<t>This TLV contains the current locally calculated network state
hash. The network state hash is derived by calculating the hash
value for each currently reachable node's Node Data TLV,
concatenating said hash values based on the ascending order of
their corresponding node identifiers, and hashing the resulting
concatenated hash values.</t>
</section>
<section anchor="node-state" title="Node State TLV">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: NODE-STATE (11) | Length: > 8 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Node Identifier |
| (length fixed in DNCP profile) |
...
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Update Sequence Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Milliseconds since Origination |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| H(node data TLV) |
| (length fixed in DNCP profile) |
...
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
</artwork>
</figure>
<t>This TLV represents the local node's knowledge about the published
state of a node in the DNCP network identified by the node identifier
field in the TLV.</t>
<t>The whole network should have roughly same idea about the time
since origination of any particular published state. Therefore
every node, including the originating one, MUST increment the
time whenever it needs to send a Node State TLV for an already
published Node Data TLV. This age value is not included within
the Node Data TLV, however, as that is immutable and used to
detect changes in the network state.</t>
</section>
<section anchor="node-data" title="Node Data TLV">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: NODE-DATA (12) | Length: > 4 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| node identifier |
| (length fixed in DNCP profile) |
...
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Update Sequence Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Nested TLVs containing node information |
</artwork>
</figure>
</section>
<section anchor="neighbor-tlv"
title="Neighbor TLV (within Node Data TLV)">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: NEIGHBOR (13) | Length: > 8 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| neighbor node identifier |
| (length fixed in DNCP profile) |
...
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Neighbor Connection Identifier |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Local Connection Identifier |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
</artwork>
</figure>
<t>This TLV indicates that the node in question vouches that the
specified neighbor is reachable by it on the specified local
connection.
The presence of this TLV at least guarantees that the node
publishing it has received traffic from the neighbor
recently. For guaranteed up-to-date bidirectional reachability,
the existence of both nodes' matching Neighbor TLVs should be
checked. </t>
</section>
<section anchor="ka-interval"
title="Keep-Alive Interval TLV (within Node Data TLV)">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: KEEP-ALIVE-INTERVAL (14)| Length: 8 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Connection Identifier |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Interval |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
</artwork>
</figure>
<t>This TLV indicates a non-default interval being used to send
keep-alive messages specified in <xref target="ka" />.</t>
<t>Connection identifier is used to identify the particular
connection for which the interval applies. If 0, it applies for
ALL connections for which no specific TLV exists.</t>
<t>Interval specifies the interval in milliseconds at which the
node sends keep-alives. A value of zero means no keep-alives are
sent at all; in that case, some lower layer mechanism that
ensures presence of nodes MUST be available and used. </t>
</section>
</section>
<section anchor="user-tlv"
title="Custom TLV (within/without Node Data TLV)">
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: CUSTOM-DATA (15) | Length: > 0 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| H(URI) |
| (length fixed in DNCP profile) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Opaque Data |
</artwork>
</figure>
<t>This TLV can be used to contain anything; the URI used should be
under control of the author of that specification.
For example:</t>
<t>V = H('http://example.com/author/json-for-dncp') .. '{"cool":
"json extension!"}'</t>
<t>or</t>
<t>V = H('mailto:author@example.com') .. '{"cool": "json
extension!"}'</t>
</section>
</section>
<section title="Security and Trust Management">
<t>If specified in the DNCP profile, either <xref
target="RFC6347">DTLS</xref> or <xref target="RFC5246">TLS</xref> may
be used to authenticate and encrypt either some (if specified
optional in the profile), or all unicast traffic. The following
methods for establishing trust are defined, but it is up to the DNCP
profile to specify which ones may, should or must be supported.</t>
<section title="Pre-Shared Key Based Trust Method">
<t>A PSK-based trust model is a simple security management
mechanism that allows an administrator to deploy devices to an
existing network by configuring them with a pre-defined key,
similar to the configuration of an administrator password or
WPA-key. Although limited in nature it is useful to provide a
user-friendly security mechanism for smaller networks. </t>
</section>
<section title="PKI Based Trust Method">
<t>A PKI-based trust-model enables more advanced management
capabilities at the cost of increased complexity and
bootstrapping effort. It however allows trust to be managed in a
centralized manner and is therefore useful for larger networks
with a need for an authoritative trust management.</t>
</section>
<section title="Certificate Based Trust Consensus Method">
<t>The certificate-based consensus model is designed to be a
compromise between trust management effort and flexibility. It is
based on X.509-certificates and allows each DNCP node to provide a
verdict on any other certificate and a consensus is found to
determine whether a node using this certificate or any
certificate signed by it is to be trusted. </t>
<t>The current effective trust verdict for any certificate is
defined as the one with the highest priority from all verdicts
announced for said certificate at the time.</t>
<section title="Trust Verdicts">
<t>Trust Verdicts are statements of DNCP nodes about the
trustworthiness of X.509-certificates. There are 5 possible
verdicts in order of ascending priority:
<list style="hanging">
<t hangText="0 Neutral">: no verdict exists but the DNCP
network should determine one.</t>
<t hangText="1 Cached Trust">: the last known effective verdict
was Configured or Cached Trust.</t>
<t hangText="2 Cached Distrust">: the last known effective
verdict was Configured or Cached Distrust.</t>
<t hangText="3 Configured Trust">: trustworthy based upon an
external ceremony or configuration.</t>
<t hangText="4 Configured Distrust">: not trustworthy based upon
an external ceremony or configuration.</t>
</list>
</t>
<t>
Verdicts are differentiated in 3 groups:
<list style="symbols">
<t>Configured verdicts are used to announce explicit verdicts
a node has based on any external trust bootstrap or
predefined relation a node has formed with a given
certificate.</t>
<t>Cached verdicts are used to retain the last known trust
state in case all nodes with configured verdicts about a
given certificate have been disconnected or turned off.</t>
<t>The Neutral verdict is used to announce a new node
intending to join the network so a final verdict for it can
be found.</t>
</list>
</t>
<t>
The current effective trust verdict for any certificate is
defined as the one with the highest priority within the set of
verdicts + announced for the certificate in the DNCP network.
A node MUST be trusted for participating in the DNCP network if
and only if the current effective verdict for its own
certificate or any one in its certificate hierarchy is (Cached
or Configured) Trust and none of the certificates in its
hierarchy have an effective verdict of (Cached or Configured)
Distrust.
In case a node has a configured verdict, which is different
from the current effective verdict for a certificate, the
current effective verdict takes precedence in deciding
trustworthiness. Despite that, the node still retains and
announces its configured verdict.
</t>
</section>
<section title="Trust Cache">
<t>Each node SHOULD maintain a trust cache containing the current
effective trust verdicts for all certificates currently announced
in the DNCP network. This cache is used as a backup of the last
known state in case there is no node announcing a configured
verdict for a known certificate. It SHOULD be saved to a
non-volatile memory at reasonable time intervals to survive a
reboot or power outage.</t>
<t>Every time a node (re)joins the network or detects the change
of an effective trust verdict for any certificate, it will
synchronize its cache, i.e. store new effective verdicts
overwriting any previously cached verdicts. Configured verdicts
are stored in the cache as their respective cached counterparts.
Neutral verdicts are never stored and do not override existing
cached verdicts.</t>
</section>
<section title="Announcement of Verdicts">
<t>A node SHOULD always announce any configured trust verdicts it
has established by itself, and it MUST do so if announcing the
configured trust verdict leads to a change in the current
effective verdict for the respective certificate. In absence of
configured verdicts, it MUST announce cached trust verdicts it has
stored in its trust cache, if one of the following conditions
applies:
<list style="symbols">
<t>The stored verdict is Cached Trust and the current effective
verdict for the certificate is Neutral or does not exist.</t>
<t>The stored verdict is Cached Distrust and the current
effective verdict for the certificate is Cached Trust.</t>
</list>
A node rechecks these conditions whenever it detects changes of
announced trust verdicts anywhere in the network.
</t>
<t>Upon encountering a node with a hierarchy of certificates for
which there is no effective verdict, a node adds a Neutral
Trust-Verdict-TLV to its node data for all certificates found in
the hierarchy, and publishes it until an effective verdict
different from Neutral can be found for any of the certificates,
or a reasonable amount of time (10 minutes is suggested) with no
reaction and no further authentication attempts has passed. Such
verdicts SHOULD also be limited in rate and number to prevent
denial-of-service attacks.</t>
<t>Trust verdicts are announced using Trust-Verdict TLVs:
<figure>
<artwork>
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type: Trust-Verdict (16) | Length: 37-100 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Verdict | (reserved) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| |
| |
| |
| SHA-256 Fingerprint |
| |
| |
| |
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Common Name |
</artwork>
</figure>
<list>
<t>Verdict represents the numerical index of the verdict.</t>
<t>(reserved) is reserved for future additions and MUST be set
to 0 when creating TLVs and ignored when parsing them.</t>
<t>SHA-256 Fingerprint contains the <xref
target="RFC6234">SHA-256</xref> hash value of the certificate
in DER-format.</t>
<t>Common Name contains the variable-length (1-64 bytes) common
name of the certificate. Final byte MUST have value of 0.</t>
</list>
</t>
</section>
<section title="Bootstrap Ceremonies">
<t>The following non-exhaustive list of methods describes
possible ways to establish trust relationships between
DNCP nodes and node certificates. Trust establishment is a
two-way process in which the existing network must trust the
newly added node and the newly added node must trust at least
one of its neighboring nodes.
It is therefore necessary that both the newly added node and an
already trusted node perform such a ceremony to successfully
introduce a node into the DNCP network. In all cases an
administrator MUST be provided with external means to identify
the node belonging to a certificate based on its fingerprint
and a meaningful common name.</t>
<section title="Trust by Identification">
<t>A node implementing certificate-based trust MUST provide
an interface to retrieve the current set of effective trust
verdicts, fingerprints and names of all certificates currently
known and set configured trust verdicts to be
announced. Alternatively it MAY provide a companion DNCP node
or application with these capabilities with which it has a
pre-established trust relationship.</t>
</section>
<section title="Preconfigured Trust">
<t>A node MAY be preconfigured to trust a certain set of
node or CA certificates. However such trust relationships
MUST NOT result in unwanted or unrelated trust for nodes not
intended to be run inside the same network (e.g. all other
devices by the same manufacturer).</t>
</section>
<section title="Trust on Button Press">
<t>A node MAY provide a physical or virtual interface to put
one or more of its internal network interfaces temporarily into
a mode in which it trusts the certificate of the first
DNCP node it can successfully establish a connection
with.</t>
</section>
<section title="Trust on First Use">
<t>A node which is not associated with any other DNCP node MAY
trust the certificate of the first DNCP node it can
successfully establish a connection with. This method MUST NOT
be used when the node has already associated with any other
DNCP node.</t>
</section>
</section>
</section>
</section>
<section anchor="profile-bits" title="DNCP Profile-Specific Definitions">
<!-- TBD - Pierre has a point - should define guidance on these,
perhaps -->
<t>Each DNCP profile MUST define following:
<list style="symbols">
<t>How the messages are secured: Not at all, optionally or always
with the TLS scheme defined here using one or more of the methods,
or with something else. If the links with DNCP nodes can be
sufficiently secured or isolated, it is possible to run DNCP in a
secure manner without using any form of authentication or
encryption.</t>
<t>Unicast and optionally multicast transport protocol(s) to be
used. If TLS scheme within this document is to be used security,
TLS or DTLS support for at least the unicast transport protocol is
mandatory.</t>
<t>Transport protocols' parameters such as port numbers to be used,
or multicast address to be used. Unicast, multicast, and secure
unicast may each require different parameters, if applicable. </t>
<t>When receiving messages, what sort of messages are dropped, as
specified in <xref target="reception" />.</t>
<t>What is the criteria for sending Trickle-based <xref
target="long-net-message">Long Network State Update message</xref>
on an interface or to a DNCP peer.</t>
<t>How to deal with node identifier collision as described in <xref
target="reception" />. Main options are either for one or both
nodes to assign new node identifiers to themselves, or to notify
someone about a fatal error condition in the DNCP network.</t>
<t>Imin, Imax and k ranges to be suggested for implementations to
be used in the Trickle algorithm. The Trickle algorithm does not
require these to be same across all implementations for it to work,
but similar orders of magnitude helps implementations of a DNCP
profile to behave more consistently and to facilitate estimation of
lower and upper bounds for behavior of the network.</t>
<t>Hash function H(x) to be used, and how many bits of the input
are actually used. The chosen hash function is used to handle both
hashing of node specific data, and network state hash, which is a
hash of node specific data hashes. SHA-256 defined in <xref
target="RFC6234" /> is the recommended default choice.</t>
<t>DNCP_NODE_IDENTIFIER_LENGTH: The fixed length of a node
identifier (in bytes).</t>
<t>DNCP_GRACE_INTERVAL: How long node data for unreachable nodes is
kept.</t>
<t>Whether to send keep-alives, and if so, on an interface, using
multicast, or directly using unicast to peers. Keep-alive has also
associated parameters:
<list style="symbols">
<t>DNCP_KEEPALIVE_INTERVAL: How often keep-alive messages are to be
sent by default (if enabled).</t>
<t>DNCP_KEEPALIVE_MULTIPLIER: How many times the
DNCP_KEEPALIVE_INTERVAL (or peer-supplied keep-alive interval
value) a node may not be heard from to be considered still
valid.</t>
</list>
</t>
</list>
</t>
</section>
<section title="Security Considerations">
<t>DNCP profiles may use multicast to indicate DNCP state changes and
for keep-alive purposes. However, no actual data TLVs will be sent
across that channel. Therefore an attacker may only learn hash values
of the state within DNCP and may be able to trigger unicast
synchronization attempts between nodes on a local link this way. A DNCP
node should therefore rate-limit its reactions to multicast
packets.</t>
<t>When using DNCP to bootstrap a network, PKI based solutions may have
issues when validating certificates due to potentially unavailable
accurate time, or due to inability to use the network to either check
Certifcate Revocation Lists or perform on-line validation.</t>
<t>The Certificate-based trust consensus mechanism defined in this
document allows for a consenting revocation, however in case of a
compromised device the trust cache may be poisoned before the actual
revocation happens allowing the distrusted device to rejoin the network
using a different identity. Stopping such an attack might require
physical intervention and flushing of the trust caches. </t>
</section>
<section anchor="iana" title="IANA Considerations">
<t>IANA should set up a registry for DNCP TLV types,
with the following initial contents:</t>
<t>0: Reserved (should not happen on wire)</t>
<t>1: Node connection</t>
<t>2: Request network state</t>
<t>3: Request node data</t>
<t>4-9: Reserved for DNCP profile use</t>
<t>10: Network state</t>
<t>11: Node state</t>
<t>12: Node data</t>
<t>13: Neighbor</t>
<t>14: Keep-alive interval</t>
<t>15: Custom</t>
<t>16: Trust-Verdict</t>
<t>17-31: Reserved for future DNCP versions.</t>
<t>192-255: Reserved for per-implementation experimentation. The
nodes using TLV types in this range SHOULD use e.g. Custom TLV to
identify each other and therefore eliminate potential conflict caused
by potential different use of same TLV numbers. </t>
<t>For the rest of the values (32-191, 256-65535), policy of 'standards
action' should be used.</t>
</section>
</middle>
<back>
<references title="Normative references">
<?rfc include="reference.RFC.2119.xml"?>
<?rfc include="reference.RFC.6206.xml"?>
<?rfc include="reference.RFC.6347.xml"?>
<?rfc include="reference.RFC.5246.xml"?>
</references>
<references title="Informative references">
<?rfc include="reference.RFC.3493.xml"?>
<?rfc include="reference.RFC.6234.xml"?>
</references>
<section title="Some Outstanding Issues">
<t>Should better forms of combined messages be defined? e.g. allow
sending both request-network-state, and currently set of known local
state at same time in one message.</t>
<t>Should some sort of fragmentation scheme be defined for the data?
Currently, DNCP uses Merkle tree of depth 2 (node data -> node hash ->
network hash). However, it essentially treats all TLVs published by
a single node as a single chunk on the protocol level. Is that a
scalability problem? Adding one more level to the tree might address
this.</t>
</section>
<section title="Some Obvious Questions and Answers">
<t>Q: Should there be nested container syntax that is actually
self-describing? (i.e. type flag that indicates container, no body
except sub-TLVs?)</t>
<t>A: Not for now, but perhaps valid design.. TBD.</t>
<t>Q: Add third case for multicast - 'medium' network state, which is
'long' one, but partial?</t>
<t>A: Drops typical convergence on large networks 5->3 packets, at
expense of some specification/implementation complexity. However, as
anything else than short network state leaks information via multicast,
it does not seem worth it as secure protocols probably want to prevent
multicast sending of anything else than short network state in any
case. Additionally, the long network state being complete set of nodes
actually facilitates light-weight nodes that do not want to do
graph-based pruning. So all in all, perhaps not worth it.</t>
<t>Q: 32-bit connection id?</t>
<t>A: Here, it would save 32 bits per neighbor if it was 16 bits (and
less is not realistic). However, TLVs defined elsewhere would not
seem to even gain that much on average. 32 bits is also used for
ifindex in various operating systems, making for simpler
implementation.</t>
<t>Q: Why have topology information at all?</t>
<t>A: It is an alternative to the more traditional seq#/TTL-based flooding
schemes. In steady state, there is no need to e.g. re-publish every now
and then.</t>
</section>
<section title="Changelog">
<t>draft-ietf-homenet-dncp-01:
<list style="symbols">
<t>Fixed keep-alive semantics to consider unicast requests also
updates of most recently consistent, and added proactive unicast
request to ensure even inconsistent keep-alive messages eventually
triggering consistency timestamp update.</t>
<t>Facilitated (simple) read-only clients by making Node Connection
TLV optional if just using DNCP for read-only purposes.</t>
<t>Added text describing how to deal with "dense" networks, but left
actual numbers and mechanics up to DNCP profiles and (local)
configurations.</t>
</list>
</t>
<t>draft-ietf-homenet-dncp-00: Split from pre-version of
draft-ietf-homenet-hncp-03 generic parts. Changes that affect
implementations:
<list style="symbols">
<t>TLVs were renumbered.</t>
<t>TLV length does not include header (=-4). This facilitates
e.g. use of DHCPv6 option parsing libraries (same encoding), and
reduces complexity (no need to handle error values of length less
than 4).</t>
<t>Trickle is reset only when locally calculated network state hash
is changes, not as remote different network state hash is seen. This
prevents e.g. attacks by multicast with one multicast packet to force
Trickle reset on every interface of every node on a link.</t>
<t>Instead of 'ping', use 'keep-alive' (optional) for dead peer
detection. Different message used!</t>
</list>
</t>
</section>
<section title="Draft Source">
<t>As usual, this draft is available at <eref
target="https://github.com/fingon/ietf-drafts/">
https://github.com/fingon/ietf-drafts/</eref>
in source format (with nice Makefile too). Feel free to send comments
and/or pull requests if and when you have changes to it! </t>
</section>
<section title="Acknowledgements">
<t>Thanks to Ole Troan, Pierre Pfister, Mark Baugher, Mark Townsley,
Juliusz Chroboczek and Jiazi Yi for their contributions to the
draft.</t>
</section>
</back>
</rfc>
| PAFTECH AB 2003-2026 | 2026-04-23 04:59:49 |