One document matched: draft-hares-i2rs-bgp-im-00.xml
<?xml version="1.0" encoding="US-ASCII"?>
<!DOCTYPE rfc SYSTEM "rfc2629.dtd" [
<!ENTITY RFC2119 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2119.xml">
<!ENTITY RFC3060 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.3060.xml">
<!ENTITY RFC3644 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.3644.xml">
<!ENTITY RFC5394 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.5394.xml">
<!ENTITY RFC5511 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.5511.xml">
<!ENTITY RFC6536 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.6536.xml">
<!ENTITY I-D.ietf-i2rs-architecture SYSTEM "http://xml.resource.org/public/rfc/bibxml3/reference.I-D.ietf-i2rs-architecture.xml">
<!ENTITY I-D.ietf-i2rs-rib-info-model SYSTEM "http://xml.resource.org/public/rfc/bibxml3/reference.I-D.ietf-i2rs-rib-info-model.xml">
<!ENTITY I-D.white-i2rs-use-case SYSTEM "http://xml.resource.org/public/rfc/bibxml3/reference.I-D.white-i2rs-use-case.xml">
<!ENTITY I-D.hares-i2rs-info-model-policy SYSTEM "http://xml.resource.org/public/rfc/bibxml3/reference.I-D.hares-i2rs-info-model-policy.xml">
<!ENTITY I-D.keyupate-i2rs-bgp-usecases SYSTEM "http://xml.resource.org/public/rfc/bibxml3/reference.I-D.keyupate-i2rs-bgp-usecases">
<!ENTITY I-D.ietf-netconf-restconf SYSTEM "http://xml.resource.org/public/rfc/bibxml3/reference.I-D.ietf-netconf-restconf.xml">
]>
<?xml-stylesheet type='text/xsl' href='rfc2629.xslt' ?>
<?rfc toc="yes" ?>
<?rfc symrefs="yes" ?>
<?rfc sortrefs="yes"?>
<?rfc compact="yes" ?>
<?rfc subcompact="no" ?>
<?rfc iprnotified="no" ?>
<?rfc strict="no" ?>
<rfc category="std" docName="draft-hares-i2rs-bgp-im-00.txt" ipr="trust200902">
<front>
<title abbrev="BGP IM">An I2RS BGP Information Model</title>
<author fullname="Susan Hares" initials="S" surname="Hares">
<organization>Huawei</organization>
<address>
<postal>
<street>7453 Hickory Hill</street>
<city>Saline</city>
<region>MI</region>
<code>48176</code>
<country>USA</country>
</postal>
<email>shares@ndzh.com</email>
</address>
</author>
<author fullname="Lixing Wang" initials="L." surname="Wang">
<organization>Huawei</organization>
<address>
<postal>
<street/>
<city>Beijing</city>
<region/>
<country>PRC</country>
</postal>
<email>wanglixing@huawei.com</email>
</address>
</author>
<author fullname="Shunwan Zhuang" initials="S." surname="Zhuang">
<organization>Huawei</organization>
<address>
<postal>
<street/>
<city>Beijing</city>
<region/>
<country>PRC</country>
</postal>
<email>Zhuangshunwan@huawei.com</email>
</address>
</author>
<date year="2014"/>
<area>Routing</area>
<workgroup>I2RS Working Group</workgroup>
<keyword>I2RS</keyword>
<keyword>Information Model</keyword>
<keyword>IM</keyword>
<abstract>
<t>This document introduces an information model for i2RS BGP protocol
and network policy that aligns with the I2RS BGP use cases. This draft
utilizes the general Policy based routing structured found in the RIB
Information Model (IM) and Policy Base Policy.</t>
</abstract>
</front>
<middle>
<section anchor="intro" title="Introduction">
<t>The Interface to the Routing System (I2RS) provides read and write
access to the information and state within the routing process within
routing elements via protocol message exchange between an I2RS Client
and an I2RS Agent associated with the a routing system. The <xref
target="I-D.ietf-i2rs-architecture"/> describes the basic interactions
of this exchange. One of the important functions of this messages
exchange is for the I2RS client to interact with one or more I2RS agents
to collect information from network routing systems.</t>
<t>One of the key information pieces that I2RS clients may collect from
I2RS agents is the BGP information regarding BGP peers, BGP routes, and
network topologies created within BGP. This document introduces a basic
information model for BGP policies. This policy model can be linked with
other information models such as the I2RS RIB informational model <xref
target="I-D.ietf-i2rs-rib-info-model"/>, and uses the generic policy
found in the Policy Routing modules in <xref
target="I-D.hares-i2rs-info-model-policy"> </xref>. This basic BGP
Information Model is an extendable policy model is a product of the
industry approach to I2RS.</t>
</section>
<section title="Definitions and Acronyms">
<t><list>
<t>BGP: Border Gateway Protocol</t>
<t>Information Model: An abstract model of a conceptual domain,
independent of a specific implementation or data representation</t>
<t>NETCONF:The Network Configuration Protocol as defined in <xref
target="RFC6536"> </xref></t>
<t>RESTCONF: The REST-like protocol that provides a programmatic
interface over HTTP for accesing data defined in YANG, using
datastores defined in NETCONF Protocol <xref
target="I-D.ietf-netconf-restconf"> </xref> as defined in RBNF:
Routing Backus-Naur Form <xref target="RFC5511"> </xref>.</t>
</list></t>
</section>
<section title="BGP Data">
<t>BGP data includes information related to the BGP protocol instances,
BGP Peer information, BGP Route information, and BGP policy. This
section describes the information associated with the I2RS agent
accessible via BGP.</t>
<section title="BGP Protocol">
<t>BGP protocol support within a router(or virtual routing device) may
have multiple instances of the BGP protocol. This section describes
the information regarding the BGP protocol available to the I2RS
Agent. At end of this section there is a RBNF form summarizing this
topology.</t>
<t><list style="hanging">
<t hangText="Protocol"><vspace blankLines="1"/> The BGP protocol
information stores a group of information associated with a BGP
protocol implementation. This group of informational includes:
<list style="symbols">
<t>Router-ID (R/W) - Router-ID for BGP protocol,</t>
<t>AS(R/W)- Autonomous System Number,</t>
<t>protocol_status (R) - administrative status of the
protocol,</t>
<t>shutdown_protocol (R/W) - Shutdown the BGP protocol,</t>
<t>ROLES (R) - a composite set of roles BGP peers play,</t>
<t>bgp-instance-list (R/W) - BGP protocol instances,</t>
</list></t>
</list> The section below describes these elements.</t>
<section title="Elements in I2RS Protocol Block">
<t>This section describes each element of the I2RS Protocol Block.
<list style="hanging">
<t hangText="Router-ID (R) "><vspace blankLines="1"/> This I2RS
BGP IM element stores the ROUTER-ID which is configured in the
BGP protocol as the default router-id. Instances of a protocol
may configure unique router-IDs. The router-ID is a 32 bit
unsigned value imported from the running BGP process.</t>
<t hangText="Router-ID (W) "><vspace blankLines="1"/> Upon a
write from the I2RS Client to this I2RS BGP IM element, this
I2RS Agent will temporarily change this default in the BGP
protocol. I2RS Agent-routing system software must be able to
handle both the original value (from node configuration
systems), and the I2RS value as valid defaults. This change in
the ROUTER_-ID default is passed to the BGP instances spawned
from the BGP protocol. After the I2RS Client stops the
relationship with the I2RS Agent, the Router-ID value reverts to
the configured value.</t>
<t hangText="AS(R/W)"><vspace blankLines="1"/> This I2RS BGP IM
element stores the default AS number configured in the BGP
implementation. This value is 8 byte value represented as a
decimal value. This value is a read-write value. The initial
value of the AS is imported by the I2RS Agent from the BGP
protocol in the BGP routing system. This the normal case. If
this data element is written to, I2RS stores the default AS
number value in the BGP protocol as the new default AS in the
BGP protocol. This does not change the default AS number in
existing BGP instances with the existing peers. New BGP
instances spawned after the I2RS changes the default AS value,
should have the I2RS specified default AS. Upon the completion
of the I2RS client-I2RS Agent association, the default AS value
should be returned to the configured value.</t>
<t hangText="protcol_status (R) "><vspace blankLines="1"/>This
I2RS BGP IM element stores an administrative status for the
protocol activity of "no-shutdown" (0x00) or "shutdown" (0x01).
If the routing entity the I2RS protocol is attached has BGP
"shutdown", no data will be available from any BGP instances.
This is a read-only variable.</t>
<t hangText="shutdown_protocol(R/W)"><vspace blankLines="1"/> A
write of "TRUE" to the "shutdown_protocol" status in the I2RS
Agent indicates the I2RS agent should request a shutdown of the
BGP protocol from the routing function. The default value is
"FALSE" for this I2RS variable. The read of the
"shutdown_protocol" indicate the current status of the
"shutdown_protocol" variable in the I2RS Agent. This is
different than the "protocol_status" which indicates the status
of the protocol: shutdown or not "shutdown" (running).</t>
<t hangText="BGP_ROLE (R)"><vspace blankLines="1"/>The BGP_ROLE
provides a list of BGP Roles found within the BGP Instances
associated with the BGP protocol function on the device
associated with the I2RS agent. These can be ASBR (AS border
router E-BGP), PE (Provider Edge E-BGP), IBGP peer, or RR (Route
Reflector).</t>
<t hangText="bgp-instance-list (R-W)"><vspace blankLines="1"/>
This element provides a list of BGP instances associated with
the BGP routing process associated with the I2RS agent. Each BGP
instance is identified by BGP_INSTANCE_NAME and
BGP_INSTANCE_TYPE (AFI,SAFI tuple) Each instances contains:
local-AS local-router-id, bgp_local_rib, bgp-peer-list described
below.</t>
</list></t>
</section>
<section title="RBNF the I2RS BGP IM Protocol Block">
<t><figure>
<artwork><![CDATA[
<bgp-protocol> ::= < ROUTER-ID >
<AS_Number>
<protocol-status>
<shutdown_protocol>
<BGP_ROLE>
<bgp-instance-list>
<BGP_ROLE> ::= [<ASBR >] [ <PE>]
[ <IBGP>] [<RR>]
<protocol_status> ::= <SHUTDOWN> | <NO_SHUTDOWN>
<shutdown_protocol> ::= <TRUE> | <FALSE>
<bgp-instance-list> ::= (<BGP_INSTANCE> ...)
]]></artwork>
</figure></t>
</section>
</section>
<section title="BGP Instance List (BGP_INSTANCE_LIST)">
<t>This section describes the list of BGP instances
(BGP_INSTANCE_LIST). Each BGP_INSTANCE has the following structure
<list style="symbols">
<t>BGP_INSTANCE_NAME (R/W) - Name for BGP instance</t>
<t>BGP_INSTANCE_CREATION_STATUS (R/W)</t>
<t>BGP_INSTANCE_TYPE (R/W) - Types of BGP Instance</t>
<t>BGP_VENDOR_TYPE (R/W) - Vendor types</t>
<t>AFI (R/W) - address family</t>
<t>SAFI (R/W) - sub-address family</t>
<t>BGP_MAX_PREFIX (R/W></t>
<t>bgp_local_RIB (R/W) -</t>
<t>bgp-peer-list (R/W)</t>
</list></t>
<section title="Elements in BGP_INSTANCE Element">
<t>The definitions of the fields in BGP Instance are: <list
style="hanging">
<t hangText="BGP_INSTANCE_NAME (R/W)"><vspace blankLines="1"/>
The BGP_INSTANCE_NAME provides a name to uniquely identify the
BGP instance. BGP_INSTANCE_NAME identifies a instance with a
particular AFI, SAFI, and BGP_VENDOR_TYPE field. A BGP instance
capabilities may be summarized in the BGP_INSTANCE_TYPE. A write
changes the BGP_INSTANCE_NAME, but</t>
<t hangText="BGP_INSTANCE_CREATION(R/W)"><vspace
blankLines="1"/> is the flag to indicate whether the I2RS client
is requesting the I2RS agent to have the BGP routing protocol
create a new BGP instance for these I2RS parameters. The
internal status includes: <list style="symbols">
<t>Not I2RS created (NOT-I2RS),</t>
<t>I2RS client requests creation
(I2RSCLIENT_CREATE_BGP_INSTANCE),</t>
<t>I2RS Agent rejects creation
(I2RSAGENT_REJECTS_BGP_INSTANCE_CREATE),</t>
<t>I2RS Agent attempting creation
(I2RSAGENT_ATTEMPTS_BGP_INSTANCE_CREATE), and</t>
<t>I2RS Agent failed creation
(I2RSAGENT_FAILS_BGP_INSTANCE_CREATE)</t>
<t>I2RS Agent created BGP instance
(I2RSAGENT_CREATED_BGP_INSTANCE)</t>
</list></t>
<t hangText="BGP_INSTANCE_TYPE (R/W)"><vspace blankLines="1"/>
contains flag field that summarizes common AFI-SAFIs within a
flag word. <list style="symbols">
<t>BGP_IPV4_UNI flag - is a flag for AFI of IPV4 and a SAFI
of Unicast (1).</t>
<t>BGP_IPV4_MULTI flag - is a flag for AFI of IPV4 and a
SAFI Multicast (2).</t>
<t>BGP_L3VPN I is a flag for AFI of IPv4 and a SAFI of MPLS
labels (4).</t>
<t>BGP_IPv4_MVPN- is a flag for AFI of IPv4 and a SAFI of
MCAST VPN (5).</t>
<t>BGP_VPLS - is a flag for an AFI of IPv4 and a SAFI VPLS
(65).</t>
<t>BGP_IPV4_MDT - is a flag for AFI of IPv4 and a SAFI of
BGP MDT (66).</t>
<t>BGP_EVPN - is a flag for APFI of IPv4 and SAFI of EVPNs
(70)</t>
<t>BGP_LS - is a flag for an AFI of IPv4 and a SAFI of BGP
link state (LS) (71).</t>
<t>BGP_IPV4_MPLS_VPN - is a flag for an AFI of IPv4 and a
SAFI MPLS-labled Unicast VPNs (128).</t>
<t>BGP_IPv4_MPLS_MVPN - is a flag for an AFI of IPv4 and a
SAFI MPLS-labeled Multicast VPNs (129).</t>
<t>BGP_IPv4_ROUTE_TARGET - is a flag for the AFI of IPv4 and
a SAFI of Route Target Constraints (132)</t>
<t>BGP_IPv4_FLOW - is a flag for an AFI of IPv4 and a SAFI
of IPv4 dissemination of flow specification rules (133).</t>
<t>BGP_IPv4_VPN_FLOW - is a flag for an AFI of IPv4 and a
SAFI of IPv4 VPN dissemination of flow specification
rules(134).</t>
<t>BGP_L2AD - is a flag for an AFI of IPv4 and a SAFI L2
Auto-discovery (140)</t>
<t>BGP_IPv6_UNI - is a flag for AFI of IPv6 and SAFI of
Unicast (1)</t>
<t>BGP_IPv6_MULTI - is a flag for API of IPv65 and a SAFI of
Multicast (2)</t>
<t>BGP_IPV6_MPLS_VPN - is a flag for AFI of IPV4 and SAFI of
Unicast MPLS VPNS.</t>
</list></t>
<t hangText="BGP_VENDOR_TYPE(R/W)"><vspace blankLines="1"/>
contains a field to specify Vendor specific additions to
identify instance.</t>
<t hangText="bgp_local_RIB_active (R/W)"><vspace
blankLines="1"/> contains the contains a local RIB for BGP
associated with this BGP instance. Every BGP instance has its
own local RIB which contains the routes received by all Peers
(Adj-RIB-IN per Peer), and selected by policy as the active
route by the local BGP speaker's Decision Process.</t>
<t hangText="AFI (R/W)"><vspace blankLines="1"/> The AFI value
from the standard values found in IANA's Address Family
Numbers.</t>
<t hangText="SAFI (R/W)"><vspace blankLines="1"/> The AFI value
from the standard values found in IANA's SAFI values.</t>
<t hangText="bgp_local_RIB (R/W)"><vspace blankLines="1"/>
contains the bgp local RIB associated with this BGP instance.
Every BGP instance has its own local RIB which contains the
routes received by all Peers (Adj-RIB-IN per Peer), and selected
by policy as the active route by the local BGP speaker's
Decision Process.</t>
<t hangText="bgp-peer-list (R/W)"><vspace blankLines="1"/>
contains a list of BGP Peer information (BGP_PEER) associated
with this BGP instance. A write to this BGP Peer list adds a BGP
Peer to this BGP instance.</t>
</list></t>
</section>
<section title="BGP_INSTANCE RBNF">
<t><figure>
<artwork><![CDATA[
<bgp-instance-list> ::= (< bgp-instance > ...)
<BGP_INSTANCE> ::= <BGP_INSTANCE_NAME>
<BGP_INSTANCE_CREATE>
<bgp-instance-type>
<BGP_VENDOR_TYPE>
<AFI>
<SAFI>
<bgp-local-rib>
<bgp-peer-list>
<BGP_INSTANCE_CREATE> ::= <Not-I2RS>
| <I2RSCLIENT_CREATE_BGP_INSTANCE>
| <I2RSAGENT_FAILS_BGP_INSTANCE_CREATE>
| <I2RSAGENT_CREATED_BGP_INSTANCE>
| <I2RSAGENT_BGP_INSTANCE_CREATE)>
| <I2RSAGENT_REJECTS_BGP_INSTANCE_CREATE>
| <I2RSAGENT_ATTEMPTS_BGP_INSTANCE_CREATE>
| <I2RSAGENT_FAILS_BGP_INSTANCE_CREATE>
| <I2RSAGENT_CREATED_BGP_INSTANCE>
< BGP_INSTANCE_TYPE > ::= < BGP_IPV4_UNI >
| < BGP_IPV4_MULTI >
| < BGP_IPV4_L3VPN >
| < BGP_IPV4_MVPN >
| < BGP_VPLS >
| < BGP_MDT >
| < BGP_EVPN >
| < BGP_LS >
| < BGP_IPV4_MPLS_VPN >
| < BGP_IPV4_MPLS_MVPN >
| < BGP_ROUTE_TARGET >
| < BGP_IPV4_FLOW >
| < BGP_IPV4_VPN_FLOW >
| < BGP_IPV4_VRF_LIST >
| < BGP_IPV6_VRF_LIST >
| < BGP_IPV6_UNI >
| < BGP_IPV6_VPNV6 >
<AFI> ::= <IPV4> | <IPV6>
< SAFI> ::= <UNICAST> | <MULTICAST>
| < L3VPN > | <MPLS> | <MVPN>
| < RT_CONSTRAIN >| < IPV4_FLOW_SPEC >
| < VPNV4_FLOW_SPEC >| < CLNS_VPN >
| < NG_MVPN >| < MDT_SAFI >
]]></artwork>
</figure></t>
</section>
</section>
<section title="BGP Peer">
<t>The following are the elements in the BGP_Peer structure <list
style="symbols">
<t>BGP_PEER_NAME (R/W)</t>
<t>Session_address(R/W)</t>
<t>Peer_MAX_PREFIX (R/W)</t>
<t>BGP_Peer_Type</t>
<t>BGP_Peer_create - request to create BGP peer</t>
<t>bgp_rib_in (R)</t>
<t>bgp_rib_out (R/W)</t>
<t>Peer_policy_in (R/W) - Policy set name</t>
<t>Peer_policy_out (R/W) - Policy set name</t>
<t>Peer_policy_in_pset - Policy set (R)</t>
<t>Peer_policy_out_pset - Policy set (R)</t>
<t>Peer_state_info <list style="symbols">
<t>Peer_current_state (R)</t>
<t>Peer_last_state (R)</t>
<t>Peer_down_reason (R)</t>
<t>Peer_Received_Route_Cnt (R)</t>
<t>Peer_Sent_Route_Cnt (R)</t>
<t>Peer_Prefix_high (R)</t>
<t>Peer_prefix_low (R)</t>
<t>Peer_prefix_ave (R)</t>
<t>Peer_prefix_time (R)</t>
<t>Peer_PREFIX_Max (R)</t>
<t>publish_version (R)</t>
</list></t>
</list></t>
<section title="Reading/Writing Policy in I2RS IM model for BGP Peer">
<t>The I2RS Agent's peer inbound and outbound policies are logically
considered to be stored in policy-sets as described in <xref
target="I-D.hares-i2rs-info-model-policy"> </xref>. Policy sets are
uniquely identified by a Policy-set name/identity which the
variables Peer_policy_in and Peer_policy_out refer to. The full
policy set can be read from the variables Peer_policy_in_pset
(inbound bgp peer policy), and Peer_policy_out_pset. A bit of
explanation is in order to describe their use in the I2RS
Client-I2RS Agent exchange.</t>
<t>Some I2RS clients will have long-term connections to I2RS Agents
associated with a BGP routing function, and may choose to transfer
the BGP Policy set or a set of the BGP Policy Components to the
client's local memory. In this case, the I2RS Client could request
the I2RS Agent to simply transfer the BGP Policy Sets name used by
for the BGP Peers to minimize travel along with an indication of the
last time the policy was updated.</t>
<t>The policy variables Peer_policy_in and Peer_policy_out allow
both read and write of the Policy set IDs. The read of the
Peer_policy_in and Peer_policy_out will provide simply the
policy-set name. The write of the Peer_policy_in the I2RS agent will
need to valid the the policy set name. After the policy set name is
validated, the policy set link is placed in Peer_policy_in_pset and
the BGP routing function updated with the policy set lists for the
BGP Peer.</t>
<t>Similarly, if a write occurs to I2RS Agent's Peer_policy_out, the
I2RS agent validates the policy-set name , replaces
Peer_policy_out_pset with the link to the policy-set identified by
the Policy-set name.</t>
<t>Reading either Peer_policy_in_pset will result in the I2RS Agent
passing the policy linked to by the Peer_policy_in name (and linked
to by Peer_policy_in_pset) to the I2RS Client in policy set form.
Similarly, reading the Peer_policy_out_pset will retrieve the policy
set pointed to by the Peer_policy_out name, and transfer the data in
policy set form.</t>
</section>
<section title="Elements in BGP Peer">
<t>The definitions of the fields in BGP peer are: <list
style="hanging">
<t hangText="BGP_PEER_NAME (R/W)"><vspace blankLines="1"/> This
entry uniquely names a BGP Peer. If multiple TCP connections are
utilized between two BGP Peers, then the BGP Peer uniquely
identifies the Peering. An I2RS Write to a BGP_PEER_NAME changes
the BGP_PEER_NAME For the duration of the I2RS client-agent
association for the run-time BGP, but it does not change the
configured BGP Peer name.</t>
<t hangText="Peer_Session_Address (R/W)"><vspace
blankLines="1"/> provides local peer address information and
remote peer address information. Please see the description
below for the details of the BGP session address. A read of this
information pulls this information from the BGP Peer data in the
routing system. The write of this information MAY set-up an TCP
session (either initial or a second in multiple TCP sessions
support</t>
<t hangText="BGP_PEER_TYPE (R/W)"><vspace blankLines="1"/>
contains the status on the type of BGP such as EBGP, IBGP, RR,
RR-Client, PE, or CE. or Vendor specific types.</t>
<t hangText="PEER_MAX_PREFIX(R/W)"><vspace blankLines="1"/>
contains the I2RS Agent's maximum prefixes allowed for this BGP
Peer. A write of the PEER_MAX_PREFIX writes a new value into the
I2RS Agent, and MAY upon configuration be read by the BGP
routing function to set a new Peer_Max_Prefix_run.</t>
<t hangText="BGP_Peer_create(R/W)"><vspace blankLines="1"/>
provides the I2RS creation status for this peer as: not-created
by I2RS, requested by client but not processed, or an I2RS
agent's response to the request: (agent reject peer creation,
agent attempting peer creation, Agent failed peer creation, and
Agent created BGP Peer).</t>
<t hangText="bgp_rib_in"><vspace blankLines="1"/> the BGP routes
received by from this BGP peer in the form of a
bgp-route-list.</t>
<t hangText="bgp_rib_out (R/W)"><vspace blankLines="1"/>
provides a a list of the routes sent to the Peer in the form of
a bgp-route-list.</t>
<t hangText="Peer_policy_in (R/W)"><vspace blankLines="1"/>
contains the policy set name for the inbound peer policy. The
Information model for the policy set is defined in <xref
target="I-D.hares-i2rs-info-model-policy"> </xref>. The I2RS
client reading a Policy_peer_in policy set name MAY match this
against the policy-set names retrieved from the I2RS Agent. The
I2RS Client writing to this Peer_policy_in writes a new policy
name for the BGP peer policy in the I2RS agent, and signals the
I2RS agent to have the bgp routing function to update the run
time policy for this bgp peer for inbound traffic. This updated
policy remains in the BGP Peer until either: a) I2RS Agents
updates it, or b) the I2RS client-agent relationship ceases and
the policy return to configured policy.</t>
<t hangText="Peer_policy_out (R/W)"><vspace blankLines="1"/>
contains of a policy-set name for the outbound. The structure of
a policy set is defined in <xref
target="I-D.hares-i2rs-info-model-policy"> </xref>. An I2RS
client reading this policy set name should match the policy-set
name to outbound policies retrieved from this I2RS Agent. A
write of a new policy-set name to the I2RS agents signals the
I2RS agent to validate the policy set name. After the Policy set
name has been validated, the I2RS Agent SHOULD update I2rs BGP
Peer_policy_out_pset, and signal the routing function to utilize
the new policy identified by the policy-set to policy sets
retrieved from the I2rs Agent. This updated policy remains in
the BGP Peer until either: a) I2RS Agents updates it, or b) the
I2RS client-agent relationship ceases and the BGP policy is
restored to the configured policy.</t>
<t hangText="Peer_policy_in_pset (R)"><vspace blankLines="1"/>
is the variable the I2RS client reads to receive the full
policy-set used for this BGP Peers inbound policy. The I2RS
Agent MUST to keep this policy variable in-sync with the I2RS
Peer-Policy_in.</t>
<t hangText="Peer_policy_out_pset (R)"><vspace blankLines="1"/>
is the variable the I2RS client reads to receive the full
policy-set used for this BGP Peers inbound policy. The I2RS
Agent MUST to keep this policy variable in-sync with the I2RS
Peer-Policy_out.</t>
<t hangText="Peer_state_info"><vspace blankLines="1"/> provides
the following readable state information copied from the BGP
routing function: <list style="symbols">
<t>Peer_current_state (R) - the the current state of the BGP
Peer in the BGP state machine.</t>
<t>Peer_last_state (R) - the the last state of the BGP peer
according to the BGP state machine</t>
<t>Peer_down_reason (R) - the error code and sub-error code
of the last peer down</t>
<t>Peer_received_update_cnt (R) - the count of the received
update packets.</t>
<t>Peer_transmit_update_cnt (R) - the count of the transmit
update packets.</t>
<t>Peer_Receive_Route_Cnt (R) - contains current total number
of unique prefix routes received from peer</t>
<t>Peer_Sent_Route_Cnt (R) - contains current total number
prefix routes sent to peer</t>
<t>MAX_PREFIX_RCV_Limit (R/W) - Maximum unique prefix routes
accepted from peer</t>
<t>MAX_PREFIX_XMT_LIMIT (R/W) - Maximum unique prefix routes
transmit to peer</t>
<t>Peer_prefix_high (R) - contains high water mark count of
total unique routes by prefix received from peer.</t>
<t>Peer_prefix_low (R) - contains the low water mark count
of total unique routes by prefix received from peer.</t>
<t>Peer_Prefix_ave (R) - contains the average number of
prefix routes from peer (running).</t>
</list></t>
</list></t>
</section>
<section title="RBNF for BGP Peer">
<t>RBNF for the BGP Peer portion of BGP IM <figure>
<artwork><![CDATA[
< bgp-peer-list > ::= (< BGP_peer > ...)
<BGP_peer> ::= <BGP_PEER_NAME>
<Peer_Session_Address>
<BGP_Peer_Type>
<PEER_MAX_PREFIX>
<BGP_Peer_create>
<bgp_rib_in>
<bgp_rib_out>
<Peer_policy_in>
<Peer_policy_out>
<Peer_policy_in_pset>
<Peer_policy_out_pset>
[<peer_state_info>]
<BGP_PEER_NAME> ::= (alpha-numeric characters)
<Peer_Session_Address> ::= [<v4-session-address>]
| [<v6-session-address>]
| [<v4-port-session-address>]
< v4-session-address> ::= <LOCAL_IPV4_ADDRESS>
<REMOTE_IPV4_ADDRESS>
<v6-session-address>::= <LOCAL_IPV6_ADDRESS>
<REMOTE_IPV6_ADDRESS>
<v4-port-session-addr>::= <LOCAL_IPV4_ADDRESS>
<LOCAL_TRANSPORT_PORT>
<REMOTE_IPV4_ADDRESS>
<REMOTE_TRANSPORT_PORT>
<BGP_Peer_Type> ::= [<ASBR>][<IBGP>]
[<RR>][<RR-Client>]
[<PE>][<CE>]
[<BGP_VENDOR_types>]
<BGP_VENDOR_type> ::= ...
<BGP_MAX_PREFIX> ::= INTEGER;
<BGP_Peer_create> ::= <Not_I2RS_Created_Peer> |
<I2RSCLIENT_BGP_PEER_CREATE_REQUEST> |
<I2RSAGENT_REJECTS_BGP_PEER_CREATE> |
<I2RSAGENT_ATTEMPTS_BGP_INSTANCE_CREATE> |
<I2RSAGENT_FAILS_BGP_INSTANCE_CREATE> |
<I2RSAGENT_CREATED_BGP_PEER>
/* details on the bgp_route_list are
* contained in the bgp route section */
<bgp_rib_in > ::= <bgp-route-list> /* Adj-RIB-IN */
<bgp_rib_out > ::= <bgp-route-list> /* Adj-RIB-OUT */
/* details on the Policy_set_name and
* Policy_set are imported from
* draft-i2rs-hares-info-model-policy
*/
<bgp_policy_in> ::= <Policy_set_name>;
<bgp_policy_out> ::= <Policy_set_name>;
<bgp_policy_in_pset> ::= <Policy_set>;
<bgp_policy_out_pset> ::= <Policy_set>;
<Peer_state_info>::= <Peer_current_state >
[<Peer_Last_state>]
[<Peer_Received_Route_Cnt>]
[<Peer_Send_Route_Cnt>]
[<PEER_DOWN_REASON>]
[<Peer_received_route_cnt>]
[<Peer_sent_route_cnt>]
[<Peer_prefix_high>]
[<Peer_prefix_low>]
[<Peer_perfix_ave>]
[<Peer_prefix_time>]
[<Peer_PREFIX_MAX>]
[<publish_version>]
<Peer_current_state> ::= < Established > | <IDLE>
| < connect > | <active>
| < open_send > | <open_confirm>
<Peer_Last_state> ::= < Established > | <IDLE>
| < connect > | <active>
| < open_send > | <open_confirm>
<PEER_DOWN_REASON> ::= <ERROR CODE> [< ERROR SUB CODE>]
<ERROR_CODE> ::= <1: Message_Header_Error>
| <2: OPEN_Message_Error>
| <3: UPDATE_Message_Error>
| <4: Hold_Timer_Expired>
| <5: Finite_State_Machine_Error >
| <6: Cease >
| <7: Route_Refresh_Error >
<ERROR_SUB_CODE>::= <HEAD_ERROR_SUB_CODE>
| <OPEN_ERROR_SUB_CODE>
| <UPDATE_ERROR_SUB_CODE>
HEAD_ERROR_SUB_CODE> ::= <1 _ Connection_Not_Synchronized>
| <2 _ Bad_Message_Length>
| <3 _ Bad_Message_Type>
<OPEN_ERROR_SUB_CODE> ::= <1 _ Unsupported Version Number>
| <2 _ Bad_Peer_AS>
| <3 _ Bad_BGP_Identifier>
| <4 _ Unsupported_Optional_Parameter>
| <5 _ Deprecated_see_Appendix_A>
| <6 _ Unacceptable_Hold_Time>
<UPDATE_ERROR_SUB_CODE> ::= <1_ Malformed Attribute List>
| <2_Unrecognized Well_known Attribute>
| <3 _ Missing_Well_known_Attribute>
| <4 _ Attribute_Flags_Error>
| <5 _ Attribute_Length_Error>
| <6 _ Invalid_ORIGIN_Attribute>
| <7 _ Deprecated_see_Appendix A>
| <8 _ Invalid_NEXT_HOP_Attribute>
| <9 _ Optional_Attribute_Error>
| <10 _ Invalid_Network_Field>
| <11 _ Malformed_AS_PATH>
< Route_Refresh_ERROR_SUB_CODE> ::= <1_Invalid_Message_Length >
<Peer_received_route_cnt> ::= COUNTER-64;
<Peer_sent_route_cnt> ::= COUNTER-64:
<Peer_prefix_high> ::= COUNTER-64;
<Peer_prefix_low> ::= COUNTER-64;
<Peer_perfix_ave> ::= COUNTER-64;
<Peer_prefix_time> ::= COUNTER-64;
<Peer_PREFIX_MAX> ::= COUNTER-64;
<publish_version> ::= COUNTER-64;
]]></artwork>
</figure></t>
</section>
</section>
<section title="BGP Route">
<t>The routes associated with each BGP instance are the BGP local RIB,
the AdjRIBIn, and the AdjRIBOut. The BGP local RIB is described below.
The AdjRIBIn is the bgp_rib_in described in the BGP_Peer section. The
policy that sets preferences on the routes received by the peer is
identified by the Peer_policy_in (an policy set name), and fully
readable by the Peer_policy_in_pset (policy set pointer) entry.</t>
<t>The AdjRIBOut is contained in bgp_rib_out (a route list) described
in the BGP Peer Section. The Policy associated with the AdjRIBOut is
identified by the Peer_policy_out (a policy set name) and fully
readable by the Peer_policy_out_pset (policy set pointer).</t>
<section title="BGP Local RIB">
<t>The BGP Local RIB listed below may be created and stored as a
prefix tree, but described below as a bgp_route-list for full
transmission from the I2RS Client to the I2RS Agent. The bgp route
list has a sequence of bgp_routes with the following elements: <list
style="symbols">
<t>BGP_ROUTE_TYPE (R/W)</t>
<t>BGP_ROUTE_CREATE (R/W)</t>
<t>ROUTE_DISTINGUISHER (R/W)</t>
<t>ip-prefix (R/W)</t>
<t>FLOW_route (R/W)</t>
<t>EVPN_route (R/W)</t>
<t>MVPN_route (R/W)</t>
<t>bgp-attribute-list (R/W)</t>
<t>rt_rib_state_info which includes: <list style="symbols">
<t>rt_rib_current_state</t>
<t>rt_rib_last_state</t>
<t>rt_rib_reject_reason</t>
<t>rt_not_preferred</t>
</list></t>
<t>BGP_ROUTE_CREATE (R/W)</t>
</list></t>
<section title="BGP_LOCAL_RIB Entries">
<t>The BGP Local RIB Entries are: <list style="hanging">
<t hangText="BGP_ROUTE_TYPE(R/W)"><vspace blankLines="1"/>
contains a ROUTE TYPE indicator.</t>
<t hangText="bgp-route-prefix (R/W) "><vspace blankLines="1"/>
contains a prefix of the type specified in ROUTE_TYPE.</t>
<t hangText="bgp-attribute-list (R/W) "><vspace
blankLines="1"/> contains a list of bgp attributes.</t>
<t hangText="bgp_route_creation (R/W) "><vspace
blankLines="1"/> provides the status of the creation of a BGP
route requested by a I2RS Client to I2RS agent to be created
in the BGP instance's local route. The status includes
"not-i2rs" routes for routes created by the bgp protocol
function and imported into the I2RS Agent for transmission to
the I2RS Client. For routes created there is a series of
status values that track the I2RS Client request for BGP route
creation.</t>
<t hangText="bgp_status_info (R/W) "><vspace blankLines="1"/>
contains a set of BGP status that inclues: <list
style="symbols">
<t>rt_rib_current_state</t>
<t>rt_rib_last_state</t>
<t>rt_rib_reject_reason</t>
<t>rt_not_preferred</t>
</list></t>
</list></t>
</section>
<section title="BGP_LOCAL_RIB RBNF">
<t>The following is the RBNF for the BGP_LOCAL_RIB: <figure>
<artwork><![CDATA[
/* bgp local rib */
<bgp_local_rib> ::= <bgp-route-list>
<bgp-route-list> ::= (<bgp_route> ...)
<bgp_route> ::= <BGP_ROUTE_TYPE>
<bgp-route-prefix>
<bgp-attribute-list>
<bgp_route_create>
<bgp_rt_state_info>
<BGP_ROUTE_TYPE> ::= <BGP_ROUTE_TYPE_RD> |
| <BGP_ROUTE_TYPE_IPV4>
| <BGP_ROUTE_TYPE_IPV6>
| <BGP_LABELED_IPV4>
| <BGP_LABELED_IPV6>
| <BGP_ROUTE_TYPE_FLOW>
| <BGP_ROUTE_TYPE_EVPN>
| <BGP_ROUTE_TYPE_MVPN>
| <BGP_ROUTE_TYPE_VPLS>
| <BGP_ROUTE_TYPE_BGP_LS>
| <BGP_ROUTE_TYPE_VPLS>
| <BGP_ROUTE_TYPE_L2VPN_SIGNALING_NLRI>
| <BGP_RT_CONSTRAINT>
| <PW_ROUTE>
<bgp-route-prefix> ::= <ipv4_route>
| <ipv6_route>
| <VPN_IPV4_route>
| <VPN_IPV6_route>
| <Labeled_IPv4>
| <Labeled_IPv6>
| <FLOW_route>
| <EVPN_route> /* draft-ietf-l2vp-evpn */
| <MVPN_route> /* RFC6514 */
| <MCAST_VPLS_route> /* RFC7117 */
| <MDT_NLRI> /* RFC6037 */
| <BGP_LS_route> /* draft-ietf-idr-ls-distribution */
| <VPLS_BGP_NLRI> /* RFC4761 */
| <L2VPN_SIGNALING_NLRI> /* RFC6064 */
| <RT_CONSTRAIN> /* RFC4684 */
| <PW_ROUTE> /* RFC7267 */
<bgp-attribute-list> ::=[<bgp_Origin>]
[<bgp_ASPath>]
[<bgp_Nexthop >]
[<bgp_MED>]
[<bgp_LocalPref> ]
[<bgp_ATOMIC_AGGREGATE >]
[<bgp_AGGREGATOR>]
[<bgp_CommAttr >]
[<bgp_ ExtCommAttr >]
[<bgp_ ClusterList >]
[<bgp_ORIGINATOR_ID>]
<ROUTE_DISTINGUISHER> ::= <16-bit ASN: 32-bit-Number >
| <32-bit IPv4_address:16-bit-Number >
| <32-bit ASN:16-bit Number >
<ipv4_route> ::= <IPV4_PREFIX_LENGTH> <IPV4_PREFIX>
<ipv6_route> ::= <IPV6_PREFIX_LENGTH> <IPV6_PREFIX>
<Labeled_IPv4> ::= <LENGTH> <MPLS_LABEL> <IPV4_PREFIX>
<Labeled_IPv6> ::= <LENGTH> <MPLS_LABEL> <IPV6_PREFIX>
<VPN_IPV4> ::= <LENGTH> <MPLS_LABEL>
<ROUTE_DISTINGUISHER> <IPV4_PREFIX>
<VPN_IPV6> ::= <LENGTH> <MPLS_LABEL>
<ROUTE_DISTINGUISHER> <IPV6_PREFIX>
<FLOW_route > ::=<flow_route_index> <acl>
<EVPN_route> ::= (<MVPN_ROUTE_TYPE> (<EVPN_MAC> | <EVPN_AD>
| <EVPN_INCLUSIVE> | <EVPN_ES>)
<EVPN_ROUTE_TYPE> ::= <Ethernet_Auto_Discovery_route>
| <MAC_advertisement_route>
| <Inclusive_Multicast_route>
| <Ethernet_Segment_route>
<MVPN_route> ::= <MVPN_ROUTE_TYPE> (<ipv4_prefix> | <ipv6_prefix> )
<MVPN_ROUTE _TYPE> :: <Intra_AS I_PMSI AD route>
| <Inter_AS_I_PMSI_AD route >
| <S_PMSI_AD route>
| <Leaf_AD route>
| <Source_Active AD route>
| <Shared_Tree_Join route>
| <Source_Tree_Join route>
| <Ethernet_Auto_Discovery route>
| <MAC_Advertisement route>
| <Inclusive_Multicast route>
| < Ethernet_Segment route>
<MCAST_VPLS_route> :: = <MVPN_ROUTE_TYPE> <Selective_Tree_AD_route>
| *lt;Leaf_AD_route>
/* MDT NLRI - IPv4 address that Identifies the PE
* that originated this route.
* The RD identifies the VRF in that PE.
* The group address MUST be a IPv4 multicast
* group address and it is used to build P-tunnel.
* All PEs attached to given MVPN must specify
* the same group address even if the group is
* an SSM group. MDT-SAFIs do not carry RTs, and the group
* address is used to associated a received
* MDT-SAFI with a VRF.
*/
<MDT_NLRI> ::= <ROUTE_DISTINGUISHER> <IPv4-address> <Group Address>
<BGPLS_ROUTE> ::= <BGPLS_ROUTE_TYPE>
(<NODE_NLRI> | <LINK_NLRI>
| <IPv4_Topology_Prefix_NLRI>
| <IPv6_Topology_Prefix_NLRI>)
<BGPLS_ROUTE_TYPE> ::= <NODE_NLRI route>
| <LINK_NLRI route>
| <IPv4_Topology_Prefix route>
| <IPv6_Topology_Prefix route>
/* VPLS_BGP_NLRI explanation
* PE participating in the VPLS must have at least one VE ID.
* VE IDs are typi8cally assigned by the network administrator
* with a scope local to VPLS. Therefore the VE ID should
* only belong to one PE unless the CE is multi-homed.
*
* if the PE is the VE, then it typical has one VE ID.
* if the PE is connected to several u-PEs, it has a distinct
* VE ID for each u-PE. The PE may additionally have a VE ID for
* itself, if asks acts as a VE for that VPLS.
* A label block with demultiplexer labels is
* used to reach the VE ID.
* A VPLS BGP NLRI has a VE ID, a VE Block offset (VBO), a
* block offset size (VBS), and a Label base (LB).
* Labels for local VE are (LB+VBS-1) and for the remote
* (VBO+VBS-1).
*/
<VPLS_BGP_NLRI> ::= <ROUTE_DISTINGUISHER>
<VE_ID> <VE_BlockOffset>
<VE_BlockSize> <Label_Base>
<L2VPN_SIGNALING_NLRI> ::= <ROUTE_DISTINGUISHER> <PE_addr>
<RT_CONSTRAIN> ::= <origin as> <route target>
/* LENGTH of PW_route is prefix length of Route Distinguisher +
* Global_ID + Prefix + AC ID in bits
* default PW route = x length
* other: length is between 96-160 bits
*/
<PW_route> ::= <LENGTH> <ROUTE_DISTINGUISHER>
<Global ID> <Prefix> <AC ID>
<BGP_ROUTE_CREATE> ::= <Not-I2RS_ROUTE> |
<I2RSCLIENT_CREATE_BGP_ROUTE> |
<I2RSAGENT_FAILS_BGP_ROUTE_CREATE> |
<I2RSAGENT_CREATED_BGP_ROUTE> |
<I2RSAGENT_BGP_ROUTE_CREATE)> |
<I2RSAGENT_REJECTS_BGP_ROUTE_CREATE> |
<I2RSAGENT_ATTEMPTS_BGP_ROUTE_CREATE> |
<I2RSAGENT_FAILS_BGP_ROUTE_CREATE> |
<I2RSAGENT_CREATED_BGP_ROUTE>
<bgp_state_info> ::= <rib_current_state>
[<rib_last_state>]
[<Rib_Rejected_REASON>]
[<Not_Preferred>]
<rib_current_state> ::= (< Active > | < InActive > )
| (< primary > | < backup >)
[ < suppress > ]
<rib_last_state> ::= (< Active > | < InActive > )
(<primary > | < backup >)
[ < suppress > ]
<Rib_Rejected_REASON> ::= <peer_withdraw>
| <Policy-Results>
| < Nexthop_Unreachable>
| < Malformed_PEER_DOWN>
| < Malformed_DISCARD>
| < Malformed_IGNORE>
<NOT_Preferred_REASON> ::= <peer_address> | <router_id>
| <Cluster_list_length> | <IGP_METRIC>
| <peer_type> | <origin>
| <WEIGHT or PREFERRED_VALUE>
| <LOCAL_PREFERENCE>
| <ROUTE_TYPE>
| <AS_PATH_LENGTH> | <MED>
]]></artwork>
</figure></t>
</section>
</section>
</section>
</section>
<section title="BGP Policy">
<t>The BGP policy uses policy sets to define the policy for
bgp_policy_in, bgp_policy_out, bgp_policy_in_rpset, and
bgp_policy_out_rpset. Policy sets are defined in <xref
target="I-D.hares-i2rs-info-model-policy"> </xref>.</t>
<t>Policy sets contain informational model definitions for Policy
groups, Policy Rules, PBR IM, local-config IM. The first section below
contains the RBNF for the Policy sets, Policy Groups, and Policy Rules.
The second section below contains the RBNF for the BGP Policy. The first
section below contains the RBNF for these models that is utilized by the
BGP Info-Model (IM).</t>
<section title="BGP IM Policy RBNF">
<t>The BGP IM policy has policy for: <list style="symbols">
<t>BGP IM Tree Matches for Roles</t>
<t>Policy-Rule Extensions for Matching</t>
</list></t>
<t><figure>
<artwork><![CDATA[
/* Role Tree Matches */
<BGP-IM_READ_list> ::= [<BGP-IM-Tree-Match ...]
<BGP-IM_WRITE_list> ::= [<BGP-IM-Tree-Match ...]
<BGP-IM-Tree-Match> ::= <BGP-IM-Tree-Match-protocol-instance>
<BGP-IM-Match-Protocol-instance> ::= (<BGP_protocol> ...)
/* Policy Rule Extensions for BGP */
<Policy-Rule_Match_Nodes_BGP-IM> ::=
(<Policy-Rule_Match_node_BGP-IM> ...)
<Policy-Rule_Match_Values_BGP-IM> ::=
(<Policy-Rule_Match_value_BGP-IM> ...)
<Policy-Rule_Match_operators_BGP-IM> ::=
(<Policy-Rule_operators_BGP-IM> ...)
<Policy-Rule_match_node_BGP-IM> ::= [<Policy-Rule_Match_acl> ]
[<as_path_filter>]
[<community_filter>]
[<MED>]
[<extcommunity_filter >]
[<interface >]
[<ipv4-route-list>]
[<ipv6_rotue_list>]
[<peer-address-list>]
[<mpls_label >]
[<rd_filter >]
[<tag >]
<Policy-Rule_Action_Vars_BGP-IM> ::= [<adc_path_list>]
[<adc_comm_list>]
[comm_mod_args]
<acl> ::= <Policy-Rule_match
<peer-address-list> ::= (Peer_Session_Address ...);
<ipv4-route-list> ::= (<ipv4_route> ..)
<ipv6-route-list> ::= (<ipv6_route> ..)
<BGP_adc_path_list>:: = [<add_as_path_list> |
<delete_as_path_list>
<modify_as_path_list>]
<adc_comm_list>::= [<add_community_list > | /* community list/
[ <delete_community_list >|
<modify_community_list >
<modify_MED >]
<comm_mod_args>::= [<modify_next_hop >]
[<modify_local_preference >]
[<modify_mpls_label >]
[<modify_origin >]
[<modify_ip_dscp >]
[<modify_preference >]
[<modify_bgp_prefference_value>]
[<modify_bgp_preference_weight>]
]]></artwork>
</figure></t>
</section>
<section title="RIB IM Policy RBNF">
<t><figure>
<artwork><![CDATA[
<RIB-IM_READ_list> ::= [<RIB-IM-Tree-Match ...]
<RIB-IM_WRITE_list> ::= [<RIB-IM-Tree-Match ...]
<RIB-IM-Tree-Match> ::= <RIB-IM-Match-routing-instance>
<RIB-IM-Match-interface-list>
<RIM-IM-Match-rib_list>
<RIB-IM-match-route-list;
]]></artwork>
</figure></t>
</section>
<section title="BNP IM Grammar">
<t><figure>
<artwork><![CDATA[
/* A few definitions imported from the policy set */
<policy_set> ::= <Policy_Set_Name> (<Policy_Group> ...)
<Policy_Group> ::= <Policy_Group_Identity>
<Policy_Group_Role>
<Policy_Group_priority>
<Policy_Group_precedence>
(<Policy-Rule>....);
[<Supporting-Policy_Group>]
[<Policy-Group-Extensions>]
<Policy_Rule> ::= <Policy-Rule_identity>
<Policy-Rule_priority>
<Policy-Rule_precedence>
<Policy-Rule_Roles>
(<Policy-Rule_Condition>
<Policy-Rule_Action> ...)
<Policy-Rule_Security_model>
<Policy-Rule_Result>
[<Policy_Rule_rule_extensions>]
<Policy_Result> ::= <Policy_set_name> <Policy_Group_Identifier>
<Policy_Rule_ID> <Policy_Action_Identifier>
<Policy-Rule_Result> <Policy-Rule_Action_result>
<Policy-Rule_identity> ::= <Rule-Name> [<Rule-Secure-Identity>]
<Policy-Rule_priority> ::= INTEGER (0..500);
<Policy-Rule_precedence> :;= INTEGER (0..500);
<Policy-Rule_Roles> ::= (<Policy_Rule_Role> ...);
<Policy-Rule_Condition> ::=<Policy-Rule_node>
(<Policy-Rule_value> ...)
<Policy-Rule_mode>
[<Policy_Rule_Match_Operator>]
[<Policy_Rule_Condition_extension>]
<Policy-Rule_Role> ::=<RESOURCES> | <Policy_Rule_Scope>
<RESOURCES> ::= [<I2RS_AGENT_RESOURCE>]
]]></artwork>
</figure></t>
</section>
</section>
<section title="BGP Grammar">
<t>This section examines the grammar.</t>
<section title="UML grammar">
<t><figure>
<artwork><![CDATA[
This will include a UML version of the grammar.
]]></artwork>
</figure></t>
</section>
<section title="BGP-IM RNBF Grammar">
<t><figure>
<artwork><![CDATA[
<BGP_protocol> ::= < Router ID >
<AS Number>
<protocol_status>
<shutdown_protocol>
<BGP_ROLE>
<bgp-instance-list>
<BGP_ROLE> ::= [<ASBR >] [ <PE>]
[ < IBGP >] [< RR >]
<protocol_status> ::= <sHUTDOWN> | <NO_SHUTDOWN>
<shutdown_protocol> ::= <TRUE> | <FALSE>
<bgp-instance-list> ::= (< BGP_INSTANCE > ...)
/* BGP Instance */
<bgp-instance-list> ::= (< BGP_INSTANCE > ...)
<BGP_INSTANCE> ::= <BGP_INSTANCE_NAME>
<BGP_INSTANCE_CREATE>
<BGP_INSTANCE_TYPE>
<BGP_VENDOR_TYPE>
<AFI>
<SAFI>
<bgp_local_rib>
<bgp-peer-list>
<BGP_INSTANCE_CREATE> ::= <Not-I2RS>
| <I2RSCLIENT_CREATE_BGP_INSTANCE>
| <I2RSAGENT_FAILS_BGP_INSTANCE_CREATE>
| <I2RSAGENT_CREATED_BGP_INSTANCE>
| <I2RSAGENT_BGP_INSTANCE_CREATE)>
| <I2RSAGENT_REJECTS_BGP_INSTANCE_CREATE>
| <I2RSAGENT_ATTEMPTS_BGP_INSTANCE_CREATE>
| <I2RSAGENT_FAILS_BGP_INSTANCE_CREATE>
| <I2RSAGENT_CREATED_BGP_INSTANCE>
< BGP_INSTANCE_TYPE > ::= < BGP_IPV4_UNI >
| < BGP_IPV4_MULTI >
| < BGP_IPV4_L3VPN >
| < BGP_IPV4_MVPN >
| < BGP_VPLS >
| < BGP_MDT >
| < BGP_EVPN >
| < BGP_LS >
| < BGP_IPV4_MPLS_VPN >
| < BGP_IPV4_MPLS_MVPN >
| < BGP_Route_Target >
| < BGP_IPV4_FLOW >
| < BGP_IPV4_VPN_FLOW >
| < BGP_IPV4_VRF_LIST >
| < BGP_IPV6_VRF_LIST >
| < BGP_IPV6_UNI >
| < BGP_IPV6_VPNV6 >
<AFI> ::= <IPV4> | <IPV6>
< SAFI> ::= <UNICAST> | <MULTICAST>
| < L3VPN > | <MPLS> | <MVPN>
| < RT_CONSTRAIN >| <IPV4_FLOW_SPEC >
| < VPNV4_FLOW_SPEC >| <CLNS_VPN >
| < NG_MVPN >| < MDT_SAFI >
/* BGP Peer information */
<bgp-peer-list> ::= (< BGP_peer > ...)
<BGP_peer> ::= <BGP_PEER_NAME>
<Peer_Session_Address>
<BGP_PEER_TYPE>
<PEER_MAX_PREFIX>
<BGP_Peer_create>
<bgp_rib_in>
<bgp_rib_out>
<Peer_policy_in>
<Peer_policy_out>
<Peer_policy_in_pset>
<Peer_policy_out_pset>
[<peer_state_info>]
<BGP_PEER_NAME> ::= (alpha-numeric characters)
<Peer_Session_Address> ::= [<v4-session-address>]
| [<v6-session-address>]
| [<v4-port-session-address>]
<v4-session-address> ::= <LOCAL_IPV4_ADDRESS>
<REMOTE_IPV4_ADDRESS>
<v6-session-address>::= <LOCAL_IPV6_ADDRESS>
<REMOTE_IPV6_ADDRESS>
<v4-port-session-addr>::= <LOCAL_IPV4_ADDRESS>
<LOCAL_TRANSPORT_PORT>
<REMOTE_IPV4_ADDRESS>
<REMOTE_TRANSPORT_PORT>
<BGP_Peer_Type> ::= [<ASBR>][<IBGP>]
[<RR>][<RR-Client>]
[<PE>][<CE>]
[<BGP_VENDOR_TYPES>]
<BGP_VENDOR_TYPE> ::= ...
<BGP_MAX_PREFIX> ::= INTEGER;
<BGP_Peer_create> ::= <Not_I2RS_Created_Peer> |
<I2RSCLIENT_BGP_PEER_CREATE_REQUEST> |
<I2RSAGENT_REJECTS_BGP_PEER_CREATE> |
<I2RSAGENT_ATTEMPTS_BGP_INSTANCE_CREATE> |
<I2RSAGENT_FAILS_BGP_INSTANCE_CREATE> |
<I2RSAGENT_CREATED_BGP_PEER>
/* details on the bgp_route_list are
* contained in the bgp route section */
<bgp_rib_in > ::= <bgp-route-list> /* Adj-RIB-IN */
<bgp_rib_out > ::= <bgp-route-list> /* Adj-RIB-OUT */
/* details on the Policy_set_name and
* Policy_set are imported from
* draft-i2rs-hares-info-model-policy
*/
<bgp_policy_in> ::= <Policy_set_name>;
<bgp_policy_out> ::= <Policy_set_name>;
<bgp_policy_in_pset> ::= <Policy_set>;
<bgp_policy_out_pset> ::= <Policy_set>;
<Peer_state_info>::= <Peer_current_state >
[<Peer_Last_state>]
[<Peer_Received_Route_Cnt>]
[<Peer_Send_Route_Cnt>]
[<PEER_DOWN_REASON>]
[<Peer_received_route_cnt>]
[<Peer_sent_route_cnt>]
[<Peer_prefix_high>]
[<Peer_prefix_low>]
[<Peer_perfix_ave>]
[<Peer_prefix_time>]
[<Peer_PREFIX_MAX>]
[<publish_version>]
<Peer_current_state> ::= < Established > | <IDLE>
| < connect > | <active>
| < open_send > | <open_confirm>
<Peer_Last_state> ::= < Established > | <IDLE>
| < connect > | <active>
| < open_send > | <open_confirm>
<PEER_DOWN_REASON> ::= <ERROR_CODE> [< ERROR_SUB_CODE>]
<ERROR_CODE> ::= <1: Message Header Error>
| <2: OPEN Message Error>
| <3: UPDATE Message Error>
| <4: Hold Timer Expired>
| <5: Finite State Machine Error >
| <6: Cease >
| <7: Route Refresh>
<ERROR_SUB_CODE>::= <HEAD_ERROR_SUB_CODE>
| <OPEN_ERROR_SUB_CODE>
| <UPDATE_ERROR_SUB_CODE>
<HEAD_ERROR SUB CODE> ::= < 1 _ Connection Not Synchronized>
| <2 _ Bad Message Length>
| <3 _ Bad Message Type>
<OPEN_ERROR_SUB_CODE> ::= <1 _ Unsupported Version Number>
| <2 _ Bad Peer AS>
| <3 _ Bad BGP Identifier>
| <4 _ Unsupported Optional Parameter>
| <5 _ [Deprecated _ see Appendix A] >
| <6 _ Unacceptable Hold Time>
<UPDATE_ERROR_SUB_CODE> ::= <1 _ Malformed Attribute List>
| <2 _ Unrecognized Well_known Attribute>
| <3 _ Missing Well_known Attribute>
| <4 _ Attribute Flags Error>
| <5 _ Attribute Length Error>
| <6 _ Invalid ORIGIN Attribute>
| <7 _ [Deprecated _ see Appendix A] >
| <8 _ Invalid NEXT_HOP Attribute>
| <9 _ Optional Attribute Error>
| <10 _ Invalid Network Field>
| <11 _ Malformed AS_PATH>
< Route_Refresh_ERROR_SUB_CODE> ::= <1_Invalid_Message_Length >
<Peer_received_route_cnt> ::= COUNTER-64;
<Peer_sent_route_cnt> ::= COUNTER-64:
<Peer_prefix_high> ::= COUNTER-64;
<Peer_prefix_low> ::= COUNTER-64;
<Peer_perfix_ave> ::= COUNTER-64;
<Peer_prefix_time> ::= COUNTER-64;
<Peer_PREFIX_MAX> ::= COUNTER-64;
<publish_version> ::= COUNTER-64;
/* bgp local rib */
<bgp_local_rib> ::= <bgp-route-list>
<bgp-route-list> ::= (<bgp_route> ...)
<bgp_route> ::= <BGP_ROUTE_TYPE>
<bgp-route-prefix>
<bgp-attribute-list>
<bgp_route_create>
<bgp_rt_state_info>
<BGP_ROUTE_TYPE> ::= <BGP_ROUTE_TYPE_RD> |
| <BGP_ROUTE_TYPE_IPV4>
| <BGP_ROUTE_TYPE_IPV6>
| <BGP_LABELED_IPV4>
| <BGP_LABELED_IPV6>
| <BGP_ROUTE_TYPE_FLOW>
| <BGP_ROUTE_TYPE_EVPN>
| <BGP_ROUTE_TYPE_MVPN>
| <BGP_ROUTE_TYPE_VPLS>
| <BGP_ROUTE_TYPE_BGP_LS>
| <BGP_ROUTE_TYPE_VPLS>
| <BGP_ROUTE_TYPE_L2VPN_SIGNALING_NLRI>
| <BGP_RT_CONSTRAINT>
| <PW_ROUTE>
<bgp-route-prefix> ::= <ipv4_route>
| <ipv6_route>
| <VPN_IPV4_route>
| <VPN_IPV6_route>
| <Labeled_IPv4>
| <Labeled_IPv6>
| <FLOW_route>
| <EVPN_route> /* draft-ietf-l2vp-evpn */
| <MVPN_route> /* RFC6514 */
| <MCAST_VPLS_route> /* RFC7117 */
| <MDT_NLRI> /* RFC6037 */
| <BGP_LS_route> /* draft-ietf-idr-ls-distribution */
| <VPLS_BGP_NLRI> /* RFC4761 */
| <L2VPN_SIGNALING_NLRI> /* RFC6064 */
| <RT_CONSTRAIN> /* RFC4684 */
| <PW_ROUTE> /* RFC7267 */
<bgp-attribute-list> ::=[<bgp_Origin>]
[<bgp_ASPath>]
[<bgp_Nexthop >]
[<bgp_MED>]
[<bgp_LocalPref> ]
[<bgp_ATOMIC_AGGREGATE >]
[<bgp_AGGREGATOR>]
[<bgp_CommAttr >]
[<bgp_ ExtCommAttr >]
[<bgp_ ClusterList >]
[<bgp_ORIGINATOR_ID>]
<ROUTE_DISTINGUISHER> ::= <16-bit ASN: 32-bit-Number >
| <32-bit IPv4_address:16-bit-Number >
| <32-bit ASN:16-bit Number >
<ipv4_route> ::= <IPV4_PREFIX_LENGTH> <IPV4_PREFIX>
<ipv6_route> ::= <IPV6_PREFIX_LENGTH> <IPV6_PREFIX>
<Labeled_IPv4> ::= <LENGTH> <MPLS_LABEL> <IPV4_PREFIX>
<Labeled_IPv6> ::= <LENGTH> <MPLS_LABEL> <IPV6_PREFIX>
<VPN_IPV4> ::= <LENGTH> <MPLS_LABEL>
<ROUTE_DISTINGUISHER> <IPV4_PREFIX>
<VPN_IPV6> ::= <LENGTH> <MPLS_LABEL>
<ROUTE_DISTINGUISHER> <IPV6_PREFIX>
<FLOW_route > ::=<flow_route_index> <acl>
<EVPN_route> ::= (<MVPN_ROUTE_TYPE> (<EVPN_MAC> | <EVPN_AD>
| <EVPN_INCLUSIVE> | <EVPN_ES>)
<EVPN_ROUTE_TYPE> ::= <Ethernet_Auto_Discovery_route>
| <MAC_Advertisement_route>
| <Inclusive_Multicast_route>
| <Ethernet_Segment_route>
<MVPN_route> ::= <MVPN_ROUTE_TYPE> (<ipv4_prefix> | <ipv6_prefix> )
<MVPN_ROUTE_TYPE> ::= <Intra_AS_I_PMSI_AD_route>
| <Inter_AS_I_PMSI_AD_route >
| <S_PMSI_AD_route>
| <Leaf_AD_route>
| <Source_Active_AD_route>
| <Shared_Tree_Join_route>
| <Source_Tree_Join_route>
<MCAST_VPLS_route> :: = <MVPN_ROUTE_TYPE> <Selective_Tree_AD_route>
| <Leaf_AD_route>
/* MDT NLRI - IPv4 address that Identifies the PE
* that originated this route.
* The RD identifies the VRF in that PE.
* The group address MUST be a IPv4 multicast
* group address and it is used to build P-tunnel.
* All PEs attached to given MVPN must specify
* the same group address even if the group is
* an SSM group. MDT-SAFIs do not carry RTs, and the group
* address is used to associated a received
* MDT-SAFI with a VRF.
*/
<MDT_NLRI> ::= <ROUTE_DISTINGUISHER> <IPV4_ADDRESS> <GROUP_ADDRESS>
<BGPLS_ROUTE> ::= <BGPLS_ROUTE_TYPE>
(<NODE_NLRI> | <LINK_NLRI>
| <IPv4_Topology_Prefix_NLRI>
| <IPv6_Topology_Prefix_NLRI>)
<BGPLS_ROUTE_TYPE> ::= <NODE_NLRI_route>
| <LINK_NLRI route>
| <IPv4_Topology_Prefix route>
| <IPv6_Topology_Prefix route>
/* VPLS_BGP_NLRI explanation
* PE participating in the VPLS must have at least one VE ID.
* VE IDs are typi8cally assigned by the network administrator
* with a scope local to VPLS. Therefore the VE ID should
* only belong to one PE unless the CE is multi-homed.
*
* if the PE is the VE, then it typical has one VE ID.
* if the PE is connected to several u-PEs, it has a distinct
* VE ID for each u-PE. The PE may additionally have a VE ID for
* itself, if asks acts as a VE for that VPLS.
* A label block with demultiplexer labels is
* used to reach the VE ID.
* A VPLS BGP NLRI has a VE ID, a VE Block offset (VBO), a
* block offset size (VBS), and a Label base (LB).
* Labels for local VE are (LB+VBS-1) and for the remote
* (VBO+VBS-1).
*/
<VPLS_BGP_NLRI< ::= <ROUTE_DISTINGUISHER>
<VE_ID> <VE_BLOCKOFFSET>
<VE_BLOCKSIZE> <LABEL_BASE>
<L2VPN_SIGNALING_NLRI> ::= <ROUTE_DISTINGUISHER> <PE_ADDRESS>
<RT_CONSTRAIN> ::= <ORIGIN_AS> <ROUTE_TARGET>
/* LENGTH of PW_route is prefix length of Route Distinguisher +
* Global_ID + Prefix + AC ID in bits
* default PW route = x length
* other: length is between 96-160 bits
*/
<PW_route> ::= <LENGTH> <ROUTE_DISTINGUISHER>
<GLOBAL_ID> <PREFIX> <AC_ID>
<BGP_ROUTE_CREATE> ::= <Not-I2RS_ROUTE> |
<I2RSCLIENT_CREATE_BGP_ROUTE> |
<I2RSAGENT_FAILS_BGP_ROUTE_CREATE> |
<I2RSAGENT_CREATED_BGP_ROUTE> |
<I2RSAGENT_BGP_ROUTE_CREATE)> |
<I2RSAGENT_REJECTS_BGP_ROUTE_CREATE> |
<I2RSAGENT_ATTEMPTS_BGP_ROUTE_CREATE> |
<I2RSAGENT_FAILS_BGP_ROUTE_CREATE> |
<I2RSAGENT_CREATED_BGP_ROUTE>
<bgp_state_info> ::= <rib_current_state>
[<rib_last_state>]
[<Rib_Rejected_REASON>]
[<Not_Preferred>]
<rib_current_state> ::= (<Active> | <InActive> )
| (<primary> | <backup>)
[ <suppress> ]
<rib_last_state> ::= (<Active> | <InActive> )
(<primary> | <backup>)
[ <suppress> ]
<Rib_Rejected_REASON> ::= <peer_withdraw>
| <Policy-Results>
| < Nexthop_Unreachable>
| < Malformed_PEER_DOWN>
| < Malformed_DISCARD>
| < Malformed_IGNORE>
<NOT_Preferred_REASON> ::= <peer_address> | <router_id>
| <Cluster_list_length> | <IGP_METRIC>
| <peer_type> | <origin>
| <weight or preferred_value>
| < Local_Preference | < route_type >
| <AS_PATH_LENGTH> | < MED >
/* BGP Policy */
<BGP-IM_READ_list> ::= [<BGP-IM-Tree-Match ...]
<BGP-IM_WRITE_list> ::= [<BGP-IM-Tree-Match ...]
<BGP-IM-Tree-Match> ::= <BGP-IM-Tree-Match-protocol-instance>
<BGP-IM-Match-Protocol-instance> ::= (<BGP_protocol> ...)
/* Policy Rule Extensions for BGP */
<Policy-Rule_Match_Nodes_BGP-IM> ::=
(<Policy-Rule_Match_node_BGP-IM> ...)
<Policy-Rule_Match_Values_BGP-IM> ::=
(<Policy-Rule_Match_value_BGP-IM> ...)
<Policy-Rule_Match_operators_BGP-IM> ::=
(<Policy-Rule_operators_BGP-IM> ...)
<Policy-Rule_match_node_BGP-IM> ::= [<Policy-Rule_Match_acl> ]
[<as-path-filter>]
[<community-filter>]
[<MED>]
[<extcommunity-filter>]
[<interface>]
[<ipv4-route-list>]
[<ipv6-rotue-list>]
[<peer-address-list>]
[<MPLS_LABEL>]
[<rd-filter>]
[<TAG>]
<Policy-Rule_Action_Vars_BGP-IM> ::= [<adc_path_list>]
[<adc_comm_list>]
[comm_mod_args]
<acl> ::= <Policy-Rule_match
/* BGP policy_lists */
<peer-address-list> ::= (Peer_Session_Address ...);
<ipv4-route-list> ::= (<ipv4_route> ..)
<ipv6-route-list> ::= (<ipv6_route> ..)
<BGP_adc_path_list>:: = [<add_as_path_list> |
<delete_as_path_list>
<modify_as_path_list>]
<adc_comm_list>::= [<add_community_list> | /* community list */
[<delete_community_list>|
<modify_community_list>
<modify_med>]
<comm_mod_args>::= [<modify_next_hop >]
[<modify_local_preference>]
[<modify_mpls_label>]
[<modify_origin>]
[<modify_ip_dscp>]
[<modify_preference>]
[<modify_bgp_prefference_value>]
[<modify_bgp_preference_weight>]
]]></artwork>
</figure></t>
</section>
</section>
<section anchor="IANA" title="IANA Considerations">
<t>This draft includes no request to IANA.</t>
</section>
<section title="Security Considerations">
<t>TBD.</t>
</section>
</middle>
<back>
<references title="Informative References">
<?rfc include="reference.RFC.2119"?>
<?rfc include="reference.RFC.3060"?>
<?rfc include="reference.RFC.3644"?>
<?rfc include="reference.RFC.5394"?>
<?rfc include="reference.RFC.5511"?>
<?rfc include="reference.RFC.6536"?>
<?rfc include='reference.I-D.ietf-i2rs-architecture'?>
<?rfc include='reference.I-D.ietf-i2rs-rib-info-model'?>
<?rfc include='reference.I-D.ietf-netconf-restconf'?>
<?rfc include='reference.I-D.white-i2rs-use-case'?>
<?rfc include='reference.I-D.keyupate-i2rs-bgp-usecases'?>
<?rfc include='reference.I-D.hares-i2rs-info-model-policy'?>
</references>
</back>
</rfc>
| PAFTECH AB 2003-2026 | 2026-04-24 04:12:51 |