One document matched: draft-ietf-sfc-control-plane-00.xml


<?xml version="1.0" encoding="US-ASCII"?>
<!-- edited with XMLSPY v5 rel. 3 U (http://www.xmlspy.com)
     by Daniel M Kohn (private) -->
<!DOCTYPE rfc SYSTEM "rfc2629.dtd" [
<!ENTITY rfc2119 PUBLIC "" "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2119.xml">
]>
<?rfc toc="yes"?>
<?rfc symrefs="yes"?>
<?rfc sortrefs="yes"?>
<?rfc rfcedstyle="yes"?>
<?rfc strict="no"?>
<?rfc iprnotified="yes"?>
<?rfc tocdepth="3"?>
<?rfc inline="yes"?>
<?rfc compact="yes"?>
<?rfc subcompact="no"?>
<?rfc comments="yes"?>
<rfc category="info" docName="draft-ietf-sfc-control-plane-00"
     ipr="trust200902">
  <front>
    <title abbrev="SFC Control Plane">Service Function Chaining (SFC) Control
    Plane Components & Requirements</title>

    <author fullname="Hongyu Li" initials="H." surname="Li">
      <organization>Huawei</organization>

      <address>
        <postal>
          <street>Huawei Industrial Base,Bantian,Longgang</street>

          <region>Shenzhen</region>

          <country>China</country>
        </postal>

        <email>hongyu.li@huawei.com</email>
      </address>
    </author>

    <author fullname="Qin Wu" initials="Q." surname="Wu">
      <organization>Huawei</organization>

      <address>
        <postal>
          <street>101 Software Avenue, Yuhua District</street>

          <city>Nanjing</city>

          <region>Jiangsu</region>

          <code>210012</code>

          <country>China</country>
        </postal>

        <email>bill.wu@huawei.com</email>
      </address>
    </author>

    <author fullname="Yong(Oliver) Huang" initials="O." surname="Huang">
      <organization>Huawei</organization>

      <address>
        <postal>
          <street>Huawei Industrial Base,Bantian,Longgang</street>

          <region>Shenzhen</region>

          <country>China</country>
        </postal>

        <email>oliver.huang@huawei.com</email>
      </address>
    </author>

    <author fullname="Mohamed Boucadair" initials="M" role="editor"
            surname="Boucadair">
      <organization>France Telecom</organization>

      <address>
        <postal>
          <street>Rennes 35000</street>

          <country>France</country>
        </postal>

        <email>mohamed.boucadair@orange.com</email>
      </address>
    </author>

    <author fullname="Christian Jacquenet" initials="C" surname="Jacquenet">
      <organization>France Telecom</organization>

      <address>
        <postal>
          <street>Rennes 35000</street>

          <country>France</country>
        </postal>

        <email>christian.jacquenet@orange.com</email>
      </address>
    </author>

    <author fullname="Walter Haeffner" initials="W." surname="Haeffner">
      <organization abbrev="Vodafone">Vodafone D2 GmbH</organization>

      <address>
        <postal>
          <street>Ferdinand-Braun-Platz 1</street>

          <region>Duesseldorf</region>

          <code>40549</code>

          <country>DE</country>
        </postal>

        <email>walter.haeffner@vodafone.com</email>
      </address>
    </author>

    <author fullname="Seungik Lee" initials="S. " surname="Lee">
      <organization abbrev="ETRI">ETRI</organization>

      <address>
        <postal>
          <street>218 Gajeong-ro Yuseung-Gu</street>

          <!-- Reorder these if your country does things differently -->

          <city>Daejeon</city>

          <region></region>

          <code>305-700</code>

          <country>Korea</country>
        </postal>

        <phone>+82 42 860 1483</phone>

        <email>seungiklee@etri.re.kr</email>

        <!-- uri and facsimile elements may also be added -->
      </address>
    </author>

    <author fullname="Ron Parker" initials="R." surname="Parker">
      <organization>Affirmed Networks</organization>

      <address>
        <postal>
          <street>Acton</street>

          <region>MA</region>

          <code>01720</code>

          <country>USA</country>
        </postal>

        <email>ron_parker@affirmednetworks.com</email>
      </address>
    </author>

    <author fullname="Linda Dunbar" initials="L." surname="Dunbar">
      <organization>Huawei Technologies</organization>

      <address>
        <postal>
          <street></street>

          <city></city>

          <region></region>

          <code></code>

          <country>USA</country>
        </postal>

        <email>ldunbar@huawei.com</email>
      </address>
    </author>

    <author fullname="Andrew Malis" initials="A." surname="Malis">
      <organization>Huawei Technologies</organization>

      <address>
        <postal>
          <street></street>

          <city></city>

          <region></region>

          <code></code>

          <country>USA</country>
        </postal>

        <email>agmalis@gmail.com</email>
      </address>
    </author>

    <author fullname="Joel M. Halpern" initials="J." surname="Halpern">
      <organization>Ericsson</organization>

      <address>
        <postal>
          <street></street>
        </postal>

        <email>joel.halpern@ericsson.com</email>
      </address>
    </author>

    <author fullname="Tirumaleswar Reddy" initials="T." surname="Reddy">
      <organization abbrev="Cisco">Cisco Systems, Inc.</organization>

      <address>
        <postal>
          <street>Cessna Business Park, Varthur Hobli</street>

          <street>Sarjapur Marathalli Outer Ring Road</street>

          <city>Bangalore</city>

          <region>Karnataka</region>

          <code>560103</code>

          <country>India</country>
        </postal>

        <email>tireddy@cisco.com</email>
      </address>
    </author>

    <author fullname="Prashanth Patil" initials="P." surname="Patil">
      <organization abbrev="Cisco">Cisco Systems, Inc.</organization>

      <address>
        <postal>
          <street></street>

          <city>Bangalore</city>

          <country>India</country>
        </postal>

        <email>praspati@cisco.com</email>
      </address>
    </author>

    <date year="" />

    <workgroup>Service Function Chaining (sfc)</workgroup>

    <keyword>differentiated services</keyword>

    <keyword>differentiated forwarding</keyword>

    <keyword>traffic steering</keyword>

    <keyword>classification</keyword>

    <keyword>service delivery</keyword>

    <keyword>service operation</keyword>

    <keyword>service innovation</keyword>

    <keyword>service agility</keyword>

    <keyword>flexible service</keyword>

    <keyword>service provisioning</keyword>

    <abstract>
      <t>This document describes requirements for conveying information
      between Service Function Chaining (SFC) control elements (including
      management components) and SFC functional elements. Also, this document
      identifies a set of control interfaces to interact with SFC-aware
      elements to establish, maintain or recover service function chains. This
      document does not specify protocols nor extensions to existing
      protocols.</t>

      <t>This document exclusively focuses on SFC deployments that are under
      the responsibility of a single administrative entity. Inter-domain
      considerations are out of scope.</t>
    </abstract>
  </front>

  <middle>
    <section anchor="intro" title="Introduction">
      <t>The dynamic enforcement of a service-derived forwarding policy for
      packets entering a network that supports advanced Service Functions
      (SFs) has become a key challenge for operators. Typically, many advanced
      Service Functions (e.g., Performance Enhancement Proxies (<xref
      target="RFC3135"></xref>), NATs <xref target="RFC3022"></xref><xref
      target="RFC6333"></xref><xref target="RFC6146"></xref>, firewalls <xref
      target="I-D.ietf-opsawg-firewalls"></xref>, etc.) are solicited for the
      delivery of value-added services, particularly to meet various service
      objectives such as IP address sharing, avoiding covert channels,
      detecting and protecting against ever increasing Denial-of-Service (DoS)
      attacks, etc.</t>

      <t>Because of the proliferation of such advanced service functions
      together with complex service deployment constraints that demand more
      agile service delivery procedures, operators need to rationalize their
      service delivery logics and master their complexity while optimising
      service activation time cycles. The overall problem space is described
      in <xref target="RFC7498"></xref>. A more in-depth discussion on use
      cases can be found in <xref
      target="I-D.ietf-sfc-use-case-mobility"></xref> and <xref
      target="I-D.ietf-sfc-dc-use-cases"></xref>.</t>

      <t><xref target="I-D.ietf-sfc-architecture"></xref> presents a model
      addressing the problematic aspects of existing service deployments,
      including topological dependence and configuration complexity. It also
      describes an architecture for the specification, creation, and ongoing
      maintenance of Service Function Chains (SFC) within a network. That is,
      how to define an ordered set of Service Functions and ordering
      constraints that must be applied to packets and/or frames and/or flows
      selected as a result of classification.</t>

      <section title="Scope">
        <t>While <xref target="I-D.ietf-sfc-architecture"></xref> focuses on
        data plane considerations, this document describes requirements for
        conveying information between SFC control elements (including
        management components) and SFC functional elements. Also, this
        document identifies a set of control interfaces to interact with
        SFC-aware elements to establish, maintain or recover service function
        chains.</t>

        <t>Both distributed and centralized control plane schemes to install
        SFC-related state and influence forwarding policies are discussed.</t>

        <t>This document does not make any assumption on the deployment use
        cases. In particular, the document implicitly covers fixed, mobile,
        data center networks and any combination thereof.</t>

        <t>This document does not make any assumption about which control
        protocol to use, whether one or multiple control protocols are
        required, or whether the same or distinct control protocols will be
        invoked for each of the control interfaces. It is out of scope of this
        document to specify a profile for an existing protocol, to define
        protocol extensions, or to select a protocol.</t>

        <t>Considerations related to the chaining of Service Functions that
        span domains owned by multiple administrative entities are out of
        scope.</t>

        <t>It is out of scope of this document to discuss SF-specific control
        and policy enforcement schemes; only SFC considerations are
        elaborated, regardless of the various connectivity services that may
        be supported in the SFC domain. Likewise, only the control of
        SFC-aware elements is discussed.</t>

        <t>Service catalogue (including guidelines for deriving service
        function chains) is out of scope.</t>
      </section>

      <section title="Terminology">
        <t>The reader should be familiar with the terms defined in <xref
        target="RFC7498"></xref> and <xref
        target="I-D.ietf-sfc-architecture"></xref>.</t>

        <t>The document makes use of the following terms:<?rfc subcompact="yes" ?><list
            style="symbols">
            <t>SFC data plane functional element: Refers to SFC-aware Service
            Function, Service Function Forwarder (SFF), SFC Proxy, or SFC
            Classifier as defined in the SFC data plane architecture <xref
            target="I-D.ietf-sfc-architecture"></xref>.</t>

            <t>SFC Control Element: A logical entity that instructs one or
            more SFC data plane functional elements on how to process packets
            within an SFC-enabled domain.</t>

            <t>SFC Classification entry: Refers to an entry maintained by an
            SFC Classifier that reflects the policies for binding an incoming
            flow/packet to a given SFC. Actions are associated with matching
            criteria. For example, packets can be marked with the appropriate
            SFC-related information to differentiate flows so that subsequent
            SFFs can forward the flows to a sequence of SFs in a given order.
            The set of classification entries maintained by a Classifier are
            referred to as in the classification policy table.</t>

            <t>SFC Forwarding Policy Table: this table reflects the
            SFC-specific traffic forwarding policy enforced by SFF components
            for every relevant incoming packet that is associated to one of
            the existing SFCs.<list style="empty">
                <t>[[Note: The question of whether the data plane operates
                just in terms of SFP IDs or needs SFC IDs, as described in
                this version of the draft, is still under discussion among the
                authors.]]</t>
              </list><?rfc subcompact="no" ?></t>
          </list></t>
      </section>

      <section title="Assumptions">
        <t>This document adheres to the assumptions listed in Section 1.2 of
        <xref target="I-D.ietf-sfc-architecture"></xref>.</t>

        <t>This document does not make any assumptions about the co-location
        of SFC data plane functional elements; this is deployment-specific.
        This document can accommodate a variety of deployment contexts such as
        (but not limited to): <?rfc subcompact="yes" ?></t>

        <t><list style="symbols">
            <t>A Service Function Forwarder (SFF) can connect instances of the
            same or distinct SFs.</t>

            <t>A SF instance can be serviced by one or multiple SFFs.</t>

            <t>One or multiple SFs can be co-located with a SFF.</t>

            <t>A boundary node (that connects one SFC-enabled domain to a node
            either located in another SFC-enabled domain or in a domain that
            is SFC-unaware) can act as an egress node and an ingress node for
            the same flow.</t>

            <t>Distinct ingress and egress nodes may be crossed by a packet
            when forwarded in an SFC-enabled domain.</t>

            <t>Distinct ingress nodes may be solicited for each traffic
            direction (e.g., upstream and downstream).</t>

            <t>An ingress node can embed a Classifier.</t>

            <t>An ingress node may not embed a Classifier, but it can be
            responsible for dispatching flows among a set of Classifiers.</t>

            <t>The same boundary node may act as an ingress node, an egress
            node, and also embed a Classifier.</t>

            <t>A Classifier can be hosted in a node that embeds one or more
            SFs.</t>

            <t>Many network elements within an SFC-enabled domain may behave
            as egress/ingress nodes.<?rfc subcompact="no" ?></t>
          </list></t>

        <t>Furthermore, the following assumptions are made:<?rfc subcompact="yes" ?><list
            style="symbols">
            <t>A Control Element can be co-located with a Classifier, SFF or
            SF.</t>

            <t>One or multiple Control Elements can be deployed in an
            SFC-enabled domain.</t>

            <t>State synchronization between Control Elements is out of
            scope.<?rfc subcompact="no" ?></t>
          </list></t>
      </section>
    </section>

    <section title="Generic Considerations">
      <t></t>

      <section title="Generic Requirements">
        <t>For deployments that would require so, SFC forwarding must be
        allowed even if no control protocols are enabled. Static configuration
        must be allowed.</t>

        <t>A permanent association between an SFC data plane element with a
        Control Element must not be required; specifically, the SFC-enabled
        domain must keep on processing incoming packets according to the SFC
        instructions even during temporary unavailability events of control
        plane components. SFC implementations that do not meet this
        requirement will suffer from another flavor of the constrained high
        availability issue, discussed in Section 2.3 of <xref
        target="RFC7498"></xref>, supposed to be solved by SFC designs.</t>
      </section>

      <section title="SFC Control Plane Bootstrapping">
        <t>The interface that is used to feed the SFC control plane with
        service objectives and guidelines is not part of the SFC control plane
        itself. Therefore, this document assumes the SFC control plane is
        provided with a set of information that is required for proper SFC
        operation with no specific assumption about how this information is
        collected/provisioned, nor about the structure of such information.
        The following information that is likely to be provided to the SFC
        control plane at bootstrapping includes (non-exhaustive list): <?rfc subcompact="yes" ?></t>

        <t><list style="symbols">
            <t>Locators for Classifiers/SFF/SFs/Proxies, etc.</t>

            <t>SFs serviced by each SFF.</t>

            <t>A list of service function chains, including how they are
            structured and unambiguously identified.</t>

            <t>Status of each SFC: active/pre-deployment phase/etc. A SFC can
            be defined at the management level and instantiated in an
            SFC-enabled domain for pre-deployment purposes (e.g., testing).
            Actions to activate, modify or withdraw an SFC are triggered by
            the control plane. Nevertheless, this document does not make any
            assumption about how an operator instructs the control plane.</t>

            <t>A list of classification guidelines and/or rules to bind flows
            to SFCs/SFPs.</t>

            <t>Optionally, (traffic/CPU/memory) load balancing objectives at
            the SFC level or on a per node (e.g., per-SF/SFF/Proxy) basis.</t>

            <t>Security credentials.</t>

            <t>Context information that needs to be shared on a per SFC
            basis.<?rfc subcompact="no" ?></t>
          </list></t>

        <t>Also, the SFC control plane may gather the following information
        from an SFC-enabled domain at bootstrapping (non-exhaustive list). How
        this information is collected is left unspecified in this document:
        <?rfc subcompact="yes" ?></t>

        <t><list style="symbols">
            <t>The list of active SFC-aware SFs (including their
            locators).</t>

            <t>The list of SFFs and the SFs that are attached to.</t>

            <t>The list of enabled SFC Proxies, and the list of SFC-unware SFs
            attached to.</t>

            <t>The list of active SFCs/SFPs as enabled in an SFC-enabled
            domain.</t>

            <t>The list of Classifiers and their locators, so as to retrieve
            the classification policy table for each Classifier, in
            particular.</t>

            <t>The SFC forwarding policy tables maintained by SFFs.<?rfc subcompact="no" ?></t>
          </list></t>

        <t>During the bootstrapping phase, a Control Element may detect a
        conflict between the running configuration in an SFC data plane
        element and the information maintained by the control plane.
        Consequently, the control plane undertakes appropriate actions to fix
        those conflicts. This is typically achieved by invoking one of the
        interfaces defined in <xref target="refint"></xref>.</t>
      </section>

      <section title="Coherent Setup of an SFC-enabled Domain">
        <t>Various transport encapsulation schemes and/or variations of SFC
        header implementations may be supported by one or several nodes of an
        SFC-enabled domain. For the sake of coherent configuration, the SFC
        control plane is responsible for instructing all the involved SFC data
        plane functional elements about the behavior to adopt to select the
        transport encapsulation scheme(s), the version of the SFC header to
        enable, etc.</t>
      </section>
    </section>

    <section title="SFC Control Plane: Reference Architecture & Interfaces">
      <t></t>

      <section title="Reference Architecture">
        <t>The SFC control plane is responsible for the following:<?rfc subcompact="yes" ?><list
            style="symbols">
            <t>Build and monitor the service-aware topology. For example, this
            can be achieved by means of dynamic SF discovery techniques. Those
            means are out of scope of this document.</t>

            <t>Maintain a repository of service function chains, SFC matching
            criteria to bind flows to a given service function chain, and
            mapping between service function chains and SFPs.</t>

            <t>Guarantee the coherency of the configuration and the operation
            of an SFC-enabled domain.</t>

            <t>Dynamically compute a service-aware forwarding path
            (distributed model, see <xref target="cd"></xref>)</t>

            <t>Determine a forwarding path in the context of a centralized
            deployment model (see <xref target="cd"></xref>).</t>

            <t>Update service function chains or adjust SFPs (e.g., for
            restoration purposes) based on various inputs (e.g., external
            policy context, path alteration, SF unavailability, SF withdrawal,
            service decommissioning, etc.).</t>

            <t>Populate SFC forwarding policy tables of involved SFC data
            plane elements and provides Classifiers with traffic
            classification rules.<?rfc subcompact="no" ?></t>
          </list></t>

        <t><xref target="arch"></xref> shows the overall SFC control plane
        architecture, including interface reference points.</t>

        <t>This document does not elaborate on the internal decomposition of
        the SFC Control & Management Plane functional blocks. The
        components within the SFC Control & Management Planes and their
        interactions are out of scope.</t>

        <t>As discussed in <xref target="cd"></xref>, the SFC control plane
        can be implemented in a (logically) centralized or distributed
        fashion.</t>

        <t><figure align="left" anchor="arch"
            title="SFC Control Plane: Overview">
            <artwork><![CDATA[
               +----------------------------------------------+
               |                                              |
               |       SFC  Control & Management Planes       |
       +-------|                                              |
       |       |                                              |
       C1      +------^-----------^-------------^-------------+
+---------------------|C3---------|-------------|-------------+
|      |            +----+        |             |             |
|      |            | SF |        |C2           |C2           |
|      |            +----+        |             |             |
| +----V--- --+       |           |             |             |
| |   SFC     |     +----+      +-|--+        +----+          |
| |Classifier |---->|SFF |----->|SFF |------->|SFF |          |
| |   Node    |<----|    |<-----|    |<-------|    |          |
| +-----------+     +----+      +----+        +----+          |
|                     |           |              |            |
|                     |C2      -------           |            |
|                     |       |       |     +-----------+ C4  |
|                     V     +----+ +----+   | SFC Proxy |-->  |
|                           | SF | |SF  |   +-----------+     |
|                           +----+ +----+                     |
|                             |C3    |C3                      |
|  SFC Data Plane Components  V      V                        |
+-------------------------------------------------------------+
]]></artwork>
          </figure></t>

        <t>Note, the SFC control plane must be able to invoke SFC OAM
        mechanisms, and to determine the results of OAM operations. </t>
      </section>

      <section anchor="cd" title="Centralized vs. Distributed">
        <t>The SFC control plane can be (logically) centralized, distributed
        or a combination thereof. Whether one or multiple SFC Control Elements
        are enabled is deployment-specific. Nevertheless, the following
        comments can be made:</t>

        <t><list style="hanging">
            <t
            hangText="SFC management (including SFC monitoring and supervision):">is
            likely to be centralized.</t>

            <t hangText="SFC Mapping Rules:">i.e., service instructions to
            bind a flow to a service function chain are likely to be managed
            by a central SFC Control Element, but the resulting policies can
            be shared among several Control Elements. Note, these policies can
            be complemented with local information (e.g., an IPv4 address/IPv6
            prefix assigned to a customer) because such information may not be
            available to the central entity but known only during network
            attachment phase.</t>

            <t hangText="Path computation:">can be either distributed or
            centralized. Distributed path computation means that the selection
            of the exact sequence of SF functions that a packet needs to
            invoke (along with instances and/or SFF locator information) is a
            result of a distributed path selection algorithm executed by
            involved nodes. For some traffic engineering proposes, the SFP may
            be constrained by the control plane; as such, some SFPs can be
            fully specified (i.e., list all the SFF/SFs that need to be
            solicited) or partially specified (e.g., exclude some nodes,
            explicitly select which instance of a given SF needs to be
            invoked, etc.).</t>

            <t hangText="SFC Resiliency (including restoration)">refers to
            mechanisms to ensure high available service function chains. It
            includes means to detect node/link/path failures. Both centralized
            and distributed mechanism to ensure SFC resiliency can be
            envisaged.</t>
          </list></t>

        <t>Implementing a (logically) centralized path computation engine
        requires information to be dynamically communicated to the central SFC
        Control Element, such as the list of available SF instances, SFF
        locators, load status, SFP availability, etc.</t>
      </section>

      <section anchor="refint" title="Interface Reference Points">
        <t>The following sub-sections describe the interfaces between the SFC
        Control & Management Planes, as well as various SFC data plane
        elements. </t>

        <section anchor="c1"
                 title="C1: Interface between SFC Control Plane & SFC Classifier">
          <t>As a reminder, a Classifier is a function that is responsible for
          classifying traffic based on (pre-defined) rules.</t>

          <t>This interface is used to install SFC classification rules in
          Classifiers. Once classification rules are populated, SFC
          Classifiers are responsible for binding incoming traffic to service
          function chains according to these classification rules. Note, the
          SFC control plane must not make any assumption on how the traffic is
          to be bound to a given SFC. In other words, classification rules are
          deployment-specific. For instance, classification can rely on a
          subset of the information carried in a received packet such as
          5-tuple classification, be subscriber-aware, be driven by traffic
          engineering considerations, or any combination thereof.</t>

          <t>The SFC control plane should be responsible for removing invalid
          (and stale) mappings from the classification tables maintained by
          the classifiers. Also, local sanity checks mechanisms may be
          supported locally by the Classifiers, but those are out of
          scope.</t>

          <t>The Classifier may be notified by the control plane about the
          available SFs (including their locators) or be part of the service
          function discovery procedure.</t>

          <t>Classification rules may be updated, deleted or disabled by the
          control plane. Criteria that would trigger those operations are
          deployment-specific.</t>

          <t>Given that service function chaining solutions may be applied to
          very large sets of traffic, any control solution should take scaling
          issues into consideration as part of the design.</t>

          <t>Below are listed some functional objectives for this interface:
          <?rfc subcompact="yes" ?><list style="symbols">
              <t>Rationalize the management of classification rules.</t>

              <t>Maintain a global view of instantiated rules in all
              Classifiers in an SFC-enabled domain.</t>

              <t>Check the consistency of instantiated classification rules
              within the same Classifier or among multiple Classifier.</t>

              <t>Assess the impact of removing or modifying a classification
              entry on packets entering an SFC-enabled domain.</t>

              <t>Aggregate classification rules for the sake of performance
              optimization (mainly reduce lookup delays).</t>

              <t>Adjust classification rules when rules are based on volatile
              identifiers (e.g., an IPv4 address, IPv6 prefix).</t>

              <t>Allow to rapidly restore SFC states during failure events
              that occurred at a Classifier (or a Control Element).<?rfc subcompact="no" ?></t>
            </list></t>

          <t>The control plane must instruct the Classifier whether it can
          trust an existing SFC marking of an incoming packet or whether it
          must be ignored.</t>

          <t>For bidirectional packet processing purposes (e.g., full or
          partial path symmetry), the control plane invokes this interface to
          configure the appropriate classification entries.</t>

          <t>A Classifier can send unsolicited messages through this interface
          to notify the SFC Control & Management Planes about specific
          events.</t>

          <t>When re-classification is allowed in an SFC-enabled domain, this
          interface can be used to control Classifiers co-resident with
          SFC-aware SFs, SFC Proxies, or SFFs to manage re-classification
          rules .</t>

          <t>SFC Classification policy entry should be bound to one single
          service function chain (or one single SFP); when an incoming packet
          matches more than one classification entry, tie-breaking criteria
          should be specified (e.g., priority). Such tie-breaking criteria
          should be instructed by the control plane.</t>

          <t>The identification of instantiated SFCs/SFPs is local to each
          administrative domain; it is policy-based and
          deployment-specific.</t>
        </section>

        <section anchor="c2"
                 title="C2: Interface between SFC Control Plane & SFF">
          <t>SFFs make traffic forwarding decisions according to the entries
          maintained in their SFC forwarding policy table. Such table is
          populated by the SFC control plane through the C2 interface.</t>

          <t>This interface is used to instruct a SFF about the SFC-aware SFs
          that it can service. This interface is also used by the SFF to
          report the connectivity to their attached (including embedded) SFs.
          Local means may be enabled between the SFC-aware SFs and SFFs to
          allow for the dynamic attachment of SFs to a SFF and/or discovery of
          SFs by a SFF but those means are unspecified in this document.</t>

          <t>The C2 interface is also used for collecting states of attributes
          (e.g., availability, workload, latency), for example, to dynamically
          adjust Service Function Paths.</t>
        </section>

        <section anchor="c3"
                 title="C3: Interface between SFC Control Plane & SFC-aware SFs">
          <t>The SFC control plane uses this interface to interact with
          SFC-aware SFs.</t>

          <t>SFs may need to output some processing results of packets to the
          SFC control plane. This information can be used by the SFC control
          plane to update the SFC classification rules and the SFC forwarding
          policy table entries.</t>

          <t>This Interface is used to collect such kind of feedback
          information from SFs. For example, the following information can be
          exchanged between a SF and the SFC control plane: <list
              style="symbols">
              <t>SF execution status: Some SFs may need to send information to
              the control plane to fine tune SFPs. For example, a
              threat-detecting SF can periodically send the threat
              characteristics via this interface, such as high probability of
              threat with packet of a given size. The control plane can then
              add an appropriate matching criteria to SFF to steer traffic to
              a scrubbing center.</t>

              <t>SF load update: When SFs are under stress that yielded the
              crossing of some performance thresholds, the SFC control plane
              needs to be notified to adjust SFPs accordingly (especially when
              the centralized path computation mode is enabled). It is out of
              scope of this document to specify the exact methods to monitor
              the performance threshold or stress level of SFs, nevertheless
              the SFC control plane can invoke those methods for its
              operations.</t>
            </list></t>

          <t>The SFC control needs the above status information for various
          tasks it undertakes, but this information may be acquired directly
          from SFs or indirectly from other management and control systems in
          the operational environment.</t>

          <t>This interface is also used to instruct an SFC-aware SF about any
          context information it needs to supply in the context of a given
          SFC.</t>

          <t>Also, this interface informs the SFC-aware SF about the semantics
          of a context information, which would otherwise have opaque meaning.
          Several attributes may be associated with a context information such
          as (but not limited to) the "scope" (e.g., per-packet, per-flow or
          per host), whether it is "mandatory" or "optional" to process flows
          bound to a given chain, etc. Note that a context may be mandatory
          for "chain 1", but optional for "chain 2".</t>

          <t>The control plane may indicate, for a given service function
          chain, an order for consuming a set of contexts supplied in a
          packet.</t>

          <t>A SFC-aware SF can also be instructed about the behavior is
          should adopt after consuming a context information that was supplied
          in the SFC header. For example, the context can be maintained or
          stripped. The SFC-aware SF can be instructed to inject a new context
          header into the SFC header.</t>

          <t>Multiple SFs may be located within the same physical node, and no
          SFF is enabled in that same node, means to unambiguously forward the
          traffic to the appropriate SF must be supported.</t>

          <t>An SF can be instructed to strip the SFC information for the
          chains it terminates.</t>
        </section>

        <section anchor="c4"
                 title="C4: Interface between SFC Control Plane & SFC Proxy">
          <t>The SFC control plane uses this interface to interact with an SFC
          Proxy.</t>

          <t>The SFC proxy can be instructed about authorized SFC-unware SFs
          it can service. A SFC Proxy can be instructed about the behavior it
          should adopt to process the context information that was supplied in
          the SFC header on behalf of a SFC-unware SF, e.g., the context can
          be maintained or stripped.</t>

          <t>The SFC proxy is also instructed about the semantics of a context
          information, which would otherwise have opaque meaning. Several
          attributes may be associated with a context information such as (but
          not limited to) the "scope" (e.g., per-packet, per-flow or per
          host), whether it is "mandatory" or "optional" to process flows
          bound to a given chain, etc.</t>

          <t>The SFC Proxy can also be instructed to add SF some new context
          information into the SFC header on behalf of a SFC-unaware SF.</t>

          <t>The C4 interface is also used for collecting attribute states
          (e.g., availability, workload, latency), for example, to dynamically
          adjust Service Function Paths.</t>
        </section>
      </section>
    </section>

    <section title="Additional Considerations">
      <t></t>

      <section title="Discovery of the SFC Control Element">
        <t>SFC data plane functional elements need to be provisioned with the
        locators of the Control Elements. This can be achieved using a variety
        if mechanisms such as static configuration or the activation of a
        service discovery mechanism. The exact specification of how this
        provisioning is achieved is out of scope.</t>
      </section>

      <section title="SF Symmetry">
        <t>Some SFs require both directions of a flow to traverse. Some
        service function chains require full symmetry. If a SF (e.g., stateful
        firewall or NAT) needs both direction of a flow, it is the SF
        instantiation that needs both direction of a flow to traverse, not the
        abstract SF (which can have many instantiations spread across the
        network).</t>
      </section>

      <section title="Pre-deploying SFCs">
        <t>Enabling service function chains should preserve some deployment
        practices adopted by Operators. Particularly, installing a service
        function chain (and its associated SFPs) should allow for
        pre-deployment testing and validation purposes (that is a restricted
        and controlled usage of such service function chain (and associated
        SFPs)).</t>
      </section>

      <section title="Withraw a Service Function (SF)">
        <t>During the lifetime of a SFC, a given SF can be decommissioned. To
        accommodate such context and any other case where a SF is to be
        withdrawn, the control plane should instruct the SFC data plane
        functional element about the behavior to adopt. Particularly: <list
            style="numbers">
            <t>a first approach would be to update the service function chains
            (and associated SFPs) where that SF is present by removing any
            reference to that SF. Doing so avoids to induce service failures
            for end users.</t>

            <t>a second approach would be to delete/deactivate any service
            function chain (and its associated SFPs) that involves that SF but
            install new service function chains.</t>
          </list></t>
      </section>

      <section title="SFC/SFP Operations">
        <t>Various actions can be executed on a service function chain (and
        associated SFPs) that is structured by the SFC Control &
        Management planes. Indeed, a service function chain (and associated
        SFPs) can be enabled, disabled, its structure modified by adding a new
        SF hop or remove an SF from the sequence of SFs to be invoked, its
        classification rules modified, etc.</t>

        <t>A modification of a service function chain can trigger control
        messages with the appropriate SFC-aware nodes accordingly.</t>
      </section>

      <section title="Unsolicited (Notification) Messages">
        <t>Involved SFC data plane functional element must be instructed to
        send unsolicited notifications when loops are detected, a problem in
        the structure of a service function chain is encountered, a long
        unavailable forwarding path time is observed, etc.</t>

        <t>Specific criteria to send unsolicited notifications to a Control
        Element should be fine tuned by the control plane using the interface
        defined in <xref target="refint"></xref>.</t>
      </section>

      <section anchor="LiveD" title="SF Liveness Detection">
        <t>The control plane must allow to detect the liveliness of SFs of an
        SFC-enabled domain. In particular, it must allow to dynamically detect
        that a SF instance is out of service and notify the relevant Control
        Element elements accordingly. The liveness information may be acquired
        directly from SFs or indirectly from other management and control
        systems in the operational environment.</t>

        <t>Liveness status records for all SF instances, and service function
        chains (including the SFPs bound to a given chain) are maintained by
        the SFC Control & Management.</t>

        <t>The Classifier may be notified by the control plane or be part of
        the liveness detection procedure.</t>

        <t>The ability of a SFC Control Element to check the liveness of each
        SF present in service function chain has several advantages,
        including:<?rfc subcompact="yes" ?><list style="symbols">
            <t>Enhanced status reporting by the control & management
            planes (i.e., an operational status for any given service chain
            derived from liveness state of its SFs).</t>

            <t>Ability to support various resiliency policies (i.e., bypass a
            node embedding an SF, use alternate node, use alternate chain,
            drop traffic, etc.) .</t>

            <t>Ability to support load balancing capabilities to solicit
            multiple SF instances that provide equivalent functions.<?rfc subcompact="no" ?></t>
          </list></t>

        <t>Because a node embedding a SF can be responsive from a reachability
        standpoint (e.g., IP level) while the function its provides may be
        broken (e.g., a NAT module may be down), additional means to assess
        whether an SF is up and running are required. These means may be
        service-specific.</t>
      </section>

      <section title="Monitoring & Counters">
        <t>SFC-specific counters and statistics must be provided using the
        interfaces defined in <xref target="refint"></xref>. These data
        include (but not limited to): <?rfc subcompact="yes" ?><list
            style="symbols">
            <t>Number of flows ever and currently assigned to a given service
            function chain and a given SFP.</t>

            <t>Number of flows, packets, bytes dropped due to policy.</t>

            <t>Number of packets and bytes in/out per service function
            chain.</t>

            <t>Number of flows, packets, bytes dropped due to unknown service
            function chain (this is valid in particular for a SF node).<?rfc subcompact="no" ?></t>
          </list></t>
      </section>

      <!--      <section title="SFC/SFP Diagnosis">
        <t><list style="empty">
            <t>[[Note: This section is expected to be removed once the working
            group adopts a document on OAM.]]</t>
          </list>The Control & Management planes should allow for the
        following:<?rfc subcompact="yes" ?></t>

        <t><list hangIndent="11" style="symbols">
            <t>Assess the status of the serviceability of a SF (i.e., the SF
            provides the service(s) it is configured for). Obviously, this
            assessment must not rely only on IP reachability to decide whether
            a SF is up and running.</t>

            <t>Diagnose the availability of a SFC (including the availability
            of a particular SFP bound to a given SFC).</t>

            <t>Retrieve the set of service function chains that are enabled
            within a domain.</t>

            <t>Assess whether an SFC-enabled domain is appropriately
            configured (including, check the configured chains are matching
            what should be configured in that domain, and ensure coherent
            classification rules are installed in and enforced by all the
            Classifiers of the SFC-enabled domain).</t>

            <t>Correlate classification policies with observed forwarding
            actions (including, assess the output of the classification rule
            applied on a packet presented to a Classifier of an SFC-enabled
            domain).</t>

            <t>Support the correlation between a service function chain and
            the actual forwarding path followed by a packet matching that
            service function chain.</t>

            <t>Notify the SFC Control Element whenever some (critical) events
            occur (for example, a malfunctioning SF instance).</t>

            <t>Re-use SF built-in diagnostic procedures specific to each
            SF.<?rfc subcompact="no" ?></t>
          </list></t>

        <t>The SFC control plane must be able to invoke SFC OAM mechanisms,
        and to determine the results of OAM operations.</t>
      </section>

-->

      <section title="Validity Lifetime">
        <t>SFC instructions communicated via the various interfaces introduced
        in <xref target="refint"></xref> may be associated with validity
        lifetimes, in which case classification entries will be automatically
        removed upon the expiry of the validity lifetime without requiring an
        explicit action from a Control Element.</t>

        <t>Lifetimes are used in particular by an SFC data plane element to
        clear invalid control entries that would be maintained in the system
        if, for some reason, no appropriate action was undertaken by the
        control plane to clear such entries.</t>

        <t>Both short and long lifetimes may be assigned.</t>
      </section>

      <section title="Considerations Specific to the Centralized Path Computation Model">
        <t>This section focuses on issues that are specific to the centralized
        deployment model (<xref target="cd"></xref>).</t>

        <section title="Service Function Path Adjustment">
          <t>A SFP is determined by composing SF instances and overlay links
          among SFFs. Thus, the status of a SFP depends on the states or
          attributes (e.g., availability, topological location, latency,
          workload, etc.) of its components. For example, failure of a single
          SF instance results in failure of the whole SFP. Since these states
          or attributes of SFP components may vary in time, their changes
          should monitored and SFPs should be dynamically adjusted.</t>

          <t>Examples of use cases for SFP adjustment are listed below:<?rfc subcompact="yes" ?></t>

          <t><list style="hanging">
              <t hangText="SFP fail-over: ">re-construct a SFP with replacing
              the failed SF instance with another instance of the same SF.</t>

              <t hangText="SFP with better latency experience:">re-construct a
              SFP with a low path stretch considering the changes in
              topological locations of SF instances and the latency induced by
              the (overlay) connectivity among SFFs.</t>

              <t hangText="Traffic engineered SFC:">re-construct SFPs to
              localize the traffic in the network considering various TE goals
              such as bypass a node, bypass a link, etc. These techniques may
              be used for planned maintenance operations on a SFC-enabled
              domain.</t>

              <t hangText="SF/SFC Load balancing: ">re-construct SFPs to
              distribute the workload among various SF instances.<?rfc subcompact="no" ?></t>
            </list></t>

          <t>For more details about the use cases, refer to <xref
          target="I-D.lee-nfvrg-resource-management-service-chain"></xref>.</t>

          <t>The procedures for SFP adjustment may be handled by the SFC
          control plane as follows:<?rfc subcompact="yes"?></t>

          <t><list style="symbols">
              <t>Collect and monitor states and attributes of SF instances and
              overlay links via the C2 interface (<xref target="c2"></xref>)
              and the C3 interface (<xref target="c3"></xref>).</t>

              <t>Evaluate SF instances and overlay links based on the
              monitoring results.</t>

              <t>Select SF instances to re-determine a SFP according to the
              evaluation results.</t>

              <t>Replace target SF instances (e.g., in a failure or overladed)
              with newly selected ones.</t>

              <t>Enforce the updated SFP for upcoming SFC traversal to SFFs
              via the C1 interface (<xref target="c1"></xref>) or the C2
              interface (<xref target="c2"></xref>).<?rfc subcompact="no"?></t>
            </list></t>
        </section>

        <section title="Head End Initiated SFP Establishment">
          <t>In some scenarios where a SFC Control Element is not connected to
          all SFFs in a SFC-enabled domain, the SFC control plane can send the
          explicit SFF-SF-sequence or SF-sequence to the SFC head-end, e.g.,
          the SFC Classifier via the C1 interface (<xref target="c1"></xref>).
          SFC head-end can use a signaling protocol to establish the
          SFF-SF-sequence based on the SF-sequence.</t>
        </section>

        <section title="(Regional) Restoration of Service Functions">
          <t>There are situations that it might not be feasible for the
          Classifier to be notified of the changes of SFF-sequence or
          SFF-SF-Sequence for a given SFP because of the time taken for the
          notification and the limited capability of the Classifiers.</t>

          <t>If a SF has a large number of instantiations, it scales better if
          the Classifier doesn't need to be notified with status of visible
          instantiations of SFs on a SFP.</t>

          <t>It might not be always feasible for the Classifier to be aware of
          the exact SF instances selected for a given SFP due to too many
          instances for each SF, notifications not being promptly sent to the
          Classifier, or other reasons. This is about multiple instances of
          the same SF attached to one SFF node; those instances can be handled
          by the SFF via local load balancing schemes.</t>

          <t>Regional restoration can take the similar approach as the global
          restoration: choosing a regional ingress node that can take over the
          responsibility of installing the new steering policies to the
          involved SFFs or network nodes. Typically, the regional ingress node
          should be:<?rfc subcompact="yes" ?><list style="symbols">
              <t>on the data path of the flow of the given SFC;</t>

              <t>in front of the relevant SFFs or network nodes that are
              impacted by the change of the SFP;</t>

              <t>capable of encoding the detailed SFP to the Service Chain
              Header of data packets of the identified flow; and</t>

              <t>capable of removing the detailed SFP encoding in data packets
              after all the impacted SFFs and network nodes completed the
              policy installation. <?rfc subcompact="no" ?></t>
            </list></t>
        </section>
      </section>
    </section>

    <section title="Security Considerations">
      <t></t>

      <section title="Secure Communications">
        <t>The SFC Control Elements and the participating SFC data plane
        elements must mutually authenticate. SFC data plane elements must
        ignore instructions received from unauthenticated SFC Control
        Elements. The credentials details used during authentication can be
        used by the SFC control plane to decide whether specific authorization
        may be granted to a Service Function with regards to some specific
        operations (e.g., authorize a given SF to access specific context
        information).</t>

        <t>In case multiple SFC data plane elements are embedded in the same
        node, the authentication mechanism may be executed as a whole; not for
        each instance.</t>

        <t>A SFC data plane element must be able to send authenticated
        unsolicited notifications to a SFC Control Element.</t>

        <t>The communication between a Control Element and SFC data plane
        elements must provide integrity and replay protection.</t>

        <t>An SFC Control Element may instruct a Service Function to include
        specific security token(s) that may be used to decrypt traffic
        upstream. The security token may be supplied by the SFC control plane
        or by an authorized Service Function (e.g., TLS proxy). The exact
        details on how authorization is granted to a specific SF, including
        via a control plane interface, should be specified.</t>

        <t>A Service Function must by default discard any action from a SFC
        Control Element that requires specific right privileges (e.g., access
        to a legal intercept log, mirror the traffic, etc.).</t>
      </section>

      <section title="Pervasive Monitoring">
        <t>The authentication mechanism should be immune to pervasive
        monitoring <xref target="RFC7258"></xref>. An attacker can intercept
        traffic by installing classification rules that would lead to redirect
        all or part of the traffic to an illegitimate network node. Means to
        protect against attacks that would lead to install, remove, or modify
        classification rules must be supported.</t>
      </section>

      <section title="Privacy">
        <t>The SFC control plane must be able to instruct SFC data plane
        elements about the information to be leaked outside an SFC-enabled
        domain. Particularly, the SFC control plane must support means to
        preserve privacy <xref target="RFC6973"></xref>. Context headers may
        indeed reveal privacy information (e.g., IMSI, user name, user
        profile, location, etc.). Those headers must not be exposed outside
        the operator's domain.</t>
      </section>

      <section title="Denial-of-Service (DoS)">
        <t>In order to protect against denial of service that would be caused
        by a misbehaving trusted SFC Control Element, SFC data plane elements
        should rate limit the messages received from an SFC Control
        Element.</t>
      </section>

      <section title="Illegitimate Discovery of SFs and SFC Control Elements">
        <t>Means to defend against soliciting illegitimate SFs/SFFs that do
        not belong to the SFC-enabled domain must be enabled. Such means must
        be defined in service function discovery and SFC Control Element
        discovery specification documents.</t>
      </section>
    </section>

    <section anchor="IANA" title="IANA Considerations">
      <t>This document does not require any IANA actions.</t>
    </section>
  </middle>

  <back>
    <references title="Normative References">
      <?rfc include='reference.I-D.ietf-sfc-architecture'?>
    </references>

    <references title="Informative References">
      <?rfc include='reference.RFC.7498'?>

      <?rfc include='reference.I-D.lee-nfvrg-resource-management-service-chain'?>

      <?rfc include='reference.I-D.lee-sfc-dynamic-instantiation'?>

      <?rfc include='reference.I-D.ietf-sfc-use-case-mobility'?>

      <?rfc include='reference.I-D.ietf-opsawg-firewalls'?>

      <?rfc include='reference.I-D.ietf-sfc-dc-use-cases'?>

      <?rfc include='reference.RFC.3135'?>

      <?rfc include='reference.RFC.3022'?>

      <?rfc include='reference.RFC.6333'?>

      <?rfc include='reference.RFC.6146'?>

      <?rfc include='reference.RFC.7258'?>

      <?rfc include='reference.RFC.6973'?>
    </references>

    <section title="RSP-related Considerations">
      <t>NOTE: This section records some contributions proposed by L. Dunbar
      and A. Malis, but have not been discussed yet among authors.</t>

      <section title="Encoding the Exact SFF-SF-sequence in Data Packets">
        <t>Encoding the exact RSP in every packet has the benefit and the
        issues associated with source routing. This approach may not be
        optimal when the SFP doesn't change very frequently, as in minutes or
        hours.</t>

        <t>There are contexts that it might not be feasible for the head end
        Classifier to be notified of the changes of SFF-sequence or
        SFF-SF-Sequence for a given SFP because of the time taken for the
        notification and the limited capability of the Classifier nodes.</t>
      </section>

      <section title="Fully Controlled SFF-SF-Sequence for a SFP">
        <t>This section describes the information that can be exchanged over
        C2 interface (<xref target="c2"></xref>) when the SFC Control Element
        explicitly passes the steering policies to all SFFs for the
        SFF-SF-Sequence of a given SFC. In this model, each SFF doesn't need
        to signal other SFFs for the SFP.</t>

        <t>Suppose the SFC ID for this SFP is “yellow”, an example
        of policy to “sff-a” is depicted in <xref
        target="rsp"></xref> (for illustration proposes)</t>

        <t><figure anchor="rsp"
            title="Example of Traffic Steering Policy to a SFF node">
            <artwork><![CDATA[           Matching                     |       Action
----------------------------------------+-------------------------
SFC ID = "yellow" & ingress = sffx-port | next-hop: "sf2" & VID
SFC ID = "yellow" & ingress = sf2-port  | next-hop: "sf3" & VID
SFC ID = "yellow" & ingress = sf3-port  | next-hop: sff-b ]]></artwork>
          </figure></t>

        <t>The SFF nodes may not be directly adjacent to each other. They can
        be interconnected by tunnels, such as GRE, VxLAN, etc. SFs are
        attached to a SFF node or SFC Proxy node via Ethernet link or other
        link types. Therefore, the steering policies to a SFF node for service
        function chain depends on if the packet comes from previous SFF or
        comes from a specific SF, i.e., the SFC Forwarding Policy Table
        entries have to be ingress port specific. There are multiple different
        steering policies for one flow within one SFF and each set of steering
        policies is specific for an ingress port.</t>

        <t>The semantics of traffic steering rules can be "Match" and
        "Action", similar to the "route" described in
        [I-D.ietf-i2rs-rib-info-model]. The “match” and
        “action” for distinct ports can be different. The matching
        criteria for SFF can be more sophisticated. For example, the matching
        criteria could be any fields in the data packets: <?rfc subcompact="yes"?><list
            style="symbols">
            <t>Ingress port</t>

            <t>Destination MAC address</t>

            <t>Source MAC address</t>

            <t>VLAN_id,</t>

            <t>Destination IP address</t>

            <t>Source IP address</t>

            <t>Source port number</t>

            <t>Destination port number</t>

            <t>DSCP</t>

            <t>Packet size, etc., or any combination thereof.</t>
          </list></t>

        <t><?rfc subcompact="no"?>A SFF node may not support some of the
        matching criteria listed above. It is important that SFC control plane
        can retrieve the supported matching criteria by SFF nodes. The
        “Actions” for traffic steering could be to steer traffic
        to the attached service function or SF instantiations via a specific
        port.</t>

        <t>The “Actions” to SFC Proxy may include a method to map
        the SFC Identifier carried in the packet header to a locally
        significant link identifier, e.g., VLAN-ID, and a method to construct
        and encapsulate the SFC header back to the packets when they come back
        from the attached SFs.</t>

        <t>This approach does not require using an end-to-end signaling
        protocol among Classier nodes and SFF nodes. However, there may be
        problems encountered if SFF nodes are not updated in the proper order
        or not at the same time. For example, if the SFF “A” and
        SFF “C” get flow steering policies at slightly different
        times, some packets might not be directed to some service functions on
        a chain.</t>
      </section>
    </section>

    <section numbered="no" title="Acknowledgments">
      <t>This document is the result of merging with <xref
      target="I-D.lee-sfc-dynamic-instantiation" />.</t>

      <t>The authors would like to thank Shibi Huang and LAC Chidung for the
      review and comments.</t>

      <t>The text about the semantic of a context information is provided by
      Dave Dolson.</t>
    </section>

    <section numbered="no" title="Contributors">
      <t>Many thanks to the following people for contributing text to the
      document: <?rfc subcompact="yes" ?> <list style="symbols">
          <t>tbc</t>
        </list></t>

      <?rfc subcompact="no" ?>
    </section>
  </back>
</rfc>

PAFTECH AB 2003-20262026-04-23 16:22:43