Network Working Group                                      A. Atlas, Ed.
Internet-Draft                                                 T. Nadeau
Intended status: Informational                          Juniper Networks
Expires: January 31, 2013                                        D. Ward
                                                           Cisco Systems
                                                           July 30, 2012

           Interface to the Routing System Problem Statement


   As modern networks grow in scale and complexity, the need for rapid
   and dynamic control increases.  With scale, the need to automate even
   the simplest operations is important, but even more critical is the
   ability to quickly interact with more complex operations such as
   policy-based controls.

   In order to enable applications to have access to and control over
   information in the Internet's routing system, we need a publically
   documented interface specification.  The interface needs to support
   real-time, transaction-based interactions using efficient data models
   and encodings.  Furthermore, the interface must support a variety of
   use cases including those where verified control feed-back loops are

   This document expands upon these statements of requirements to
   provide a problem statement for an interface to the Internet routing

Status of this Memo

   This Internet-Draft is submitted in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-
   Drafts is at

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   This Internet-Draft will expire on January 31, 2013.

Atlas, et al.           Expires January 31, 2013                [Page 1]

Internet-Draft            IRS Problem Statement                July 2012

Copyright Notice

   Copyright (c) 2012 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   ( in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
   2.  IRS Model and Problem Area for The IETF  . . . . . . . . . . .  3
   3.  Standard Data-Models of Routing State for Installation . . . .  5
   4.  Learning Router Information  . . . . . . . . . . . . . . . . .  5
   5.  Desired Aspects of a Protocol for IRS  . . . . . . . . . . . .  6
   6.  Existing Management Interfaces . . . . . . . . . . . . . . . .  7
   7.  Acknowledgements . . . . . . . . . . . . . . . . . . . . . . .  8
   8.  IANA Considerations  . . . . . . . . . . . . . . . . . . . . .  8
   9.  Security Considerations  . . . . . . . . . . . . . . . . . . .  8
   10. Informative References . . . . . . . . . . . . . . . . . . . .  9
   Appendix A.  Gaps and Concerns for SNMP  . . . . . . . . . . . . .  9
   Appendix B.  Gaps and Concerns with NetConf  . . . . . . . . . . . 10
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 11

Atlas, et al.           Expires January 31, 2013                [Page 2]

Internet-Draft            IRS Problem Statement                July 2012

1.  Introduction

   As modern networks grow in scale and complexity, the need for rapid
   and dynamic control increases.  With scale, the need to automate even
   the simplest operations is important, but even more critical is the
   ability to quickly interact with more complex operations such as
   policy-based controls.

   With complexity comes the need for more sophisticated automated
   applications and orchestration software that can process large
   quantities of data, run complex algorithms, and adjust the routing
   state as required in order to support the applications, their
   calculations and their policies.  Changes made to the routing state
   of a network by external applications must be verifiable by those
   applications to ensure that the correct state has been installed in
   the right places.

   Mechanisms to support the requirements outlined above have been
   developed piecemeal as proprietary solutions to specific situations
   and needs.  A standard protocol, clear operations that an application
   can initiate with that protocol, and data-models to support such
   actions would facilitate wide-scale deployment of interoperable
   applications and routing systems.  That a protocol designed to
   facilitate rapid, isolated, secure, and dynamic routing changes is
   needed motivates the creation of an Interface to The Routing System

2.  IRS Model and Problem Area for The IETF

   Managing a network of deployed devices running a variety of routing
   protocols involves interactions among multiple different functions
   and components that exist within the network.  Some of these
   components are virtual while some are physical; all should be made
   available to be managed and manipulated by applications, given that
   appropriate access, authentication, and policy hurdles have been
   crossed.  The management of only some of these components requires
   standardization, as others have already been standardized.  The IRS
   model is intended to incorporate existing mechanisms where
   appropriate, and to build extensions and new protocols where needed.
   The IRS model and problem area proposed for IETF work is illustrated
   in Figure 1.

Atlas, et al.           Expires January 31, 2013                [Page 3]

Internet-Draft            IRS Problem Statement                July 2012

        +***************+   +***************+   +***************+
        *  Application  *   *  Application  *   *  Application  *
        +***************+   +***************+   +***************+
                 ^                   ^                  ^
                 *                   *                  *
                 *                   *   ****************
                 *                   *   *
                 v                   v   v
         +---------------+   +---------------+
         |  IRS Client   |   |  IRS Client   |
         +---------------+   +---------------+
                 ^                   ^
                 |________________   |
                                  |  |  <== IRS Protocol
                                  |  |
       .                          v  v                                 .
       . +*************+     +---------------+      +****************+ .
       . *    Policy   *     |               |      *   Routing  &   * .
       . *   Database  *<***>|  IRS Agent    |<****>*   Signaling    * .
       . +*************+     |               |      *   Protocols    * .
       .                     +---------------+      +****************+ .
       .                        ^   ^     ^                  ^         .
       . +*************+        *   *     *                  *         .
       . *  Topology   *        *   *     *                  *         .
       . *  Database   *<*******+   *     *                  v         .
       . +*************+            *     *         +****************+ .
       .                            *     +********>*  RIB Manager   * .
       .                            *               +****************+ .
       .                            *                        ^         .
       .                            v                        *         .
       .                 +*******************+               *         .
       .                 * Subscription &    *               *         .
       .                 * Configuration     *               v         .
       .                 * Templates for     *      +****************+ .
       .                 * Measurements,     *      *  FIB Manager   * .
       .                 * Events, QoS, etc. *      *  & Data Plane  * .
       .                 +*******************+      +****************+ .

     <-->  interfaces inside the scope of IRS
     +--+  objects inside the scope of IRS

     <**>  interfaces NOT within the scope of IRS
     +**+  objects NOT within the scope of IRS

     ....  boundary of a router participating in the IRS

Atlas, et al.           Expires January 31, 2013                [Page 4]

Internet-Draft            IRS Problem Statement                July 2012

                   Figure 1: IRS model and Problem Area

   A critical aspect of IRS is defining a suitable protocol or protocols
   to carry messages between the IRS Clients and the IRS Agent, and
   defining the encapsulation of data within those messages.  This
   should provide a clear transfer syntax that is straightforward for
   applications to use (e.g., a Web Services design paradigm), and
   should provide the key features specified in Section 5.

   The second critical aspect is semantic-aware data-models for
   information in the routing system and in a topology database.  The
   data-models should be separable across different features of the
   managed components, versioned, and combine to provide a network data-

3.  Standard Data-Models of Routing State for Installation

   There is a need to be able to precisely control routing and signaling
   state based upon policy or external measures.  This can range from
   simple static routes to policy-based routing to static multicast
   replication and routing state.  This means that the data model
   employed needs to handle indirection as well as different types of
   tunneling and encapsulation.  The relevant MIB modules (for example
   [RFC4292]) lack the necessary generality and flexibility.  In
   addition, by having IRS focus initially on interfaces to the RIB
   layer (e.g.  RIB, LFIB, multicast RIB, policy-based routing), the
   ability to use routing indirection allows flexibility and
   functionality that can't be as easily obtained at the forwarding

   Efforts to provide this level of control have focused on
   standardizing data models that describe the forwarding plane (e.g.
   ForCES [RFC3746]).  IRS recognizes that the routing system and a
   router's OS provide useful mechanisms that applications could
   usefully harness to accomplish application-level goals.

   In addition to interfaces to the RIB layer, there is a need to
   configure the various routing and signaling protocols with differing
   dynamic state based upon application-level policy decisions.  The
   range desired is not available via MIBs at the present time.

4.  Learning Router Information

   A router has information that applications may require so that they
   can understand the network, verify that programmed state is installed
   in the forwarding plane, measure the behavior of various flows, and

Atlas, et al.           Expires January 31, 2013                [Page 5]

Internet-Draft            IRS Problem Statement                July 2012

   understand the existing configuration and state of the router.  IRS
   provides a framework for applications to register for asynchronous
   notifications and for them to make specific requests for information.

   Although there are efforts to extend the topological information
   available, even the best of these (e.g., BGP-LS
   [I-D.gredler-idr-ls-distribution]) still provides only the current
   active state as seen at the IGP layer and above.  Detailed
   topological state that provides more information than the current
   functional status is needed by applications; only the active paths or
   links are known versus those desired or unknown to the routing

   For applications to have a feedback loop that includes awareness of
   the relevant traffic, an application must be able to request the
   measurement and timely, scalable reporting of data.  While a
   mechanism such as IPFIX [RFC5470] may be the facilitator for
   delivering the data, the need for an application to be able to
   dynamically request that measurements be taken and data delivered is

   There are a wide range of events that applications could use for
   either verification of router state before other network state is
   changed (e.g. that a route has been installed), to act upon changes
   to relevant routes by others, or upon router events (e.g. link up/
   down).  While a few of these (e.g. link up/down) may be available via
   MIB Notifications today, the full range is not - nor is there the
   ability to set up the router to trigger different actions upon an
   event's occurrence.

5.  Desired Aspects of a Protocol for IRS

   This section describes required aspects of a protocol that could
   support IRS.  Whether such a protocol is built upon extending
   existing mechanisms or requires a new mechanism requires further

   The key aspects needed in an interface to the routing system are:

   Multiple Simultaneous Asynchronous Operations:   A single application
      should be able to send multiple operations to IRS without needing
      to wait for each to complete before sending the next.

   Configuration Not Re-Processed:   When an IRS operation is processed,
      it does not require that any of the configuration be processed.
      I.e., the desired behavior is orthogonal to the static

Atlas, et al.           Expires January 31, 2013                [Page 6]

Internet-Draft            IRS Problem Statement                July 2012

   Duplex:   Communications can be established by either the router or
      the application.  Similarly, events, acknowledgements, failures,
      operations, etc. can be sent at any time by both the router and
      the application.  The IRS is not a pure pull-model where only the
      application queries to pull responses.

   High-Throughput:   At a minimum, the IRS Agent and associated router
      should be able to handle hundreds of simple operations per second.

   Responsive:   It should be possible to complete simple operations
      within a sub-second time-scale.

   Multi-Channel:   It should be possible for information to be
      communicated via the interface from different components in the
      router without requiring going through a single channel.  For
      example, for scaling, some exported data or events may be better
      sent directly from the forwarding plane, while other interactions
      may come from the control-plane.  Thus a single TCP session would
      not be a good match.

   Timing of State Installation and Expiration:   The ability to have
      state installed with different lifetimes and different start-times
      is very valuable.  In particular, the ability of an IRS client to
      request that a pre-sent operation be started based upon a dynamic
      event would provide a powerful functionality.

   To extract information in a scalable fashion that is more easily used
   by applications, the ability to specify filtering constructs in an
   operation requesting data or requesting an asynchronous notification
   is very valuable.

6.  Existing Management Interfaces

   This section discusses the combination of the abstract data models,
   their representation in a data language, and the transfer protocol
   commonly used with them as a single entity.  While other combinations
   are possible, the combinations described are those that have
   significant deployment.

   There are three basic ways that routers are managed.  The most
   popular is the command line interface (CLI), which allows both
   configuration and learning of device state.  This is a proprietary
   interface resembling a UNIX shell that allows for very customized
   control and observation of a device, and, specifically of interest in
   this case, its routing system.  Some form of this interface exists on
   almost every device (virtual or otherwise).  Processing of
   information returned to the CLI (called "screen scraping") is a

Atlas, et al.           Expires January 31, 2013                [Page 7]

Internet-Draft            IRS Problem Statement                July 2012

   burdensome activity because the data is normally formatted for use by
   a human operator, and because the layout of the data can vary from
   device to device, and between different software versions.  Despite
   its ubiquity, this interface has never been standardized and is
   unlikely to ever be standardized.  IRS does not involve CLI

   The second most popular interface for interrogation of a device's
   state, statistics, and configuration is The Simple Network Management
   Protocol (SNMP) and a set of relevant standards-based and proprietary
   Management Information Base (MIB) modules.  SNMP has a strong history
   of being used by network managers to gather statistical and state
   information about devices, including their routing systems.  However,
   SNMP is very rarely used to configure a device or any of its systems
   for reasons that vary depending upon the network operator.  Some
   example reasons include complexity, the lack of desired configuration
   semantics (e.g., configuration "roll-back", "sandboxing" or
   configuration versioning), and the difficulty of using the semantics
   (or lack thereof) as defined in the MIB modules to configure device
   features.  Therefore, SNMP is not considered as a candidate solution
   for the problems motivating IRS.

   Finally, the IETF's Network Configuration (or NetConf) protocol has
   made many strides at overcoming most of the limitations around
   configuration that were just described.  However, the lack of
   standard data models have hampered the adoption of NetConf.

7.  Acknowledgements

   The authors would like to thank Ken Gray for their suggestions and

8.  IANA Considerations

   This document includes no request to IANA.

9.  Security Considerations

   Security is a key aspect of any protocol that allows state
   installation and extracting of detailed router state.  More
   investigation remains to fully define the security requirements, such
   as authorization and authentication levels.

Atlas, et al.           Expires January 31, 2013                [Page 8]

Internet-Draft            IRS Problem Statement                July 2012

10.  Informative References

              Gredler, H., Medved, J., Previdi, S., and A. Farrel,
              "North-Bound Distribution of Link-State and TE Information
              using BGP", draft-gredler-idr-ls-distribution-02 (work in
              progress), July 2012.

   [RFC3746]  Yang, L., Dantu, R., Anderson, T., and R. Gopal,
              "Forwarding and Control Element Separation (ForCES)
              Framework", RFC 3746, April 2004.

   [RFC4292]  Haberman, B., "IP Forwarding Table MIB", RFC 4292,
              April 2006.

   [RFC5470]  Sadasivan, G., Brownlee, N., Claise, B., and J. Quittek,
              "Architecture for IP Flow Information Export", RFC 5470,
              March 2009.

Appendix A.  Gaps and Concerns for SNMP

   Though SNMP can allow state to be written, the overhead of the
   required infrastructure is quite high.  Clients and servers that wish
   to use SNMP must build in and understand a large number of MIB
   modules, including many proprietary modules.  Even when ignoring the
   overhead in building the SNMP processing and handling functions into
   an application, these properties lend themselves well to read-only
   operations.  A critical lack in MIB modules for read-write (i.e., for
   configuration) operations is that there is no semantic understanding
   of the objects defined in the modules encoded in those modules.  Any
   required semantic knowledge must be specifically hand-coded into
   applications or ignored.  Further, many MIB modules do not allow the
   writing of state, and this limits coverage; owing to the cumbersome
   nature, there has not been interest in increasing coverage.

   An additional deficiency in using SNMP MIB modules either for reading
   or writing comes in the inherent co-mingling of semantics and syntax
   in the form of indexing requirements.  SNMP MIB modules contain
   tables that also define an index format.  This format is then
   translated - often mapped onto - a device's actual implementation.
   Such a mapping means an implementation either matches the module's
   indexing during searches or not; if not, then an implementation is
   slowed down when it searches for objects.  Even in not-so-extreme
   cases, such slow performance can result in the SNMP manager's request
   timing-out owing to the delay of the SNMP agent's response.

   For example, if a search of N*M objects is stipulated as (N, M) in

Atlas, et al.           Expires January 31, 2013                [Page 9]

Internet-Draft            IRS Problem Statement                July 2012

   the standard MIB module, but the implementation happens to choose to
   index its tables internally as (M, N), this will result in search
   times of O(N^2).  When N or M become large, as they do in routing
   tables, this results in wasted processing cycles for the device, and
   either extremely long wait times for queries, or simply a lack of
   answers to queries.  It is a clear requirement for the IRS to not
   suffer from this issue.

   In addition to these difficulties, SNMP matches up to the key needed
   aspects as follows:

   Multiple Simultaneous Asynchronous Operations:   Supported, but
      difficult for configuration.

   Configuration Not Re-Processed:   Supported

   Duplex:   The manager must initiate communications with the SNMP
      agent on the router.  With the limited exception of Notifications
      and InformRequests defined in a MIB module, this is a pull model.

   High-Throughput:   Possible

   Responsive:   Possible

   Multi-Channel:   Possible

   The key gaps identified for SNMP are:

   a.  Infrastructure Overhead

   b.  Lack of Semantic Information in the Data-Model

   c.  Required Indexing, from mingling of semantics and syntax, badly
       impacting performance or driving implementation decisions.

   d.  Limited interest and use for configuration

   e.  Communication model isn't naturally duplex.

Appendix B.  Gaps and Concerns with NetConf

   While NetConf has solved many of the deficiencies present in SNMP in
   terms of configuration, it still does not satisfy a number of
   requirements needed to manage today's routing information.  First,
   the lack of standard data models have hampered the adoption of
   NetConf; a significant amount of per-vendor customization is still
   needed.  The transport mechanisms that are currently defined (e.g.,

Atlas, et al.           Expires January 31, 2013               [Page 10]

Internet-Draft            IRS Problem Statement                July 2012

   SOAP/BEEP) for NetConf are not those commonly used by modern
   applications (e.g., ReST or JSON).

   NetConf primarily facilitates configuration rather than reading of
   state or handling asynchronous events.

   NetConf matches up to the key needed aspects as follows:

   Multiple Simultaneous Asynchronous Operations:   Not Possible

   Configuration Not Re-Processed:   Not Possible

   Duplex:   Not Possible - strict pull model.

   High-Throughput:   Unlikely - Can depend on configuration size

   Responsive:   Unlikely - Can depend on configuration size

   Multi-Channel:   Not Possible

Authors' Addresses

   Alia Atlas (editor)
   Juniper Networks
   10 Technology Park Drive
   Westford, MA  01886


   Thomas Nadeau
   Juniper Networks
   1194 N. Mathilda Ave.
   Sunnyvale, CA  94089


Atlas, et al.           Expires January 31, 2013               [Page 11]

Internet-Draft            IRS Problem Statement                July 2012

   Dave Ward
   Cisco Systems
   Tasman Drive
   San Jose, CA  95134


Atlas, et al.           Expires January 31, 2013               [Page 12]