Operational Considerations Regarding Reputation Services
draft-ietf-repute-considerations-01

REPUTE                                                      M. Kucherawy
Internet-Draft                                               May 5, 2013
Intended status: Informational
Expires: November 6, 2013

        Operational Considerations Regarding Reputation Services
                  draft-ietf-repute-considerations-01

Abstract

   The use of reputation systems is has become a common tool in many
   applications that seek to apply collected intelligence about traffic
   sources.  Often this is done because it is common or even expected
   operator practice.  It is therefore important to be aware of a number
   of considerations for both operators and consumers of the data.  This
   document includes a collection of the best advice available regarding
   providers and consumers of reputation data, based on experience to
   date.  Much of this is based on experience with email reputation
   systems, but the concepts are generally applicable.

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 http://datatracker.ietf.org/drafts/current/.

   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 November 6, 2013.

Copyright Notice

   Copyright (c) 2013 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
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect

Kucherawy               Expires November 6, 2013                [Page 1]
Internet-Draft            Reputation Operations                 May 2013

   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.  Background  . . . . . . . . . . . . . . . . . . . . . . . . . . 3
   3.  Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
   4.  Reputation Clients  . . . . . . . . . . . . . . . . . . . . . . 4
   5.  Reputation Service Providers  . . . . . . . . . . . . . . . . . 6
   6.  Security Considerations . . . . . . . . . . . . . . . . . . . . 7
   7.  IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 7
   8.  Informative References  . . . . . . . . . . . . . . . . . . . . 8
   Appendix A.  Acknowledgements . . . . . . . . . . . . . . . . . . . 8

Kucherawy               Expires November 6, 2013                [Page 2]
Internet-Draft            Reputation Operations                 May 2013

1.  Introduction

   Reputation services involve collecting feedback from the community
   about sources of Internet traffic and aggregating that feedback into
   a rating of some kind.  Common examples include feedback about
   traffic associated with specific email addresses, URIs or parts of
   URIs, IP addresses, etc.  The specific collection, analysis, and
   rating methods vary from one service to the next and one problem
   domain to the next, but several operational concepts appear to be
   common to all of these.

   The promise of the protection that reputation services offers can be
   enticing, and many users and operators alike typically engage those
   services merely because it is expected of them.  A critical notion,
   however, is that doing so explicitly involves a third party in the
   flow of data those parties receive.  This is often taken for granted,
   with potentially disastrous results.

   This document highlights this and other considerations in providing
   and consuming reputation data services.

2.  Background

   The community has historically focused on identifying sources that
   misbehave, i.e., that earn negative reputations.  The purpose here is
   to identify and filter traffic from bad actors.  This grew out of
   operational need.  As the Internet grew, so did the occurence of
   problematic traffic, especially in email.  The pragmatics of email
   (i.e., the fact that the total IP address space is more constrained
   than the total email address space) drove the focus on using IP
   addresses as the focus of reputation, in addition to the fact that IP
   addresses have a degree of validation (via the TCP/IP infrastructure)
   where email addresses have had none.

   A specific example of a reputation service in common use in the email
   space is the DNS blacklist [DNSBL].  This is a method of querying a
   database as to whether a source of incoming [SMTP] email traffic
   should be allowed to relay email, based on previous observations and
   feedback.  The method uses the IP address of the source as the basis
   for a query to the database using the Domain Name System [DNS] as the
   interface.  [DNSBL] includes several points in its Security
   Considerations document that are repeated and further developed here.

   However, regardless of the identifier used as the identifier for a
   reputation, bad actors can evade detection or the effects of their
   observed behavior by changing identifiers (e.g., move to a new IP
   address, register a new domain name, use a sub-domain).  This makes
   the problem space effectively boundless, especially as IPv6 rolls

Kucherawy               Expires November 6, 2013                [Page 3]
Internet-Draft            Reputation Operations                 May 2013

   out.

3.  Evolution

   More modern thinking is evolving toward the identification of good
   actors rather than bad actors, and giving them preferential
   treatment.  This drastically reduces the problem space: There are
   vastly more IP addresses and email addresses used by bad actors to
   generate problematic traffic than are used by good actors to generate
   desirable traffic.

   Moreover, good actors tend to be represented by stable names and
   addresses, allowing users to rely on these to identify and give
   preferential treatment to their traffic.  Good actors have no need to
   hop around to different addresses, and already work to keep their
   traffic clean.

   This notion has only been tried to date using manually edited
   whitelists, but has shown promising results on that scale.

4.  Reputation Clients

   Operators that choose to make use of reputation services to influence
   content allowed to pass into or through their infrastructures need to
   understand that they are granting a third party (the reputation
   service provider, or RSP) the ability to affect incoming traffic, for
   better or worse.  Of course, this is the whole point of engaging an
   RSP when everything is working properly, but a number of issues are
   worthy of consideration before establishing such a relationship.

   Some cases have occurred where an RSP made the unilateral decision to
   terminate its service.  To encourage its clients to stop issuing
   queries, it began reporting a maximally negative reputation about all
   subjects, causing rejection of all incoming traffic during the
   incident period.  Although one would hope such incidents to be rare,
   automated means to detect such unfortunate returns (malicious or
   otherwise) and take remedial should be considered.

   RSPs will be the subject of attacks once it is understood that sucess
   in doing so will allow malicious content to evade detection and
   filtering.  Users of RSPs need to be aware of possible interruptions
   in service availability or quality.

   Similarly, some actors will try to "game" the service, which is to
   say that such actors will attempt to determine patterns of behavior
   that result in the reporting of favorable reputations, and in doing
   so, acquire artifically inflated reputations.  One could reasonably
   assume that a reputation service is inherently fragile.  For

Kucherawy               Expires November 6, 2013                [Page 4]
Internet-Draft            Reputation Operations                 May 2013

   operational clients, this should prompt balanced and comparative,
   rather than unilateral, use of the service.

   It is suggested that, when engaging an RSP, an operator should try to
   learn the following things about the RSP in order to understand the
   exposure potential:

   o  the RSP's basis for listing or not listing particular subjects;

   o  if an RSP is paid by its listees, the rate and criteria for
      rejection from being listed;

   o  how the RSP collects data about subjects;

   o  how many data points are input to the reported reputation;

   o  whether reputation is based on a reliable identifier;

   o  how the RSP establishes reliability and authenticity of those
      data;

   o  how data validity is maintained (e.g., on-going monitoring of the
      reported data and sources);

   o  how actively data validity is tracked (e.g., how changes are
      detected);

   o  how disputed reputations are handled;

   o  how often input data expire;

   o  whether older information more or less influential than newer;

   o  whether the reported reputation a scalar, a Boolean value, a
      collection of values, or something else;

   o  when transitioning among RSPs, the differences between them among
      these above points; that is, whether a particular score from one
      means the same thing from another.

   An operator using an RSP would be wise to ensure it has the
   capability to effect local overrides for cases where the client
   expects to disagree with the reported reputation.

   An operator should be able limit the impact of a negative reputation
   on content acceptance.  For example, rather than rejecting content
   outright when a negative reputation is returned, simply subject it to
   additional (i.e., more thorough) local analyis before permitting the

Kucherawy               Expires November 6, 2013                [Page 5]
Internet-Draft            Reputation Operations                 May 2013

   traffic to pass.

   A sensible default should apply when the RSP is not available.  This
   may also be a query to a different RSP known to be less robust than
   the primary one.

   Recent proposals have focused on tailoring operation to prefer or
   emphasize content whose sources have positive reputations.  As stated
   above, negative reputations are easy to shed, and the universe of
   things that will earn and maintain positive reputations is relatively
   small.  Designing a filtering system that observes these notions is
   expected to be more lightweight to operate and harder to game.

   One choice is to query and cross-referencing multiple RSPs.  This can
   help to detect which ones under comparison are reliable, and offsets
   the effect of anomalous replies.

5.  Reputation Service Providers

   Operators intending to provide a reptuation service need to consider
   that there are many flavors of clients.  There will be clients that
   are prepared to make use of a reputation service blindly, while
   others will be interested in understanding more fully the nature of
   the service being provided.  An operator of an RSP should be prepared
   to answer as may of the questions identified in Section 4 as
   possible, not only because wise clients will ask, but also because
   they reflect issues that have arisen over the years, and exploration
   of the points they raise will result in a more robust reputation
   service.

   Obviously, in computing reputations via traffic analysis, some
   private algorithms may come into play.  For some RSPs, such "secret
   sauce" comprises their competitive advantage over others in the same
   space.  This document is not suggesting that all private algorithms
   need to be exposed for a reputation service to be acceptable.
   Instead, it is anticipated that enough of the above details need to
   be available to ensure consumers (and in some cases, industry or the
   general public) that the RSP can be trusted to influence key local
   policy decisions.

   Reptuations should be based on accurate identifiers, i.e., some
   property of the content under analysis that is difficult to falsify.
   For example, in the realm of email, the address found in the From:
   field of a message is typically not verifiable, while the domain name
   found in a validated domain-level signature is.  In this case,
   constructing a reputation system based on the domain name is more
   useful than one based on the From: field.

Kucherawy               Expires November 6, 2013                [Page 6]
Internet-Draft            Reputation Operations                 May 2013

   The biggest frustration with most RSPs to date has been the absence
   of a visible, accessible, and transparent process for remediating the
   errant addition of an identifier to a negative reputation list.  An
   RSP in widespread use is perceived to have enormous power when its
   results are used to reject traffic outright; when a "bad" entry is
   added referencing a good actor, it can have destructive effects, so
   an effective mechanism to fix such problems needs to exist.

   To accommodate clients with varying sensitivities, it is advisable
   for the query mechanism used to access the RSP to provide the ability
   to request details in the returned result about how the result was
   reached, allowing the client to decide if the result should be
   applied.  For example, it shoudl be possible for the reply to
   contain:

   o  the result itself;

   o  the number of data points used to compute the result;

   o  the age range of the data;

   o  source diversity of the input data;

   o  currency of the result (i.e., when it was computed);

   o  basis of the result (i.e., which identifier was used).

   The systems and algorithms used by the RSP to compute the reported
   reputation will need to be hardened as much as practicable against
   gaming or other forms of data poisoning.  Larger source diversities
   are harder to overcome with poisoned input, but are expensive to
   build in terms of both infrastructure and time.

   Systems focused on assigning positive reputations rather than negtive
   ones are promising since positive reputations, if made difficult to
   earn, put a large cost on bad actors, which may be enough to dissuade
   them entirely.

6.  Security Considerations

   Several points are raised above that can be described as threats to
   the delivery of valid user data.  This document highlights and
   discusses those matters, but introduces no new security issues.

7.  IANA Considerations

   This memo contains no actions for IANA.

Kucherawy               Expires November 6, 2013                [Page 7]
Internet-Draft            Reputation Operations                 May 2013

   [RFC Editor: Please remove this section prior to publication.]

8.  Informative References

   [DNS]    Mockapetris, P., "Domain Names -- Concepts and Facilities",
            RFC 1034, November 1987.

   [DNSBL]  Levine, J., "DNS Blacklists and Whitelists", RFC 5782,
            February 2010.

   [SMTP]   Klensin, J., "Simple Mail Transfer Protocol", RFC 5321,
            October 2008.

Appendix A.  Acknowledgements

   The author wishes to acknowledge the following for their review and
   constructive criticism of this proposal: Chris Barton, Vincent
   Schonau

Author's Address

   Murray S. Kucherawy

   EMail: superuser@gmail.com

Kucherawy               Expires November 6, 2013                [Page 8]