RFC 8117






Internet Engineering Task Force (IETF)                        C. Huitema
Request for Comments: 8117                          Private Octopus Inc.
Category: Informational                                        D. Thaler
ISSN: 2070-1721                                                Microsoft
                                                               R. Winter
                                 University of Applied Sciences Augsburg
                                                              March 2017


              Current Hostname Practice Considered Harmful

Abstract



   Giving a hostname to your computer and publishing it as you roam from
   one network to another is the Internet's equivalent of walking around
   with a name tag affixed to your lapel.  This current practice can
   significantly compromise your privacy, and something should change in
   order to mitigate these privacy threats.

   There are several possible remedies, such as fixing a variety of
   protocols or avoiding disclosing a hostname at all.  This document
   describes some of the protocols that reveal hostnames today and
   sketches another possible remedy, which is to replace static
   hostnames by frequently changing randomized values.

Status of This Memo



   This document is not an Internet Standards Track specification; it is
   published for informational purposes.

   This document is a product of the Internet Engineering Task Force
   (IETF).  It represents the consensus of the IETF community.  It has
   received public review and has been approved for publication by the
   Internet Engineering Steering Group (IESG).  Not all documents
   approved by the IESG are a candidate for any level of Internet
   Standard; see Section 2 of RFC 7841.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at
   http://www.rfc-editor.org/info/rfc8117.











Huitema, et al.               Informational                     [Page 1]

RFC 8117                Harmful Hostname Practice             March 2017


Copyright Notice



   Copyright (c) 2017 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
   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.  Naming Practices  . . . . . . . . . . . . . . . . . . . . . .   3
   3.  Partial Identifiers . . . . . . . . . . . . . . . . . . . . .   4
   4.  Protocols That Leak Hostnames . . . . . . . . . . . . . . . .   5
     4.1.  DHCP  . . . . . . . . . . . . . . . . . . . . . . . . . .   5
     4.2.  DNS Address to Name Resolution  . . . . . . . . . . . . .   5
     4.3.  Multicast DNS . . . . . . . . . . . . . . . . . . . . . .   6
     4.4.  Link-Local Multicast Name Resolution  . . . . . . . . . .   6
     4.5.  DNS-Based Service Discovery . . . . . . . . . . . . . . .   7
     4.6.  NetBIOS-over-TCP  . . . . . . . . . . . . . . . . . . . .   7
   5.  Randomized Hostnames as a Remedy  . . . . . . . . . . . . . .   8
   6.  Security Considerations . . . . . . . . . . . . . . . . . . .   9
   7.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .   9
   8.  Informative References  . . . . . . . . . . . . . . . . . . .   9
   Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . .  12
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .  12


















Huitema, et al.               Informational                     [Page 2]

RFC 8117                Harmful Hostname Practice             March 2017


1.  Introduction



   There is a long established practice of giving names to computers.
   In the Internet protocols, these names are referred to as "hostnames"
   [RFC7719].  Hostnames are normally used in conjunction with a domain
   name suffix to build the Fully Qualified Domain Name (FQDN) of a host
   [RFC1983].  However, it is common practice to use the hostname
   without further qualification in a variety of applications from file
   sharing to network management.  Hostnames are typically published as
   part of domain names and can be obtained through a variety of name
   lookup and discovery protocols.

   Hostnames have to be unique within the domain in which they are
   created and used.  They do not have to be globally unique
   identifiers, but they will always be at least partial identifiers, as
   discussed in Section 3.

   The disclosure of information through hostnames creates a problem for
   mobile devices.  Adversaries that monitor a remote network such as a
   Wi-Fi hot spot can obtain the hostname through passive monitoring or
   active probing of a variety of Internet protocols, such as DHCP or
   Multicast DNS (mDNS).  They can correlate the hostname with various
   other information extracted from traffic analysis and other
   information sources, and they can potentially identify the device,
   device properties, and its user [TRAC2016].

2.  Naming Practices



   There are many reasons to give names to computers.  This is
   particularly true when computers operate on a network.  Operating
   systems like Microsoft Windows or Unix assume that computers have a
   "hostname."  This enables users and administrators to do things such
   as ping a computer, add its name to an access control list, remotely
   mount a computer disk, or connect to the computer through tools such
   as telnet or remote desktop.  Other operating systems maintain
   multiple hostnames for different purposes, e.g., for use with certain
   protocols such as mDNS.

   In most consumer networks, naming is pretty much left to the
   discretion of the user.  Some will pick names of planets or stars,
   others will pick names of fruits or flowers, and still others will
   pick whatever suits their mood when they unwrap the device.  As long
   as users are careful to not pick a name already in use on the same
   network, anything goes.  Very often, however, the operating system
   suggests a hostname at the time of installation, which can contain
   the user name, the login name, and information learned from the
   device itself such as the brand, model, or maker of the device
   [TRAC2016].



Huitema, et al.               Informational                     [Page 3]

RFC 8117                Harmful Hostname Practice             March 2017


   In large organizations, collisions are more likely and a more
   structured approach is necessary.  In theory, organizations could use
   multiple DNS subdomains to ease the pressure on uniqueness, but in
   practice many don't and insist on unique flat names, if only to
   simplify network management.  To ensure unique names, organizations
   will set naming guidelines and enforce some kind of structured
   naming.  For example, within the Microsoft corporate network,
   computer names are derived from the login name of the main user,
   which leads to names like "huitema-test2" for a machine that one of
   the authors used to test software.

   There is less pressure to assign names to small devices including,
   for example, smart phones, as these devices typically do not enable
   sharing of their disks or remote login.  As a consequence, these
   devices often have manufacturer-assigned names, which vary from
   generic names like "Windows Phone" to completely unique names like
   "BrandX-123456-7890-abcdef" and often contain the name of the device
   owner, the device's brand name, and often also a hint as to which
   language the device owner speaks [TRAC2016].

3.  Partial Identifiers



   Suppose an adversary wants to track the people connecting to a
   specific Wi-Fi hot spot, for example, in a railroad station.  Assume
   that the adversary is able to retrieve the hostname used by a
   specific laptop.  That, in itself, might not be enough to identify
   the laptop's owner.  Suppose, however, that the adversary observes
   that the laptop name is "dthaler-laptop" and that the laptop has
   established a VPN connection to the Microsoft corporate network.  The
   two pieces of information, put together, firmly point to Dave Thaler,
   employed by Microsoft.  The identification is successful.

   In the example, we saw a login name inside the hostname, and that
   certainly helped identification.  But generic names like "jupiter" or
   "rosebud" also provide partial identification, especially if the
   adversary is capable of maintaining a database recording, among other
   information, the hostnames of devices used by specific users.
   Generic names are picked from vocabularies that include thousands of
   potential choices.  Finding the name reduces the scope of the search
   significantly.  Other information such as the visited sites will
   quickly complement that data and can lead to user identification.

   Also, the special circumstances of the network can play a role.
   Experiments on operational networks such as the IETF meeting network
   have shown that, with the help of external data such as the publicly
   available IETF attendees list or other data sources such as





Huitema, et al.               Informational                     [Page 4]

RFC 8117                Harmful Hostname Practice             March 2017


   Lightweight Directory Access Protocol (LDAP) servers on the network
   [TRAC2016], the identification of the device owner can become trivial
   given only partial identifiers in a hostname.

   Unique names assigned by manufacturers do not directly encode a user
   identifier, but they have the property of being stable and unique to
   the device in a large context.  A unique name like "BrandX-
   123456-7890-abcdef" allows efficient tracking across multiple
   domains.  In theory, this only allows tracking of the device but not
   of the user.  However, an adversary could correlate the device to the
   user through other means, for example, the one-time capture of some
   cleartext traffic.  Adversaries could then maintain databases linking
   a unique hostname to a user identity.  This will allow efficient
   tracking of both the user and the device.

4.  Protocols That Leak Hostnames



   Many IETF protocols can leak the "hostname" of a computer.  A non-
   exhaustive list includes DHCP, DNS address to name resolution,
   Multicast DNS, Link-local Multicast Name Resolution, and DNS service
   discovery.

4.1.  DHCP



   Shortly after connecting to a new network, a host can use DHCP
   [RFC2131] to acquire an IPv4 address and other parameters [RFC2132].
   A DHCP query can disclose the "hostname."  DHCP traffic is sent to
   the broadcast address and can be easily monitored, enabling
   adversaries to discover the hostname associated with a computer
   visiting a particular network.  DHCPv6 [RFC3315] shares similar
   issues.

   The problems with the hostname and FQDN parameters in DHCP are
   analyzed in [RFC7819] and [RFC7824].  Possible mitigations are
   described in [RFC7844].

4.2.  DNS Address to Name Resolution



   The domain name service design [RFC1035] includes the specification
   of the special domain "in-addr.arpa" for resolving the name of the
   computer using a particular IPv4 address, using the PTR format
   defined in [RFC1033].  A similar domain, "ip6.arpa", is defined in
   [RFC3596] for finding the name of a computer using a specific IPv6
   address.

   Adversaries who observe a particular address in use on a specific
   network can try to retrieve the PTR record associated with that
   address and thus the hostname of the computer, or even the FQDN of



Huitema, et al.               Informational                     [Page 5]

RFC 8117                Harmful Hostname Practice             March 2017


   that computer.  The retrieval may not be useful in many IPv4 networks
   due to the prevalence of NAT, but it could work in IPv6 networks.
   Other name lookup mechanisms, such as [RFC4620], share similar
   issues.

4.3.  Multicast DNS



   Multicast DNS (mDNS) is defined in [RFC6762].  It enables hosts to
   send DNS queries over multicast and to elicit responses from hosts
   participating in the service.

   If an adversary suspects that a particular host is present on a
   network, the adversary can send mDNS requests to find, for example,
   the A or AAAA records associated with the hostname in the ".local"
   domain.  A positive reply will confirm the presence of the host.

   When a new responder starts, it must send a set of multicast queries
   to verify that the name that it advertises is unique on the network
   and to populate the caches of other mDNS hosts.  Adversaries can
   monitor this traffic and discover the hostname of computers as they
   join the monitored network.

   mDNS further allows queries to be sent via unicast to port 5353.  An
   adversary might decide to use unicast instead of multicast in order
   to hide from, e.g., intrusion detection systems.

4.4.  Link-Local Multicast Name Resolution



   Link-Local Multicast Name Resolution (LLMNR) is defined in [RFC4795].
   The specification did not achieve consensus as an IETF standard, but
   it is widely deployed.  Like mDNS, it enables hosts to send DNS
   queries over multicast and to elicit responses from computers
   implementing the LLMNR service.

   Like mDNS, LLMNR can be used by adversaries to confirm the presence
   of a specific host on a network by issuing a multicast request to
   find the A or AAAA records associated with the hostname in the
   ".local" domain.

   When an LLMNR responder starts, it sends a set of multicast queries
   to verify that the name that it advertises is unique on the network.
   Adversaries can monitor this traffic and discover the hostname of
   computers as they join the monitored network.








Huitema, et al.               Informational                     [Page 6]

RFC 8117                Harmful Hostname Practice             March 2017


4.5.  DNS-Based Service Discovery



   DNS-based Service Discovery (DNS-SD) is described in [RFC6763].  It
   enables participating hosts to retrieve the location of services
   proposed by other hosts.  It can be used with DNS servers or in
   conjunction with mDNS in a serverless environment.

   Participating hosts publish a service described by an "instance
   name", which is typically chosen by the user responsible for the
   publication.  While this is obviously an active disclosure of
   information, privacy aspects can be mitigated by user control.
   Services should only be published when deciding to do so, and the
   information disclosed in the service name should be well under the
   control of the device's owner.

   In theory, there should not be any privacy issue, but in practice the
   publication of a service also forces the publication of the hostname
   due to a chain of dependencies.  The service name is used to publish
   a PTR record announcing the service.  The PTR record typically points
   to the service name in the local domain.  The service names, in turn,
   are used to publish TXT records describing service parameters and SRV
   records describing the service location.

   SRV records are described in [RFC2782].  Each record contains four
   parameters: priority, weight, port number, and hostname.  While the
   service name published in the PTR record is chosen by the user, the
   "hostname" in the SRV record is indeed the hostname of the device.

   Adversaries can monitor the mDNS traffic associated with DNS-SD and
   retrieve the hostname of computers advertising any service with DNS-
   SD.

4.6.  NetBIOS-over-TCP



   Amongst other things, NetBIOS-over-TCP [RFC1002] implements a name
   registration and resolution mechanism called the NetBIOS Name
   Service.  In practice, NetBIOS resource names are often based on
   hostnames.

   NetBIOS allows an application to register resource names and to
   resolve such names to IP addresses.  In environments without a
   NetBIOS Name Server, the protocol makes extensive use of broadcasts
   from which resource names can be easily extracted.  NetBIOS also
   allows querying for the names registered by a node directly (node
   status).






Huitema, et al.               Informational                     [Page 7]

RFC 8117                Harmful Hostname Practice             March 2017


5.  Randomized Hostnames as a Remedy



   There are several ways to remedy the hostname practices.  We could
   instruct people to just turn off any protocol that leaks hostnames,
   at least when they visit some "insecure" place.  We could also
   examine each particular standard that publishes hostnames and somehow
   fix the corresponding protocols.  Or, we could attempt to revise the
   way devices manage the hostname parameter.

   There is a lot of merit in turning off unneeded protocols when
   visiting insecure places.  This amounts to attack-surface reduction
   and is clearly beneficial -- this is an advantage of the stealth mode
   defined in [RFC7288].  However, there are two issues with this
   advice.  First, it relies on recognizing which networks are secure or
   insecure.  This is hard to automate, but relying on end-user judgment
   may not always provide good results.  Second, some protocols such as
   DHCP cannot be turned off without losing connectivity, which limits
   the value of this option.  Also, the services that rely on protocols
   that leak hostnames such as mDNS will not be available when switched
   off.  In addition, not always are hostname-leaking protocols well-
   known, as they might be proprietary and come with an installed
   application instead of being provided by the operating system.

   It may be possible in many cases to examine a protocol and prevent it
   from leaking hostnames.  This is, for example, what is attempted for
   DHCP in [RFC7844].  However, it is unclear that we can identify,
   revisit, and fix all the protocols that publish hostnames.  In
   particular, this is impossible for proprietary protocols.

   We may be able to mitigate most of the effects of hostname leakage by
   revisiting the way platforms handle hostnames.  In a way, this is
   similar to the approach of Media Access Control (MAC) address
   randomization described in [RFC7844].  Let's assume that the
   operating system, at the time of connecting to a new network, picks a
   random hostname and starts publicizing that random name in protocols
   such as DHCP or mDNS, instead of the static value.  This will render
   monitoring and identification of users by adversaries much more
   difficult without preventing protocols such as DNS-SD from operating
   as expected.  This, of course, has implications on the applications
   making use of such protocols, e.g., when the hostname is being
   displayed to users of the application.  They will not as easily be
   able to identify, e.g., network shares or services based on the
   hostname carried in the underlying protocols.  Also, the generation
   of new hostnames should be synchronized with the change of other
   tokens used in network protocols such as the MAC or IP address to
   prevent correlation of this information.  For example, if the IP





Huitema, et al.               Informational                     [Page 8]

RFC 8117                Harmful Hostname Practice             March 2017


   address changes but the hostname stays the same, the new IP address
   can be correlated to belong to the same device based on a leaked
   hostname.

   Some operating systems, including Windows, support "per network"
   hostnames, but some other operating systems only support "global"
   hostnames.  In that case, changing the hostname may be difficult if
   the host is multihomed, as the same name will be used on several
   networks.  Other operating systems already use potentially different
   hostnames for different purposes, which might be a good model to
   combine both static hostnames and randomized hostnames based on their
   potential use and threat to a user's privacy.

   Obviously, further studies are required before the idea of randomized
   hostnames can be implemented.

6.  Security Considerations



   This document does not introduce any new protocol.  It does point to
   potential privacy issues in a set of existing protocols.

   There are obvious privacy gains to changing to randomized hostnames
   and also to changing these names frequently.  However, wide
   deployment might affect security functions or current practices.  For
   example, incident response using hostnames to track the source of
   traffic might be affected.  It is common practice to include
   hostnames and reverse lookup information at various times during an
   investigation.

7.  IANA Considerations



   This document does not require any IANA actions.

8.  Informative References



   [RFC1002]  NetBIOS Working Group in the Defense Advanced Research
              Projects Agency, Internet Activities Board, and End-to-End
              Services Task Force, "Protocol standard for a NetBIOS
              service on a TCP/UDP transport: Detailed specifications",
              STD 19, RFC 1002, DOI 10.17487/RFC1002, March 1987,
              <http://www.rfc-editor.org/info/rfc1002>.

   [RFC1033]  Lottor, M., "Domain Administrators Operations Guide",
              RFC 1033, DOI 10.17487/RFC1033, November 1987,
              <http://www.rfc-editor.org/info/rfc1033>.






Huitema, et al.               Informational                     [Page 9]

RFC 8117                Harmful Hostname Practice             March 2017


   [RFC1035]  Mockapetris, P., "Domain names - implementation and
              specification", STD 13, RFC 1035, DOI 10.17487/RFC1035,
              November 1987, <http://www.rfc-editor.org/info/rfc1035>.

   [RFC1983]  Malkin, G., Ed., "Internet Users' Glossary", FYI 18,
              RFC 1983, DOI 10.17487/RFC1983, August 1996,
              <http://www.rfc-editor.org/info/rfc1983>.

   [RFC2131]  Droms, R., "Dynamic Host Configuration Protocol",
              RFC 2131, DOI 10.17487/RFC2131, March 1997,
              <http://www.rfc-editor.org/info/rfc2131>.

   [RFC2132]  Alexander, S. and R. Droms, "DHCP Options and BOOTP Vendor
              Extensions", RFC 2132, DOI 10.17487/RFC2132, March 1997,
              <http://www.rfc-editor.org/info/rfc2132>.

   [RFC2782]  Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for
              specifying the location of services (DNS SRV)", RFC 2782,
              DOI 10.17487/RFC2782, February 2000,
              <http://www.rfc-editor.org/info/rfc2782>.

   [RFC3315]  Droms, R., Ed., Bound, J., Volz, B., Lemon, T., Perkins,
              C., and M. Carney, "Dynamic Host Configuration Protocol
              for IPv6 (DHCPv6)", RFC 3315, DOI 10.17487/RFC3315, July
              2003, <http://www.rfc-editor.org/info/rfc3315>.

   [RFC3596]  Thomson, S., Huitema, C., Ksinant, V., and M. Souissi,
              "DNS Extensions to Support IP Version 6", RFC 3596,
              DOI 10.17487/RFC3596, October 2003,
              <http://www.rfc-editor.org/info/rfc3596>.

   [RFC4620]  Crawford, M. and B. Haberman, Ed., "IPv6 Node Information
              Queries", RFC 4620, DOI 10.17487/RFC4620, August 2006,
              <http://www.rfc-editor.org/info/rfc4620>.

   [RFC4795]  Aboba, B., Thaler, D., and L. Esibov, "Link-local
              Multicast Name Resolution (LLMNR)", RFC 4795,
              DOI 10.17487/RFC4795, January 2007,
              <http://www.rfc-editor.org/info/rfc4795>.

   [RFC6762]  Cheshire, S. and M. Krochmal, "Multicast DNS", RFC 6762,
              DOI 10.17487/RFC6762, February 2013,
              <http://www.rfc-editor.org/info/rfc6762>.

   [RFC6763]  Cheshire, S. and M. Krochmal, "DNS-Based Service
              Discovery", RFC 6763, DOI 10.17487/RFC6763, February 2013,
              <http://www.rfc-editor.org/info/rfc6763>.




Huitema, et al.               Informational                    [Page 10]

RFC 8117                Harmful Hostname Practice             March 2017


   [RFC7288]  Thaler, D., "Reflections on Host Firewalls", RFC 7288,
              DOI 10.17487/RFC7288, June 2014,
              <http://www.rfc-editor.org/info/rfc7288>.

   [RFC7719]  Hoffman, P., Sullivan, A., and K. Fujiwara, "DNS
              Terminology", RFC 7719, DOI 10.17487/RFC7719, December
              2015, <http://www.rfc-editor.org/info/rfc7719>.

   [RFC7819]  Jiang, S., Krishnan, S., and T. Mrugalski, "Privacy
              Considerations for DHCP", RFC 7819, DOI 10.17487/RFC7819,
              April 2016, <http://www.rfc-editor.org/info/rfc7819>.

   [RFC7824]  Krishnan, S., Mrugalski, T., and S. Jiang, "Privacy
              Considerations for DHCPv6", RFC 7824,
              DOI 10.17487/RFC7824, May 2016,
              <http://www.rfc-editor.org/info/rfc7824>.

   [RFC7844]  Huitema, C., Mrugalski, T., and S. Krishnan, "Anonymity
              Profiles for DHCP Clients", RFC 7844,
              DOI 10.17487/RFC7844, May 2016,
              <http://www.rfc-editor.org/info/rfc7844>.

   [TRAC2016] Faath, M., Winter, R., and F. Weisshaar, "How Broadcast
              Data Reveals Your Identity and Social Graph", IEEE,
              Wireless Communications and Mobile Computing Conference
              (IWCMC), 2016 International,
              DOI 10.1109/IWCMC.2016.7577084, September 2016.
























Huitema, et al.               Informational                    [Page 11]

RFC 8117                Harmful Hostname Practice             March 2017


Acknowledgments

   Thanks to the members of the INTAREA Working Group for discussions
   and reviews.

Authors' Addresses



   Christian Huitema
   Private Octopus Inc.
   Friday Harbor, WA  98250
   United States of America

   Email: huitema@huitema.net


   Dave Thaler
   Microsoft
   Redmond, WA  98052
   United States of America

   Email: dthaler@microsoft.com


   Rolf Winter
   University of Applied Sciences Augsburg
   Augsburg
   DE

   Email: rolf.winter@hs-augsburg.de






















Huitema, et al.               Informational                    [Page 12]