Network Time Protocol (ntp) Working Group F. Gont
Internet-Draft G. Gont
Updates: rfc5905 (if approved) SI6 Networks
Intended status: Standards Track M. Lichvar
Expires: September 10, 2020 Red Hat
March 9, 2020
Port Randomization in the Network Time Protocol Version 4
draft-ietf-ntp-port-randomization-01
Abstract
The Network Time Protocol can operate in several modes. Some of
these modes are based on the receipt of unsolicited packets, and
therefore require the use of a service/well-known port as the local
port number. However, in the case of NTP modes where the use of a
service/well-known port is not required, employing such well-known/
service port unnecessarily increases the ability of attackers to
perform blind/off-path attacks. This document formally updates
RFC5905, recommending the use of port randomization for those modes
where use of the NTP service port is not required.
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 https://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 September 10, 2020.
Copyright Notice
Copyright (c) 2020 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
(https://trustee.ietf.org/license-info) in effect on the date of
Gont, et al. Expires September 10, 2020 [Page 1]
Internet-Draft NTP Port Randomization March 2020
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 . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3
3. Considerations About Port Randomization in NTP . . . . . . . 3
3.1. Mitigation Against Off-path Attacks . . . . . . . . . . . 3
3.2. Effects on Path Selection . . . . . . . . . . . . . . . . 4
3.3. Filtering of NTP traffic . . . . . . . . . . . . . . . . 4
3.4. Effect on NAT devices . . . . . . . . . . . . . . . . . . 5
3.5. Relation to Other Mitigations for Off-Path Attacks . . . 5
4. Update to RFC5905 . . . . . . . . . . . . . . . . . . . . . . 5
5. Implementation Status . . . . . . . . . . . . . . . . . . . . 6
6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7
7. Security Considerations . . . . . . . . . . . . . . . . . . . 7
8. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 7
9. References . . . . . . . . . . . . . . . . . . . . . . . . . 8
9.1. Normative References . . . . . . . . . . . . . . . . . . 8
9.2. Informative References . . . . . . . . . . . . . . . . . 8
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 9
1. Introduction
The Network Time Protocol (NTP) is one of the oldest Internet
protocols, and currently specified in [RFC5905]. Since its original
implementation, standardization, and deployment, a number of
vulnerabilities have been found both in the NTP specification and in
some of its implementations [NTP-VULN]. Some of these
vulnerabilities allow for off-path/blind attacks, where an attacker
can send forged packets to one or both NTP peers for achieving Denial
of Service (DoS), time-shifts, and other undesirable outcomes. Many
of these attacks require the attacker to guess or know at least a
target NTP association, typically identified by the tuple {srcaddr,
srcport, dstaddr, dstport, keyid}. Some of these parameters may be
easily known or guessed.
NTP can operate in several modes. Some of these modes rely on the
ability of nodes to receive unsolicited packets, and therefore
require the use of a service/well-known port number. However, for
modes where the use of a service/well-known port is not required,
employing such well-known/service port improves the ability of an
attacker to perform blind/off-path attacks (since knowledge of such
Gont, et al. Expires September 10, 2020 [Page 2]
Internet-Draft NTP Port Randomization March 2020
port number is typically required for such attacks). A recent study
[NIST-NTP] that analyzes the port numbers employed by NTP clients
suggests that a considerable number of NTP clients employ the NTP
service/well-known port as their local port, or select predictable
ephemeral port numbers, thus improving the ability of attackers to
perform blind/off-path attacks against NTP.
BCP 156 [RFC6056] already recommends the randomization of transport-
protocol ephemeral ports. This document aligns NTP with the
recommendation in BCP 156 [RFC6056], by formally updating [RFC5905]
such that port randomization is employed for those NTP modes for
which the use of the NTP service port is not required.
2. Terminology
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in [RFC2119].
3. Considerations About Port Randomization in NTP
The following subsections analyze a number of considerations about
transport-protocol port randomization when applied to NTP.
3.1. Mitigation Against Off-path Attacks
There has been a fair share of work in the area of off-path/blind
attacks against transport protocols and upper-layer protocols, such
as [RFC5927] and [RFC4953]. Whether the target of the attack is a
transport protocol instance (e.g., TCP connection) or an upper-layer
protocol instance (e.g., an application protocol instance), the
attacker is required to know or guess the five-tuple {Protocol, IP
Source Address, IP Destination Address, Source Port, Destination
Port} that identifies the target transport protocol instance or the
transport protocol instance employed by the target upper-layer
protocol instance. Therefore, increasing the difficulty of guessing
this five-tuple helps mitigate blind/off-path attacks.
As a result of this considerations, BCP 156 [RFC6056] recommends the
randomization of transport-protocol ephemeral ports. And as such,
this document aims to bring the NTP specification [RFC5905] in line
with the aforementioned recommendation.
We note that the use of port randomization is a transport-layer
mitigation against off-path/blind attacks, and does not preclude (nor
is it precluded by) other possible mitigations for off-path attacks
that might be implemented by an application protocol (e.g.
[I-D.ietf-ntp-data-minimization]). For instance, some of the
Gont, et al. Expires September 10, 2020 [Page 3]
Internet-Draft NTP Port Randomization March 2020
aforementioned mitigations may be ineffective against some off-path
attacks [NTP-FRAG] or may benefit from the additional entropy
provided by port randomization [NTP-security].
3.2. Effects on Path Selection
Intermediate systems implementing the Equal-Cost Multi-Path (ECMP)
algorithm may select the outgoing link by computing a hash over a
number of values, that include the transport-protocol source port.
Thus, as discussed in [NTP-CHLNG], the selected client port may have
an influence on the measured delay and jitter values.
This might mean, for example, that two clients in the same network
synchronized with the same NTP server using a stable source port
(selected randomly or not) have a significant offset between their
clocks due to their NTP exchanges taking paths with different
asymmetry in the network delay.
If the clients changed their source port with each request, packets
in different exchanges would take different paths. The measured
delay and offset would be less stable, but the offset between the
clients' clocks would be smaller. The impact on stability of the
clocks would be mitigated by the clock filter algorithm, which
prefers samples with shorter delay.
On the other hand, if port-randomization is applied on a per-
association basis, request/responses to the same association would
likely follow the same path, since the IP addresses and transport
port numbers employed for an association would not change. This
would thus result in more stability of the measurements.
Section 4 recommends NTP implementations to randomize the ephemeral
port number of non-symmetrical associations on a per-association
basis (as opposed to "per-request"), since this is believed to be the
more conservative approach. However, implementations may override
this setting and perform port-randomization on a per-request basis.
3.3. Filtering of NTP traffic
In a number of scenarios (such as when mitigating DDoS attacks), a
network operator may want to differentiate between NTP requests sent
by clients, and NTP responses sent by NTP servers. If an
implementation employs the NTP service port for the client port
number, requests/responses cannot be readily differentiated by
inspecting the source and destination port numbers. Implementation
of port randomization for non-symmetrical modes allows for simple
differentiation of NTP requests and responses, and for the
Gont, et al. Expires September 10, 2020 [Page 4]
Internet-Draft NTP Port Randomization March 2020
enforcement of security policies that may be valuable for the
mitigation of DDoS attacks.
3.4. Effect on NAT devices
Some NAT devices will not translate the source port of a packet when
a privileged port number is employed. In networks where such NAT
devices are employed, use of the NTP service port for the client port
will essentially limit the number of hosts that may successfully
employ NTP client implementations.
In the case of NAT devices that will translate the source port even
when a privileged port is employed, packets reaching the external
realm of the NAT will not employ the NTP service port as the local
port, since the local port will normally be translated by the NAT
device possibly, but not necessarily, with a random port.
3.5. Relation to Other Mitigations for Off-Path Attacks
Ephemeral Port Randomization is a best current practice (BCP 156)
that helps mitigate off-path attacks at the transport-layer. It is
orthogonal to other possible mitigations for off-path attacks that
may be implemented at other layers (such as the use of timestamps in
NTP) which may or may not be effective against some off-path attacks
(see e.g. [NTP-FRAG]. This document aligns NTP with the existing
best current practice on ephemeral port selection, irrespective of
other techniques that may (and should) be implemented for mitigating
off-path attacks.
The following text from Section 9.1 ("Peer Process Variables") of
[RFC5905]:
dstport: UDP port number of the client, ordinarily the NTP port
number PORT (123) assigned by the IANA. This becomes the source
port number in packets sent from this association.
is replaced with:
dstport: UDP port number of the client. In the case of broadcast
server mode (5) and symmetric modes (1 and 2), it SHOULD contain
the NTP port number PORT (123) assigned by the IANA. In other
cases, it SHOULD contain a randomized port number, as specified in
[RFC6056]. The value in this variable becomes the source port
number of packets sent from this association.
Gont, et al. Expires September 10, 2020 [Page 5]
Internet-Draft NTP Port Randomization March 2020
NOTES:
When port randomization is employed, the port number SHOULD be
randomized on a per-association basis. That is, a random port
number SHOULD selected when an association is first mobilized,
and the selected port number is expected to remain constant
during the life of an association. An implementation MAY,
however, override this setting and employ port randomization on
a per-request basis.
On most current operating systems, which implement ephemeral
port randomization [RFC6056], an NTP client may normally rely
on the operating system to perform port randomization. For
example, NTP implementations using POSIX sockets may achieve
port randomization by *not* binding the socket with the bind()
function, or binding it to port 0, which has a special meaning
of "any port". connect()ing the docket will make the port
inaccessible by other systems (that is, only packets from the
specified remote socket will be received by the application).
5. Implementation Status
[RFC Editor: Please remove this section before publication of this
document as an RFC.]
This section records the status of known implementations of the
protocol defined by this specification at the time of posting of this
Internet-Draft, and is based on a proposal described in [RFC7942].
The description of implementations in this section is intended to
assist the IETF in its decision processes in progressing drafts to
RFCs. Please note that the listing of any individual implementation
here does not imply endorsement by the IETF. Furthermore, no effort
has been spent to verify the information presented here that was
supplied by IETF contributors. This is not intended as, and must not
be construed to be, a catalog of available implementations or their
features. Readers are advised to note that other implementations may
exist.
OpenNTPD:
[OpenNTPD] has never explicitly set the local port of NTP clients,
and thus employs the ephemeral port selection algorithm
implemented by the operating system. Thus, on all operating
systems that implement port randomization (such as current
versions of OpenBSD, Linux, and FreeBSD), OpenNTPD will employ
port randomization for client ports.
chrony:
[chrony] by default does not set the local client port, and thus
employs the ephemeral port selection algorithm implemented by the
Gont, et al. Expires September 10, 2020 [Page 6]
Internet-Draft NTP Port Randomization March 2020
operating system. Thus, on all operating systems that implement
port randomization (such as current versions of OpenBSD, Linux,
and FreeBSD), chrony will employ port randomization for client
ports.
nwtime.org's sntp client:
sntp does not explicitly set the local port, and thus employs the
ephemeral port selection algorithm implemented by the operating
system. Thus, on all operating systems that implement port
randomization (such as current versions of OpenBSD, Linux, and
FreeBSD), it will employ port randomization for client ports.
6. IANA Considerations
There are no IANA registries within this document. The RFC-Editor
can remove this section before publication of this document as an
RFC.
7. Security Considerations
The security implications of predictable numeric identifiers
[I-D.gont-predictable-numeric-ids] (and of predictable transport-
protocol port numbers [RFC6056] in particular) have been known for a
long time now. However, the NTP specification has traditionally
followed a pattern of employing common settings and code even when
not strictly necessary, which at times has resulted in negative
security and privacy implications (see e.g.
[I-D.ietf-ntp-data-minimization]). The use of the NTP service port
(123) for the srcport and dstport variables is not required for all
operating modes, and such unnecessary usage comes at the expense of
reducing the amount of work required for an attacker to successfully
perform off-path/blind attacks against NTP. Therefore, this document
formally updates [RFC5905], recommending the use of transport-
protocol port randomization when use of the NTP service port is not
required.
This issue has been tracked by US-CERT with VU#597821, and has been
assigned CVE-2019-11331.
8. Acknowledgments
Watson Ladd raised the problem of DDoS mitigation when the NTP
service port is employed as the client port (discussed in Section 3.3
of this document).
The authors would like to thank (in alphabetical order) Ivan Arce,
Todd Glassey, Watson Ladd, Aanchal Malhotra, Danny Mayer, Gary E.
Gont, et al. Expires September 10, 2020 [Page 7]
Internet-Draft NTP Port Randomization March 2020
Miller, Dieter Sibold, Steven Sommars, and Ulrich Windl, for
providing valuable comments on earlier versions of this document.
The authors would like to thank Harlan Stenn for answering questions
about nwtime.org's NTP implementation.
Fernando would like to thank Nelida Garcia and Jorge Oscar Gont, for
their love and support.
9. References
9.1. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>.
[RFC5905] Mills, D., Martin, J., Ed., Burbank, J., and W. Kasch,
"Network Time Protocol Version 4: Protocol and Algorithms
Specification", RFC 5905, DOI 10.17487/RFC5905, June 2010,
<https://www.rfc-editor.org/info/rfc5905>.
[RFC6056] Larsen, M. and F. Gont, "Recommendations for Transport-
Protocol Port Randomization", BCP 156, RFC 6056,
DOI 10.17487/RFC6056, January 2011,
<https://www.rfc-editor.org/info/rfc6056>.
9.2. Informative References
[chrony] "chrony", <https://chrony.tuxfamily.org/>.
[I-D.gont-predictable-numeric-ids]
Gont, F. and I. Arce, "Security and Privacy Implications
of Numeric Identifiers Employed in Network Protocols",
draft-gont-predictable-numeric-ids-03 (work in progress),
March 2019.
[I-D.ietf-ntp-data-minimization]
Franke, D. and A. Malhotra, "NTP Client Data
Minimization", draft-ietf-ntp-data-minimization-04 (work
in progress), March 2019.
[NIST-NTP]
Sherman, J. and J. Levine, "Usage Analysis of the NIST
Internet Time Service", Journal of Research of the
National Institute of Standards and Technology Volume 121,
March 2016, <https://tf.nist.gov/general/pdf/2818.pdf>.
Gont, et al. Expires September 10, 2020 [Page 8]
Internet-Draft NTP Port Randomization March 2020
[NTP-CHLNG]
Sommars, S., "Challenges in Time Transfer Using the
Network Time Protocol (NTP)", Proceedings of the 48th
Annual Precise Time and Time Interval Systems and
Applications Meeting, Monterey, California pp. 271-290,
January 2017, <http://leapsecond.com/ntp/
NTP_Paper_Sommars_PTTI2017.pdf>.
[NTP-FRAG]
Malhotra, A., Cohen, I., Brakke, E., and S. Goldberg,
"Attacking the Network Time Protocol", NDSS'17, San Diego,
CA. Feb 2017, 2017,
<http://www.cs.bu.edu/~goldbe/papers/NTPattack.pdf>.
[NTP-security]
Malhotra, A., Van Gundy, M., Varia, V., Kennedy, H.,
Gardner, J., and S. Goldberg, "The Security of NTP's
Datagram Protocol", Cryptology ePrint Archive Report
2016/1006, 2016, <https://eprint.iacr.org/2016/1006>.
[NTP-VULN]
Network Time Foundation, "Security Notice", Network Time
Foundation's NTP Support Wiki ,
<https://support.ntp.org/bin/view/Main/SecurityNotice>.
[OpenNTPD]
"OpenNTPD Project", <https://www.openntpd.org>.
[RFC4953] Touch, J., "Defending TCP Against Spoofing Attacks",
RFC 4953, DOI 10.17487/RFC4953, July 2007,
<https://www.rfc-editor.org/info/rfc4953>.
[RFC5927] Gont, F., "ICMP Attacks against TCP", RFC 5927,
DOI 10.17487/RFC5927, July 2010,
<https://www.rfc-editor.org/info/rfc5927>.
[RFC7942] Sheffer, Y. and A. Farrel, "Improving Awareness of Running
Code: The Implementation Status Section", BCP 205,
RFC 7942, DOI 10.17487/RFC7942, July 2016,
<https://www.rfc-editor.org/info/rfc7942>.
Authors' Addresses
Gont, et al. Expires September 10, 2020 [Page 9]
Internet-Draft NTP Port Randomization March 2020
Fernando Gont
SI6 Networks
Evaristo Carriego 2644
Haedo, Provincia de Buenos Aires 1706
Argentina
Phone: +54 11 4650 8472
Email: fgont@si6networks.com
URI: https://www.si6networks.com
Guillermo Gont
SI6 Networks
Evaristo Carriego 2644
Haedo, Provincia de Buenos Aires 1706
Argentina
Phone: +54 11 4650 8472
Email: ggont@si6networks.com
URI: https://www.si6networks.com
Miroslav Lichvar
Red Hat
Purkynova 115
Brno 612 00
Czech Republic
Email: mlichvar@redhat.com
Gont, et al. Expires September 10, 2020 [Page 10]
