| Internet-Draft | NTPv5 use cases and requirements | October 2021 |
| Gruessing | Expires 13 April 2022 | [Page] |
This document describes the use cases, requirements, and considerations that should be factored in the design of a successor protocol to supersede version 4 of the NTP protocol [RFC5905] presently referred to as NTP version 5 ("NTPv5"). This document is non-exhaustive and does not in its current version represent working group consensus.¶
RFC Editor: please remove this section before publication¶
Source code and issues for this draft can be found at https://github.com/fiestajetsam/I-D/tree/main/draft-gruessing-ntp-ntpv5-requirements.¶
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 13 April 2022.¶
Copyright (c) 2021 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 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.¶
NTP version 4 [RFC5905] has seen active use for over a decade, and within this time period the protocol has not only been extended to support new requirements but also fallen victim to vulnerabilities that have made it used for distributed denial of service (DDoS) amplification attacks.¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
There are several common scenarios for existing NTPv4 deployments; publicly accessible NTP services such as the NTP Pool [ntppool] are used to offer clock synchronisation for end users and embedded devices, ISP provided servers to synchronise devices such as customer-premises equipment where reduced accuracy may be tolerable. Depending on the network and path these deployments may be affected by variable latency as well as throttling or blocking by providers.¶
Data centres and cloud computing providers also have deployed and offer NTP services both for internal use and for customers, particularly where the network is unable to offer or does not require PTP [IEEE-1588-2008]. As these deployments are less likely to be constrained by network latency or power the potential for higher levels of accuracy and precision within the bounds of the protocol are possible.¶
At a high level, NTPv5 should be a protocol that is capable of operating in both local networks and also over public internet connections where packet loss, delay, and even filtering may occur.¶
Historically there have been many documented instances of NTP servers taking a large increase in unauthorised traffic [ntp-misuse] and the design of NTPv5 must ensure the risk of these can be minimised to the fullest extent.¶
Servers SHOULD have a new identifier that peers use as reference, this SHOULD NOT be a FQDN, an IP address or identifier tied to a public certificate. Servers SHOULD be able to migrate and change their identifiers as stratum topologies or network configuration changes occur.¶
The specification MUST have support for servers to notify clients that the service is unavailable, and clients MUST have clearly defined behaviours honouring this signalling. In addition to this servers SHOULD be able to communicate to clients that they should reduce their query interval rate when the server is under high bandwidth or has reduced capacity.¶
Clients SHOULD re-establish connections with servers at an interval to prevent attempting to maintain connectivity to a dead host and give network operators the ability to move traffic away from hosts in a timely manner.¶
The specification SHOULD have provisions for deployments where Network Address Translation occurs, and define behaviours when NAT rebinding occurs. This should also not compromise any DDoS mitigation(s) that the protocol may define.¶
Algorithms describing functions such as clock filtering, selection and clustering SHOULD be omitted from the protocol specification; the specification should instead only provide what is necessary to describe protocol semantics and normative behaviours.¶
The working group should consider creating a separate informational document to describe an algorithm to assist with implementation, and to consider adopting future documents which describe new algorithms as they are developed. Specifying client algorithms separately from the protocol allows will allow NTPv5 to meet the needs of applications with a variety of network properties and performance requirements. It also allows for innovation in implementations without sacrificing basic interoperability.¶
The protocol SHOULD adopt a linear, monotonic timescale as the basis for communicating time. The format should meet sufficient scale and precision with resolution either meeting or exceeding NTPv4, and have a rollover date sufficiently far enough into the future that the protocol's complete obsolescence is most likely to occur first. The specification should continue to use the existing NTP epoch unless requirements on precision, resolution, or backwards compatibility cannot be met.¶
The timescale in addition to any other time sensitive information must be sufficient to calculate representations of both UTC and TAI. Through extensions the protocol SHOULD support additional timescale representations outside of the main specification, and all transmissions of time data SHALL indicate the timescale in use.¶
Support for UTC leap second information MUST be included in the protocol specification in order for clients to generate a UTC representation but must be transmitted as separate information to the timescale. The specification SHOULD also be capable of transmitting upcoming leap seconds greater than 1 calendar day in advance.¶
Leap second smearing SHOULD NOT be part of the wire specification, however this should not prevent implementers from applying leap second smearing between the client and any clock it is training but MUST NOT be applied to downstream clients.¶
The support for compatibility with other protocols should not prevent addressing issues that have previous caused issues in deployments or cause ossification of the protocol.¶
Protocol ossification MUST be addressed to prevent existing NTPv4 deployments which incorrectly respond to clients posing as NTPv5 from causing issues. Forward prevention of ossification (for a potential NTPv6 protocol in the future) should also be taken into consideration.¶
The model for backward compatibility is servers that support multiple versions NTP and send a response in the same version as the request. This does not preclude servers from acting as a client in one version of NTP and a server in another.¶
Many other documents make use of NTP's data formats ([RFC5905] Section 6) for representing time, notably for media and packet timestamp measurements. Any changes to the data formats should consider the potential implementation complexity that may be incurred.¶
The protocol MUST have the capability to be extended, and that implementations MUST ignore unknown extensions. Unknown extensions received by a server from a lower stratum server SHALL not be added to response messages sent by the server receiving these extensions.¶
Considerations should be made about the future of the existing IANA registry for NTPv4 parameters. If NTPv5 becomes incompatible with these parameters a new registry SHOULD be created.¶
Encryption and authentication MUST be provided by the protocol specification as a default and MUST be resistant to downgrade attacks. The encryption used must have agility, allowing for the protocol to update as more secure cryptography becomes known and vulnerabilities are discovered.¶
The specification MAY consider leaving room for middleboxes which may deliberately modify packets in flight for legitimate purposes. Thought must be given around how this will be incorporated into any applicable trust model. Downgrading attacks that could lead to an adversary disabling or removing encryption or authentication MUST NOT be possible in the design of the protocol.¶
Detection and reporting of server malfeasance SHOULD remain out of scope of this specification as [I-D.ietf-ntp-roughtime] already provides this capability as a core functionality of the protocol.¶
TODO: Describe the assumptions which support this model, separating those which can be generic to any deployment, and those that are more industry or deployment specific.¶
The risk that an on-path attacker can delay packets between a client and server exists in all time protocols operating on insecure networks and its mitigations are limited within the protocol with a clock which is not yet synchronised. Increased path diversity and protocol support for synchronisation across multiple heterogeneous sources are likely the most effective mitigations.¶
Conversely on-path attackers who can manipulate timestamps could also speed up a client's clock, also resulting into drift-related malfunctions and errors such as incorrect expiration of public certificates on the affected hosts. An attacker may also manipulate other data in flight to disrupt service and cause de-synchronisation. In both cases having message authentication with a regular key rotation interval should mitigate; however consideration should be made for hardware based timestamping.¶
NTPv4 has previously suffered from DDoS amplification attacks using a combination of IP address spoofing with a private mode commands used in many NTP implementations, leading to an attacker being able to orders of magnitude of traffic to a victim IP address. Current mitigation uses a combination of disabling the use of private mode commands, in addition to encouraging network operators to implement BCP 38 [RFC2827]. Additional mitigations in future protocol specification should reduce the amplification factor in request/response payload sizes [drdos-amplification] through the use of padding and consideration of payload data.¶
The author would like to thank Doug Arnold and Hal Murray for contributions to this document, and would like to acknowledge Daniel Franke, Watson Ladd, Miroslav Lichvar for their existing documents and ideas. The author would also like to thank Angelo Moriondo, Franz Karl Achard, and Malcom McLean for providing the author with motivation.¶