Internet-Draft QUIC on Streams February 2024
Oku & Pardue Expires 19 August 2024 [Page]
Workgroup:
QUIC
Internet-Draft:
draft-kazuho-quic-quic-on-streams-00
Published:
Intended Status:
Standards Track
Expires:
Authors:
K. Oku
Fastly
L. Pardue
Cloudflare

QUIC on Streams

Abstract

This document specifies a polyfill of QUIC version 1 that runs on top of bi-directional streams such as TLS.

Discussion Venues

This note is to be removed before publishing as an RFC.

Discussion of this document takes place on the QUIC Working Group mailing list (quic@ietf.org), which is archived at https://mailarchive.ietf.org/arch/browse/quic/.

Source for this draft and an issue tracker can be found at https://github.com/kazuho/draft-kazuho-quic-quic-on-streams.

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 19 August 2024.

Table of Contents

1. Introduction

QUIC version 1 [QUIC] is a bi-directional, authenticated transport-layer protocol built on top of UDP [UDP]. The protocol provides multiplexed flow-controlled streams without head-of-line blocking as a core service. It also offers low-latency connection establishment and efficient loss recovery.

However, there are downsides to QUIC.

One downside is that QUIC, being based on UDP, is not as universally accessible as TCP [TCP], due to occasionally being blocked by middleboxes.

Another downside is that QUIC is computationally more expensive compared to TLS [TLS13] over TCP. This increased cost is partly because QUIC encrypts each packet, which is smaller than the encryption unit of TLS, leading to more overhead, and partly because UDP is less optimized within computing infrastructures.

Due to these limitations, applications are often served using both QUIC and TCP. QUIC is employed to provide the optimal user experience, while TCP acts as a fallback for ensuring network reachability and computational efficiency as needed.

One such example is HTTP, which has different bindings for QUIC (HTTP/3 [HTTP3]) and TCP (HTTP/2 [HTTP2]). Recently, security concerns have prompted proposals to revise HTTP/2 ([h2-stream-limits]), which has sparked discussions about the costs of maintaining multiple HTTP versions.

Another example is WebTransport, a superset of HTTP. Because HTTP has different bindings for QUIC and TCP, WebTransport defines its own extensions for the two HTTP variants ([webtrans-h3], [webtrans-h2]).

To reduce or eliminate the costs associated with duplicated efforts in providing services on top of both transport protocols, this document specifies a polyfill that allows application protocols built on QUIC to run on transport protocols that provide single bi-directional, byte-oriented stream such as TCP or TLS.

The specified polyfill provides a compatibility layer for the set of operations (i.e., API) required by QUIC, as specified in Section 2.4 and Section 5.3 of [QUIC].

2. Conventions and Definitions

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.

3. The Protocol

QUIC on Streams can be used on any transport that provides bi-directional, byte-oriented stream that is ordered and reliable; for details, see Section 3.1.

QUIC frames are sent directly on top of the transport.

The frames are not encrypted. It is the task of the transport (e.g., TLS) to provide confidentially and integrity.

QUIC packet headers are not used.

For exchanging the Transport Parameters, a new frame called QS_TRANSPORT_PARAMETERS frame is defined.

3.1. Properties of Underlying Transport

QUIC on Streams is designed to work on top of transport layer protocols that provide the following capabilities:

In-order delivery of bytes in both direction:

Underlying transport provides a byte-oriented and bi-directional stream that deliver the bytes in order; i.e., bytes that were sent in one order become available to the receiving side in the same order.

Guaranteed delivery:

If the transport runs on top of a lossy network, that transport recovers the bytes lost; e.g., by retransmitting them. This requires buffering and reassembly, in order to achieve the first bullet point (in-order delivery).

Congestion control:

When used on a shared network, the transport is congestion controlled. Implementations of QUIC on Streams simply write outgoing frames to the transport when that transport permits to.

Confidentially and Integrity:

Unless used upon endpoints between which tampering or monitoring is a non-concern, the transport provides confidentially and integrity protection.

TLS over TCP provides all these capabilities.

UNIX sockets are an example that provides only the first two. Congestion control is not employed, as UNIX sockets do not face a shared bottleneck. Confidentiality and integrity protection are deemed unnecessary in environments where the operating system is trusted.

4. QUIC Frames

In QUIC on Streams, the following QUIC frames can be used, as if they were sent or received in the application packet number space:

The frame formats are identical to those in QUIC version 1. Likewise, the meaning and requirements for the use of these frames are consistent with QUIC version 1, with the exception to the specific changes made to the STREAM frames, as detailed in Section 4.1.

Use of other frames defined in QUIC version 1 is prohibited. Namely, ACK frames are not used, because the underlying transport guarantees delivery. Use of frames that communicate Connection IDs and those related to path migration is forbidden.

If an endpoint receives one of the prohibited frames, the endpoint MUST close the connection with an error of type FRAME_ENCODING_ERROR.

4.1. STREAM Frames

While the frame format remains unchanged, there are two differences in the handling of STREAM frames between QUIC version 1 and QUIC on Streams.

4.1.1. STREAM Frames without the Length Field

In QUIC on Streams, when a STREAM frame that omits the Length field is used, the size of that STREAM frame is determined by the maximum frame size, as regulated by the max_frame_size Transport Parameter (Section 5.2).

This behavior contrasts with that of QUIC version 1, where the absence of the Length field implies that the STREAM frame extends to the end of the QUIC packet payload.

This variation arises due to the characteristics of the underlying transports of QUIC on Streams, which may not have, or provide visibility into, the packet boundaries.

4.1.2. Ordering of STREAM frames

For each stream being sent, senders MUST send stream payload in order.

When receiving a STREAM frame that carries a payload not immediately following the payload of the previous STREAM frame for the same Stream ID, receivers MUST close connection with an error of type PROTOCOL_VIOLATION_ERROR.

This change from QUIC version 1 eliminates the need for implementations to buffer and reassemble the stream payload. As a result, the payload being received can be directly passed to the application as it is read from the transport. This efficiency is due to the underlying transport's guarantee of in-order delivery.

These changes do not impact the senders' capability to interleave STREAM frames from multiple streams.

4.2. QS_TRANSPORT_PARAMETERS Frames

In QUIC on Streams, Transport Parameters are exchanged as frames.

QS_TRANSPORT_PARAMETERS frames are formatted as shown in Figure 1.

QS_TRANSPORT_PARAMETERS Frame {
  Type (i) = 0x3f5153300d0a0d0a,
  Length (i),
  Transport Parameters (..),
}
Figure 1: QS_TRANSPORT_PARAMETERS Frame Format

QS_TRANSPORT_PARAMETERS frames contain the following fields:

Length:

A variable-length integer specifying the length of the Transport Parameters field in this QS_TRANSPORT_PARAMETERS frame.

Transport Parameters:

The Transport Parameters. The encoding of the payload is as defined in Section 18 of [QUIC].

The QS_TRANSPORT_PARAMETERS frame is the first frame being sent by endpoints. Endpoints MUST send the QS_TRANSPORT_PARAMETERS frame as soon as the underlying transport becomes available. Note neither endpoint needs to wait for the peer's Transport Parameters before sending its own, as Transport Parameters are a unilateral declaration of an endpoint's capabilities (Section 7.4 of [QUIC]).

If the first frame being received by an endpoint is not a QS_TRANSPORT_PARAMETERS frame, the endpoint MUST close the connection with an error of type TRANSPORT_PARAMETER_ERROR.

The frame type (0x3f5153300d0a0d0a; "\xffQS0\r\n\r\n" on wire) has been chosen so that it can be used to disambiguate QUIC on Streams from HTTP/1.1 [HTTP1] and HTTP/2.

4.3. QS_PING Frames

In QUIC on Streams, QS_PING frames allow endpoints to test peer reachability above the underlying transport.

QS_PING frames are formatted as shown in Figure 2.

QS_PING Frame {
  Type (i) = 0xTBD..0xTBD+1,
  Sequence Number (i),
}
Figure 2: QS_PING Frame Format

Type 0xTBD is used for sending a ping (i.e., request the peer to respond). Type 0xTBD+1 is used in response.

QS_PING frames contain the following fields:

Sequence Number:

A variable-length integer used to identify the ping.

When sending QS_PING frames of type 0xTBD, endpoints MUST send monotonically increasing values in the Sequence Number field, since that allows the endpoints to identify to which ping the peer has responded.

When sending QS_PING frames of type 0xTBD+1 in response, endpoints MUST echo the Sequence Number that they received.

When receiving multiple QS_PING frames of type 0xTBD before having the chance to respond, an endpoint MAY only respond with one QS_PING frame of type 0xTBD+1 carrying the largest Sequence Number that the endpoint has received.

5. Transport Parameters

QUIC on Streams uses a subset of Transport Parameters defined in QUIC version 1. Also, one new Transport Parameter specific to QUIC on Streams is defined.

5.1. Permitted and Forbidden Transport Parameters

In QUIC on Streams, use of the following Transport Parameters is allowed.

  • max_idle_timeout

  • initial_max_data

  • initial_max_stream_data_bidi_local

  • initial_max_stream_data_bidi_remote

  • initial_max_stream_data_uni

  • initial_max_streams_bidi

  • initial_max_streams_uni

The definition of these Transport Parameters are unchanged.

Use of other Transport Parameters defined in QUIC version 1 is prohibited. When an endpoint receives one of the prohibited Transport Parameters, the endpoint MUST close the connection with an error of type TRANSPORT_PARAMETER_ERROR.

Endpoints MUST NOT send Transport Parameters that extend QUIC version 1, unless they are specified to be compatible with QUIC on Streams.

When receiving Transport Parameters not defined in QUIC version 1, receivers MUST ignore them unless they are specified to be usable on QUIC on Streams.

5.2. max_frame_size Transport Parameter

The max_frame_size Transport Parameter (0xTBD) is a variable-length integer specifying the maximum size of the QUIC frame that the peer can send, in the unit of bytes.

The initial value of the max_frame_size Transport Parameter is 16384.

By sending the Transport Parameter, the maximum frame size can only be increased. When receiving a value below the initial value, receivers MUST close the connection with an error of type TRANSPORT_PARAMETER_ERROR.

Endpoints MUST NOT send QUIC frames that exceed the maximum declared by the peer.

When receiving QUIC frames that exceed the declared maximum, receivers MUST close the connection with an error of type FRAME_ENCODING_ERROR.

6. Closing the Connection

As is with QUIC version 1, a connection can be closed either by a CONNECTION_CLOSE frame or by an idle timeout.

Unlike QUIC version 1, there is no draining period; once an endpoint sends or receives the CONNECTION_CLOSE frame or reaches the idle timeout, all the resources allocated for the Service are freed and the underlying transport is closed immediately.

7. Using 0-RTT

TLS 1.3 introduced the concept of early data (also knows as 0-RTT data).

When using QUIC on Streams on top of TLS that supports early data, clients MAY use early data when resuming a connection, by reusing certain Transport Parameters as defined in Section 7.4.1 of [QUIC].

Similarly, when accepting early data, the servers MUST send Transport Parameters that obey to the restrictions defined in Section 7.4.1 of [QUIC].

8. Extensions

Not all the extensions of QUIC version 1 can be used. Each extension have to define its mapping for QUIC on Streams, or explicitly allow the use; see Section 5.1.

As is the case with QUIC version 1, use of extension frames have to be negotiated before use; see Section 19.21 of [QUIC].

This specification defines the mapping of the Unreliable Datagram Extension.

8.1. Unreliable Datagram Extension

The use of the Unreliable Datagram Extension [QUIC_DATAGRAM] is permitted, with one modification:

Similar to STREAM frames, when employing DATAGRAM frames of type 0x30 (i.e., DATAGRAM frames without the Length field), their size is determined by the max_frame_size Transport Parameter (Section 5.2).

Apart from this, the encoding and semantics of the Unreliable Datagram Extension remain unchanged. The use of the extension is negotiated via the Transport Parameters.

As discussed in Section 5 of [QUIC_DATAGRAM], senders can drop DATAGRAM frames if the transport is blocked by flow or congestion control.

9. Version Agility

Unlike QUIC, QUIC on Streams does not define a mechanism for version negotiation.

In large-scale deployments requiring service and protocol version discovery, QUIC on Streams can and is likely to be implemented over TLS. The Application-Layer Protocol Negotiation Extension of TLS [ALPN] is the favored mechanism to negotiate between an application protocol based on this specification and others.

When ALPN is unavailable, first 8 bytes exchanged on the transport (i.e., the type field of the QS_TRANSPORT_PARAMETERS frame in the encoded form) can be used to identify if QUIC on Streams is in use.

10. Implementation Considerations

Similar to HTTP/3 with Extensible Priorities [HTTP_PRIORITY], application protocols using QUIC may employ stream multiplexing along with a system to tune the delivery sequence of QUIC streams.

To alternate between QUIC streams of varying priorities in a timely manner, it is advisable for QUIC on Streams implementations to avoid creating deep buffers holding QUIC frames. Instead, endpoints should wait for the transport layer to be ready for writing. Upon becoming writable, they should write QUIC frames according to the latest prioritization signals.

Additionally, implementations may consider monitoring or adjusting the flow and congestion control parameters of the underlying transport. This approach aims to minimize data buffering within the transport layer before transmission. However, improper adjustment of these parameters could potentially lead to lower throughput.

11. Security Considerations

TODO Security

12. IANA Considerations

TODO

13. References

13.1. Normative References

[QUIC]
Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based Multiplexed and Secure Transport", RFC 9000, DOI 10.17487/RFC9000, , <https://www.rfc-editor.org/rfc/rfc9000>.
[QUIC_DATAGRAM]
Pauly, T., Kinnear, E., and D. Schinazi, "An Unreliable Datagram Extension to QUIC", RFC 9221, DOI 10.17487/RFC9221, , <https://www.rfc-editor.org/rfc/rfc9221>.
[RFC2119]
Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, , <https://www.rfc-editor.org/rfc/rfc2119>.
[RFC8174]
Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, , <https://www.rfc-editor.org/rfc/rfc8174>.
[TLS13]
Rescorla, E., "The Transport Layer Security (TLS) Protocol Version 1.3", RFC 8446, DOI 10.17487/RFC8446, , <https://www.rfc-editor.org/rfc/rfc8446>.

13.2. Informative References

[ALPN]
Friedl, S., Popov, A., Langley, A., and E. Stephan, "Transport Layer Security (TLS) Application-Layer Protocol Negotiation Extension", RFC 7301, DOI 10.17487/RFC7301, , <https://www.rfc-editor.org/rfc/rfc7301>.
[h2-stream-limits]
Thomson, M. and L. Pardue, "Using HTTP/3 Stream Limits in HTTP/2", Work in Progress, Internet-Draft, draft-thomson-httpbis-h2-stream-limits-00, , <https://datatracker.ietf.org/doc/html/draft-thomson-httpbis-h2-stream-limits-00>.
[HTTP1]
Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, Ed., "HTTP/1.1", STD 99, RFC 9112, DOI 10.17487/RFC9112, , <https://www.rfc-editor.org/rfc/rfc9112>.
[HTTP2]
Thomson, M., Ed. and C. Benfield, Ed., "HTTP/2", RFC 9113, DOI 10.17487/RFC9113, , <https://www.rfc-editor.org/rfc/rfc9113>.
[HTTP3]
Bishop, M., Ed., "HTTP/3", RFC 9114, DOI 10.17487/RFC9114, , <https://www.rfc-editor.org/rfc/rfc9114>.
[HTTP_PRIORITY]
Oku, K. and L. Pardue, "Extensible Prioritization Scheme for HTTP", RFC 9218, DOI 10.17487/RFC9218, , <https://www.rfc-editor.org/rfc/rfc9218>.
[TCP]
Eddy, W., Ed., "Transmission Control Protocol (TCP)", STD 7, RFC 9293, DOI 10.17487/RFC9293, , <https://www.rfc-editor.org/rfc/rfc9293>.
[UDP]
Postel, J., "User Datagram Protocol", STD 6, RFC 768, DOI 10.17487/RFC0768, , <https://www.rfc-editor.org/rfc/rfc768>.
[webtrans-h2]
Frindell, A., Kinnear, E., Pauly, T., Thomson, M., Vasiliev, V., and G. Xie, "WebTransport over HTTP/2", Work in Progress, Internet-Draft, draft-ietf-webtrans-http2-07, , <https://datatracker.ietf.org/doc/html/draft-ietf-webtrans-http2-07>.
[webtrans-h3]
Frindell, A., Kinnear, E., and V. Vasiliev, "WebTransport over HTTP/3", Work in Progress, Internet-Draft, draft-ietf-webtrans-http3-08, , <https://datatracker.ietf.org/doc/html/draft-ietf-webtrans-http3-08>.

Acknowledgments

TODO acknowledge.

Authors' Addresses

Kazuho Oku
Fastly
Lucas Pardue
Cloudflare