]> Akamai

rmarx@akamai.com

Meta

lniccolini@meta.com

martenseemann@gmail.com

Cloudflare

lucas@lucaspardue.com

Transport QUIC Internet-Draft This document describes concrete qlog event definitions and their metadata for QUIC events. These events can then be embedded in the higher level schema defined in . Note to Readers

• Note to RFC editor: Please remove this section before publication.

Feedback and discussion are welcome at https://github.com/quicwg/qlog. Readers are advised to refer to the "editor's draft" at that URL for an up-to-date version of this document. Concrete examples of integrations of this schema in various programming languages can be found at https://github.com/quiclog/qlog/.

Introduction This document describes the values of the qlog name ("category" + "event") and "data" fields and their semantics for the QUIC protocol (see , , and ) and some of its extensions (see and ).

Notational Conventions 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 when, and only when, they appear in all capitals, as shown here. The event and data structure definitions in ths document are expressed in the Concise Data Definition Language and its extensions described in . The following fields from are imported and used: name, category, type, data, group_id, protocol_type, importance, RawInfo, and time-related fields. As is the case for , the qlog schema definitions in this document are intentionally agnostic to serialization formats. The choice of format is an implementation decision.

Overview This document describes how the QUIC protocol can be expressed in qlog using the schema defined in . QUIC protocol events are defined with a category, a name (the concatenation of "category" and "event"), an "importance", an optional "trigger", and "data" fields. Some data fields use complex datastructures. These are represented as enums or re-usable definitions, which are grouped together on the bottom of this document for clarity. When any event from this document is included in a qlog trace, the protocol_type qlog array field MUST contain an entry with the value "QUIC". When the qlog group_id field is used, it is recommended to use QUIC's Original Destination Connection ID (ODCID, the CID chosen by the client when first contacting the server), as this is the only value that does not change over the course of the connection and can be used to link more advanced QUIC packets (e.g., Retry, Version Negotiation) to a given connection. Similarly, the ODCID should be used as the qlog filename or file identifier, potentially suffixed by the vantagepoint type (For example, abcd1234_server.qlog would contain the server-side trace of the connection with ODCID abcd1234).

Raw packet and frame information QUIC packets always include an AEAD authentication tag at the end. In general, the length of the AEAD tag depends on the TLS cipher suite, although all cipher suites used in QUIC v1 use a 16 byte tag.

Note:

As QUIC appends an authentication tag after the packet payload, the packet header_lengths can be calculated as:

header_length = length - payload_length - 16

For UDP datagrams, the calculation is simpler:

header_length = length - payload_length

Note:

In some cases, the length fields are also explicitly reflected inside of packet headers. For example, the QUIC STREAM frame has a length field indicating its payload size. Similarly, the QUIC Long Header has a length field which is equal to the payload length plus the packet number length. In these cases, those fields are intentionally preserved in the event definitions. Even though this can lead to duplicate data when the full RawInfo is logged, it allows a more direct mapping of the QUIC specifications to qlog, making it easier for users to interpret.

Events not belonging to a single connection A single qlog event trace is typically associated with a single QUIC connection. However, for several types of events (for example, a event with trigger value of connection_unknown), it can be impossible to tie them to a specific QUIC connection, especially on the server. There are various ways to handle these events, each making certain tradeoffs between file size overhead, flexibility, ease of use, or ease of implementation. Some options include:

• Log them in a separate endpoint-wide trace (or use a special group_id value) not associated with a single connection.
• Log them in the most recently used trace.
• Use additional heuristics for connection identification (for example use the four-tuple in addition to the Connection ID).
• Buffer events until they can be assigned to a connection (for example for version negotiation and retry events).

QUIC Event Overview QUIC connections consist of different phases and interaction events. In order to model this, QUIC event types are divided into general categories: connectivity (), security (), quic , and recovery . As described in , the qlog name field is the concatenation of category and type. summarizes the name value of each event type that is defined in this specification. QUIC Events

Name value	Importance	Definition
connectivity:server_listening	Extra
connectivity:connection_started	Base
connectivity:connection_closed	Base
connectivity:connection_id_updated	Base
connectivity:spin_bit_updated	Base
connectivity:connection_state_updated	Base
connectivity:path_assigned	Base
connectivity:mtu_updated	Extra
quic:version_information	Core
quic:alpn_information	Core
quic:parameters_set	Core
quic:parameters_restored	Base
quic:packet_sent	Core
quic:packet_received	Core
quic:packet_dropped	Base
quic:packet_buffered	Base
quic:packets_acked	Extra
quic:datagrams_sent	Extra
quic:datagrams_received	Extra
quic:datagram_dropped	Extra
quic:stream_state_updated	Base
quic:frames_processed	Extra
quic:stream_data_moved	Base
quic:datagram_data_moved	Base
quic:migration_state_updated	Extra
security:key_updated	Base
security:key_discarded	Base
recovery:parameters_set	Base
recovery:metrics_updated	Core
recovery:congestion_state_updated	Base
recovery:loss_timer_updated	Extra
recovery:packet_lost	Core
recovery:marked_for_retransmit	Extra
recovery:ecn_state_updated	Extra

QUIC events extend the $ProtocolEventData extension point defined in .

QuicEventData definition and ProtocolEventData extension

Connectivity events

server_listening Emitted when the server starts accepting connections. It has Extra importance level; see .

ConnectivityServerListening definition

Some QUIC stacks do not handle sockets directly and are thus unable to log IP and/or port information.

connection_started The connection_started event is used for both attempting (client-perspective) and accepting (server-perspective) new connections. Note that while there is overlap with the connection_state_updated event, this event is separate event in order to capture additional data that can be useful to log. It has Base importance level; see .

ConnectivityConnectionStarted definition

Some QUIC stacks do not handle sockets directly and are thus unable to log IP and/or port information.

connection_closed The connection_closed event is used for logging when a connection was closed, typically when an error or timeout occurred. It has Base importance level; see . Note that this event has overlap with the connection_state_updated event, as well as the CONNECTION_CLOSE frame. However, in practice, when analyzing large deployments, it can be useful to have a single event representing a connection_closed event, which also includes an additional reason field to provide more information. Furthermore, it is useful to log closures due to timeouts, which are difficult to reflect using the other options. In QUIC there are two main connection-closing error categories: connection and application errors. They have well-defined error codes and semantics. Next to these however, there can be internal errors that occur that may or may not get mapped to the official error codes in implementation-specific ways. As such, multiple error codes can be set on the same event to reflect this.

ConnectivityConnectionClosed definition

connection_id_updated The connection_id_updated event is emitted when either party updates their current Connection ID. As this typically happens only sparingly over the course of a connection, using this event is more efficient than logging the observed CID with each and every packet_sent or packet_received events. It has Base importance level; see . The connection_id_updated event is viewed from the perspective of the endpoint applying the new ID. As such, when the endpoint receives a new connection ID from the peer, the owner field will be "remote". When the endpoint updates its own connection ID, the owner field will be "local".

ConnectivityConnectionIDUpdated definition

spin_bit_updated The spin_bit_updated event conveys information about the QUIC latency spin bit; see . The event is emitted when the spin bit changes value, it SHOULD NOT be emitted if the spin bit is set without changing its value. It has Base importance level; see .

ConnectivitySpinBitUpdated definition

connection_state_updated The connection_state_updated event is used to track progress through QUIC's complex handshake and connection close procedures. It has Base importance level; see . It is intended to provide exhaustive options to log each state individually, but also provides a more basic, simpler set for implementations less interested in tracking each smaller state transition. As such, users should not expect to see all these states reflected in all qlogs and implementers should focus on support for the SimpleConnectionState set.

ConnectivityConnectionStateUpdated definition

These states correspond to the following transitions for both client and server: Client:

• send initial
  • state = attempted
• get initial
  • state = validated (not really "needed" at the client, but somewhat useful to indicate progress nonetheless)
• get first Handshake packet
  • state = handshake_started
• get Handshake packet containing ServerFinished
  • state = handshake_complete
• send ClientFinished
  • state = early_write (1RTT can now be sent)
• get HANDSHAKE_DONE
  • state = handshake_confirmed

Server:

• get initial
  • state = attempted
• send initial (TODO don't think this needs a separate state, since some handshake will always be sent in the same flight as this?)
• send handshake EE, CERT, CV, ...
  • state = handshake_started
• send ServerFinished
  • state = early_write (1RTT can now be sent)
• get first handshake packet / something using a server-issued CID of min length
  • state = validated
• get handshake packet containing ClientFinished
  • state = handshake_complete
• send HANDSHAKE_DONE
  • state = handshake_confirmed

Note:

connection_state_changed with a new state of attempted is the same conceptual event as the connection_started event above from the client's perspective. Similarly, a state of closing or draining corresponds to the connection_closed` event.

path_assigned Importance: Base This event is used to associate a single PathID's value with other parameters that describe a unique network path. As described in , each qlog event can be linked to a single network path by means of the top-level "path" field, whose value is a PathID. However, since it can be cumbersome to encode additional path metadata (such as IP addresses or Connection IDs) directly into the PathID, this event allows such an association to happen separately. As such, PathIDs can be short and unique, and can even be updated to be associated with new metadata as the connection's state evolves. Definition:

ConnectivityPathAssigned definition

Choosing the different path_id values is left up to the implementation. Some options include using a uniquely incrementing integer, using the (first) Destination Connection ID associated with a path (or its sequence number), or using (a hash of) the two endpoint IP addresses. It is important to note that the empty string ("") is a valid PathID and that it is the default assigned to events that do not explicitly set a "path" field. Put differently, the initial path of a QUIC connection on which the handshake occurs (see also ) is implicitly associated with the PathID with value "". Associating metadata with this default path is possible by logging the ConnectivityPathAssigned event with a value of "" for the path_id field. As paths and their metadata can evolve over time, multiple ConnectivityPathAssigned events can be emitted for each unique PathID. The latest event contains the most up-to-date information for that PathID. As such, the first time a PathID is seen in a ConnectivityPathAssigned event, it is an indication that the path is created. Subsequent occurrences indicate the path is updated, while a final occurrence with both path_local and path_remote fields omitted implicitly indicates the path has been abandoned.

mtu_updated The mtu_updated event indicates that the estimated Path MTU was updated. This happens as part of the Path MTU discovery process. It has Extra importance level; see .

ConnectivityMTUUpdated definition

QUIC events

version_information The version_information event supports QUIC version negotiation; see . It has Core importance level; see . QUIC endpoints each have their own list of QUIC versions they support. The client uses the most likely version in their first initial. If the server does not support that version, it replies with a Version Negotiation packet, which contains its supported versions. From this, the client selects a version. The version_information event aggregates all this information in a single event type. It also allows logging of supported versions at an endpoint without actual version negotiation needing to happen.

QUICVersionInformation definition

Intended use:

• When sending an initial, the client logs this event with client_versions and chosen_version set
• Upon receiving a client initial with a supported version, the server logs this event with server_versions and chosen_version set
• Upon receiving a client initial with an unsupported version, the server logs this event with server_versions set and client_versions to the single-element array containing the client's attempted version. The absence of chosen_version implies no overlap was found
• Upon receiving a version negotiation packet from the server, the client logs this event with client_versions set and server_versions to the versions in the version negotiation packet and chosen_version to the version it will use for the next initial packet. If the client receives a set of server_versions with no viable overlap with its own supported versions, this event should be logged without the chosen_version set

alpn_information The alpn_information event supports Application-Layer Protocol Negotiation (ALPN) over the QUIC transport; see and . It has Core importance level; see . QUIC endpoints are configured with a list of supported ALPN identifiers. Clients send the list in a TLS ClientHello, and servers match against their list. On success, a single ALPN identifier is chosen and sent back in a TLS ServerHello. If no match is found, the connection is closed. ALPN identifiers are byte sequences, that may be possible to present as UTF-8. The ALPNIdentifier` type supports either format. Implementations SHOULD log at least one format, but MAY log both or none.

QUICALPNInformation definition

Intended use:

• When sending an initial, the client logs this event with client_alpns set
• When receiving an initial with a supported alpn, the server logs this event with server_alpns set, client_alpns equalling the client-provided list, and chosen_alpn to the value it will send back to the client.
• When receiving an initial with an alpn, the client logs this event with chosen_alpn to the received value.
• Alternatively, a client can choose to not log the first event, but wait for the receipt of the server initial to log this event with both client_alpns and chosen_alpn set.

parameters_set The parameters_set event groups settings from several different sources (transport parameters, TLS ciphers, etc.) into a single event. This is done to minimize the amount of events and to decouple conceptual setting impacts from their underlying mechanism for easier high-level reasoning. The event has Core importance level; see . Most of these settings are typically set once and never change. However, they are usually set at different times during the connection, so there will regularly be several instances of this event with different fields set. Note that some settings have two variations (one set locally, one requested by the remote peer). This is reflected in the owner field. As such, this field MUST be correct for all settings included a single event instance. If you need to log settings from two sides, you MUST emit two separate event instances. In the case of connection resumption and 0-RTT, some of the server's parameters are stored up-front at the client and used for the initial connection startup. They are later updated with the server's reply. In these cases, utilize the separate parameters_restored event to indicate the initial values, and this event to indicate the updated values, as normal.

QUICParametersSet definition

The generic $$quic-parametersset-extension is defined here as a CDDL extension point (a "group socket"). It can be used to support additional, unknown, custom, and greased parameters. An example of such an extension can be found in .

quic-parametersset-extension example

parameters_restored When using QUIC 0-RTT, clients are expected to remember and restore the server's transport parameters from the previous connection. The parameters_restored event is used to indicate which parameters were restored and to which values when utilizing 0-RTT. It has Base importance level; see . Note that not all transport parameters should be restored (many are even prohibited from being re-utilized). The ones listed here are the ones expected to be useful for correct 0-RTT usage.

QUICParametersRestored definition

The generic $$quic-parametersrestored-extension is defined here as a CDDL extension point (a "group socket"). It can be used to support additional and custom parameters.

packet_sent The packet_sent event indicates a QUIC-level packet was sent. It has Core importance level; see .

QUICPacketSent definition

The encryption_level and packet_number_space are not logged explicitly: the header.packet_type specifies this by inference (assuming correct implementation) For more details on datagram_id, see . It is only needed when keeping track of packet coalescing.

packet_received The packet_received event indicates a QUIC-level packet was received. It has Core importance level; see .

QUICPacketReceived definition

The encryption_level and packet_number_space are not logged explicitly: the header.packet_type specifies this by inference (assuming correct implementation) For more details on datagram_id, see . It is only needed when keeping track of packet coalescing.

packet_dropped The packet_dropped event indicates a QUIC-level packet was dropped. It has Base importance level; see . The trigger field indicates a general reason category for dropping the packet, while the details field can contain additional implementation-specific information.

QUICPacketDropped definition any} ? trigger: "internal_error" / "rejected" / "unsupported" / "invalid" / "duplicate" / "connection_unknown" / "decryption_failure" / "key_unavailable" / "general" } ]]>

Some example situations for each of the trigger categories include:

• internal_error: not initialized, out of memory
• rejected: limits reached, DDoS protection, unwilling to track more paths, duplicate packet
• unsupported: unknown or unsupported version. See also .
• invalid: packet parsing or validation error
• duplicate: duplicate packet
• connection_unknown: packet does not relate to a known connection or Connection ID
• decryption_failure: decryption failed
• key_unavailable: decryption key was unavailable
• general: situations not clearly covered in the other categories

For more details on datagram_id, see .

packet_buffered The packet_buffered event is emitted when a packet is buffered because it cannot be processed yet. Typically, this is because the packet cannot be parsed yet, and thus only the full packet contents can be logged when it was parsed in a packet_received event. The event has Base importance level; see .

QUICPacketBuffered definition

For more details on datagram_id, see . It is only needed when keeping track of packet coalescing.

packets_acked The packets_acked event is emitted when a (group of) sent packet(s) is acknowledged by the remote peer for the first time. It has Extra importance level; see . This information could also be deduced from the contents of received ACK frames. However, ACK frames require additional processing logic to determine when a given packet is acknowledged for the first time, as QUIC uses ACK ranges which can include repeated ACKs. Additionally, this event can be used by implementations that do not log frame contents.

QUICPacketsAcked definition

If packet_number_space is omitted, it assumes the default value of application_data, as this is by far the most prevalent packet number space a typical QUIC connection will use.

datagrams_sent When one or more UDP-level datagrams are passed to the socket. This is useful for determining how QUIC packet buffers are drained to the OS. The event has Extra importance level; see .

QUICDatagramsSent definition

Since QUIC implementations rarely control UDP logic directly, the raw data excludes UDP-level headers in all fields. The datagram_id is a qlog-specific concept to allow tracking of QUIC packet coalescing inside UDP datagrams. Since QUIC generates many UDP datagrams, unique identifiers are required to be able to track them individually in qlog traces. However, neither UDP nor QUIC exchanges datagram identifiers on the wire. Selecting identifier values is thus left to qlog implementations, which should consider how to generate unique values within the scope of their created traces.

datagrams_received When one or more UDP-level datagrams are received from the socket. This is useful for determining how datagrams are passed to the user space stack from the OS. The event has Extra importance level; see .

QUICDatagramsReceived definition

For more details on datagram_ids, see .

datagram_dropped When a UDP-level datagram is dropped. This is typically done if it does not contain a valid QUIC packet. If it does, but the QUIC packet is dropped for other reasons, the packet_dropped event () should be used instead. The event has Extra importance level; see .

QUICDatagramDropped definition

stream_state_updated The stream_state_updated event is emitted whenever the internal state of a QUIC stream is updated; see . Most of this can be inferred from several types of frames going over the wire, but it's much easier to have explicit signals for these state changes. The event has Base importance level; see .

QUICStreamStateUpdated definition

QUIC implementations SHOULD mainly log the simplified bidirectional (HTTP/2-alike) stream states (e.g., idle, open, closed) instead of the more fine-grained stream states (e.g., data_sent, reset_received). These latter ones are mainly for more in-depth debugging. Tools SHOULD be able to deal with both types equally.

frames_processed The frame_processed event is intended to prevent a large proliferation of specific purpose events (e.g., packets_acknowledged, flow_control_updated, stream_data_received). It has Extra importance level; see . Implementations have the opportunity to (selectively) log this type of signal without having to log packet-level details (e.g., in packet_received). Since for almost all cases, the effects of applying a frame to the internal state of an implementation can be inferred from that frame's contents, these events are aggregated into this single frames_processed event. The frame_processed event can be used to signal internal state change not resulting directly from the actual "parsing" of a frame (e.g., the frame could have been parsed, data put into a buffer, then later processed, then logged with this event). The packet_received event can convey all constituent frames. It is not expected that the frames_processed event will also be used for a redundant purpose. Rather, implementations can use this event to avoid having to log full packets or to convey extra information about when frames are processed (for example, if frame processing is deferred for any reason). Note that for some events, this approach will lose some information (e.g., for which encryption level are packets being acknowledged?). If this information is important, the packet_received event can be used instead. In some implementations, it can be difficult to log frames directly, even when using packet_sent and packet_received events. For these cases, the frames_processed event also contains the packet_numbers field, which can be used to more explicitly link this event to the packet_sent/received events. The field is an array, which supports using a single frames_processed event for multiple frames received over multiple packets. To map between frames and packets, the position and order of entries in the frames and packet_numbers is used. If the optional packet_numbers field is used, each frame MUST have a corresponding packet number at the same index.

QUICFramesProcessed definition

For example, an instance of the frames_processed event that represents four STREAM frames received over two packets would have the fields serialized as:

stream_data_moved The stream_data_moved event is used to indicate when QUIC stream data moves between the different layers. This helps make clear the flow of data, how long data remains in various buffers, and the overheads introduced by individual layers. The event has Base importance level; see . For example, it can be useful to understand when when data moves from an application protocol (e.g., HTTP) to QUIC stream buffers and vice versa. Similarly, when data moves from the application protocol layer into a user-facing application such as a web browser. The stream_data_moved event can provide insight into whether received data on a QUIC stream is moved to the application protocol immediately (for example per received packet) or in larger batches (for example, all QUIC packets are processed first and afterwards the application layer reads from the streams with newly available data). This can help identify bottlenecks, flow control issues, or scheduling problems. This event is only for data in QUIC streams. For data in QUIC Datagram Frames, see the datagram_data_moved event defined in .

QUICStreamDataMoved definition

datagram_data_moved The datagram_data_moved event is used to indicate when QUIC Datagram Frame data (see ) moves between the different layers. This helps make clear the flow of data, how long data remains in various buffers, and the overheads introduced by individual layers. The event has Base importance level; see . For example, passing from the application protocol (e.g., WebTransport) to QUIC Datagram Frame buffers and vice versa. Similarly, when data moves from the application protocol layer into a user-facing application such as a web browser. The datagram_data_moved event can provide insight into whether received data in a QUIC Datagram Frame is moved to the application protocol immediately (for example per received packet) or in larger batches (for example, all QUIC packets are processed first and afterwards the application layer reads all Datagrams at once). This can help identify bottlenecks, flow control issues, or scheduling problems. This event is only for data in QUIC Datagram Frames. For data in QUIC streams, see the stream_data_moved event defined in .

QUICDatagramDataMoved definition

migration_state_updated Importance: Extra Use to provide additional information when attempting (client-side) connection migration. While most details of the QUIC connection migration process can be inferred by observing the PATH_CHALLENGE and PATH_RESPONSE frames, in combination with the ConnectivityPathAssigned event, it can be useful to explicitly log the progression of the migration and potentially made decisions in a single location/event. Generally speaking, connection migration goes through two phases: a probing phase (which is not always needed/present), and a migration phase (which can be abandoned upon error). Implementations that log per-path information in a QUICMigrationStateUpdated, SHOULD also emit QUICPathAssigned events, to serve as a ground-truth source of information. Definition:

QUICMigrationStateUpdated definition

Security Events

key_updated The key_updated event has Base importance level; see .

SecurityKeyUpdated definition

Note that the key_phase is the full value of the key phase (as indicated by @M and @N in Figure 9 of ). The key phase bit used on the packet header is the least significant bit of the key phase.

key_discarded The key_discarded event has Base importance level; see .

SecurityKeyDiscarded definition

Recovery events Most of the events in this category are kept generic to support different recovery approaches and various congestion control algorithms. Tool creators SHOULD make an effort to support and visualize even unknown data in these events (e.g., plot unknown congestion states by name on a timeline visualization).

parameters_set The parameters_set event groups initial parameters from both loss detection and congestion control into a single event. It has Base importance level; see . All these settings are typically set once and never change. Implementation that do, for some reason, change these parameters during execution, MAY emit the parameters_set event more than once.

RecoveryParametersSet definition

Additionally, this event can contain any number of unspecified fields to support different recovery approaches.

metrics_updated The metrics_updated event is emitted when one or more of the observable recovery metrics changes value. It has Core importance level; see . This event SHOULD group all possible metric updates that happen at or around the same time in a single event (e.g., if min_rtt and smoothed_rtt change at the same time, they should be bundled in a single metrics_updated entry, rather than split out into two). Consequently, a metrics_updated event is only guaranteed to contain at least one of the listed metrics.

RecoveryMetricsUpdated definition

In order to make logging easier, implementations MAY log values even if they are the same as previously reported values (e.g., two subsequent RecoveryMetricsUpdated entries can both report the exact same value for min_rtt). However, applications SHOULD try to log only actual updates to values. Additionally, the metrics_updated event can contain any number of unspecified fields to support different recovery approaches.

congestion_state_updated The congestion_state_updated event signifies when the congestion controller enters a significant new state and changes its behaviour. It has Base importance level; see . The event is generic to support different Congestion Control algorithms. For example, for the algorithm defined in the Recovery draft ("enhanced" New Reno), the following states are defined:

• slow_start
• congestion_avoidance
• application_limited
• recovery

RecoveryCongestionStateUpdated definition

The trigger field SHOULD be logged if there are multiple ways in which a state change can occur but MAY be omitted if a given state can only be due to a single event occurring (e.g., slow start is exited only when ssthresh is exceeded).

loss_timer_updated The loss_timer_updated event is emitted when a recovery loss timer changes state. It has Extra importance level; see . The three main event types are:

• set: the timer is set with a delta timeout for when it will trigger next
• expired: when the timer effectively expires after the delta timeout
• cancelled: when a timer is cancelled (e.g., all outstanding packets are acknowledged, start idle period)

In order to indicate an active timer's timeout update, a new set event is used.

RecoveryLossTimerUpdated definition

packet_lost The packet_lost event is emitted when a packet is deemed lost by loss detection. It has Core importance level; see . It is RECOMMENDED to populate the optional trigger field in order to help disambiguate among the various possible causes of a loss declaration.

RecoveryPacketLost definition

marked_for_retransmit The marked_for_retransmit event indicates which data was marked for retransmission upon detection of packet loss (see packet_lost). It has Extra importance level; see . Similar to the reasoning for the frames_processed event, in order to keep the amount of different events low, this signal is grouped into in a single event based on existing QUIC frame definitions for all types of retransmittable data. Implementations retransmitting full packets or frames directly can just log the constituent frames of the lost packet here (or do away with this event and use the contents of the packet_lost event instead). Conversely, implementations that have more complex logic (e.g., marking ranges in a stream's data buffer as in-flight), or that do not track sent frames in full (e.g., only stream offset + length), can translate their internal behaviour into the appropriate frame instance here even if that frame was never or will never be put on the wire. Much of this data can be inferred if implementations log packet_sent events (e.g., looking at overlapping stream data offsets and length, one can determine when data was retransmitted).

RecoveryMarkedForRetransmit definition

ecn_state_updated The ecn_state_updated event indicates a progression in the ECN state machine as described in section A.4 of . It has Extra importance level; see .

ECNStateUpdated definition

QUIC data field definitions

QuicVersion

QuicVersion definition

ConnectionID

ConnectionID definition

Owner

Owner definition

IPAddress and IPVersion

IPAddress definition

IPVersion definition

PathEndpointInfo PathEndpointInfo indicates a single half/direction of a path. A full path is comprised of two halves. Firstly: the server sends to the remote client IP + port using a specific destination Connection ID. Secondly: the client sends to the remote server IP + port using a different destination Connection ID. As such, structures logging path information SHOULD include two different PathEndpointInfo instances, one for each half of the path.

PathEndpointInfo definition

PacketType

PacketType definition

PacketNumberSpace

PacketNumberSpace definition

PacketHeader

PacketHeader definition

Token

Token definition any } ? raw: RawInfo } ]]>

The token carried in an Initial packet can either be a retry token from a Retry packet, or one originally provided by the server in a NEW_TOKEN frame used when resuming a connection (e.g., for address validation purposes). Retry and resumption tokens typically contain encoded metadata to check the token's validity when it is used, but this metadata and its format is implementation specific. For that, Token includes a general-purpose details field.

Stateless Reset Token

Stateless Reset Token definition

The stateless reset token is carried in stateless reset packets, in transport parameters and in NEW_CONNECTION_ID frames.

KeyType

KeyType definition

ECN

ECN definition

The ECN bits carried in the IP header.

QUIC Frames The generic $QuicFrame is defined here as a CDDL extension point (a "socket" or "plug"). It can be extended to support additional QUIC frame types.

QuicFrame plug definition any } ]]>

The QUIC frame types defined in this document are as follows:

QuicBaseFrames definition

PaddingFrame In QUIC, PADDING frames are simply identified as a single byte of value 0. As such, each padding byte could be theoretically interpreted and logged as an individual PaddingFrame. However, as this leads to heavy logging overhead, implementations SHOULD instead emit just a single PaddingFrame and set the payload_length property to the amount of PADDING bytes/frames included in the packet.

PaddingFrame definition

PingFrame

PingFrame definition

AckFrame

AckFrame definition

Note that the packet ranges in AckFrame.acked_ranges do not necessarily have to be ordered (e.g., [[5,9],[1,4]] is a valid value). Note that the two numbers in the packet range can be the same (e.g., [120,120] means that packet with number 120 was ACKed). However, in that case, implementers SHOULD log [120] instead and tools MUST be able to deal with both notations.

ResetStreamFrame

ResetStreamFrame definition

StopSendingFrame

StopSendingFrame definition

CryptoFrame

CryptoFrame definition

NewTokenFrame

NewTokenFrame definition

StreamFrame

StreamFrame definition

MaxDataFrame

MaxDataFrame definition

MaxStreamDataFrame

MaxStreamDataFrame definition

MaxStreamsFrame

MaxStreamsFrame definition

DataBlockedFrame

DataBlockedFrame definition

StreamDataBlockedFrame

StreamDataBlockedFrame definition

StreamsBlockedFrame

StreamsBlockedFrame definition

NewConnectionIDFrame

NewConnectionIDFrame definition

RetireConnectionIDFrame

RetireConnectionIDFrame definition

PathChallengeFrame

PathChallengeFrame definition

PathResponseFrame

PathResponseFrame definition

ConnectionCloseFrame An endpoint that receives unknown error codes can record it in the error_code field using the numerical value without variable-length integer encoding. When the connection is closed due a connection-level error, the trigger_frame_type field can be used to log the frame that triggered the error. For known frame types, the appropriate string value is used. For unknown frame types, the numerical value without variable-length integer encoding is used.

ConnectionCloseFrame definition

HandshakeDoneFrame

HandshakeDoneFrame definition

UnknownFrame The frame_type_value field is the numerical value without VLIE encoding.

UnknownFrame definition

DatagramFrame The QUIC DATAGRAM frame is defined in .

DatagramFrame definition

TransportError

TransportError definition

ApplicationError By definition, an application error is defined by the application-level protocol running on top of QUIC (e.g., HTTP/3). As such, it cannot be defined here directly. Applications MAY use the provided extension point through the use of the CDDL "socket" mechanism. Application-level qlog definitions that wish to define new ApplicationError strings MUST do so by extending the $ApplicationError socket as such:

CryptoError These errors are defined in the TLS document as "A TLS alert is turned into a QUIC connection error by converting the one-byte alert description into a QUIC error code. The alert description is added to 0x100 to produce a QUIC error code from the range reserved for CRYPTO_ERROR." This approach maps badly to a pre-defined enum. As such, the crypto_error string is defined as having a dynamic component here, which should include the hex-encoded and zero-padded value of the TLS alert description.

CryptoError definition

Security and Privacy Considerations The security and privacy considerations discussed in apply to this document as well.

IANA Considerations There are no IANA considerations.

References Normative References Akamai Meta Protocol Labs Cloudflare This document defines qlog, an extensible high-level schema for a standardized logging format. It allows easy sharing of data, benefitting common debug and analysis methods and tooling. The high- level schema is independent of protocol; separate documents extend qlog for protocol-specific data. The schema is also independent of serialization format, allowing logs to be represented in many ways such as JSON, CSV, or protobuf. Note to Readers Note to RFC editor: Please remove this section before publication. Feedback and discussion are welcome at https://github.com/quicwg/qlog (https://github.com/quicwg/qlog). Readers are advised to refer to the "editor's draft" at that URL for an up-to-date version of this document. Concrete examples of integrations of this schema in various programming languages can be found at https://github.com/quiclog/ qlog/ (https://github.com/quiclog/qlog/). This document defines the core of the QUIC transport protocol. QUIC provides applications with flow-controlled streams for structured communication, low-latency connection establishment, and network path migration. QUIC includes security measures that ensure confidentiality, integrity, and availability in a range of deployment circumstances. Accompanying documents describe the integration of TLS for key negotiation, loss detection, and an exemplary congestion control algorithm. This document describes loss detection and congestion control mechanisms for QUIC. This document describes how Transport Layer Security (TLS) is used to secure QUIC. This document defines an extension to the QUIC transport protocol to add support for sending and receiving unreliable datagrams over a QUIC connection. This document describes a method for negotiating the ability to send an arbitrary value for the second-most significant bit in QUIC packets. In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements. RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings. This document proposes a notational convention to express Concise Binary Object Representation (CBOR) data structures (RFC 7049). Its main goal is to provide an easy and unambiguous way to express structures for protocol messages and data formats that use CBOR or JSON. This document defines an extension to the QUIC transport protocol to add support for sending and receiving unreliable datagrams over a QUIC connection. Informative References This document describes a Transport Layer Security (TLS) extension for application-layer protocol negotiation within the TLS handshake. For instances in which multiple application protocols are supported on the same TCP or UDP port, this extension allows the application layer to negotiate which protocol will be used within the TLS connection.

Acknowledgements Much of the initial work by Robin Marx was done at the Hasselt and KU Leuven Universities. Thanks to Jana Iyengar, Brian Trammell, Dmitri Tikhonov, Stephen Petrides, Jari Arkko, Marcus Ihlar, Victor Vasiliev, Mirja Kuehlewind, Jeremy Laine, Kazu Yamamoto, and Christian Huitema for their feedback and suggestions.

Change Log

Since draft-ietf-qlog-quic-events-06:

• Added PathAssigned and MigrationStateUpdated events (#336)
• Added extension points to parameters_set and parameters_restored (#400)
• Removed error_code_value from connection_closed (#386, #392)
• Renamed generation to key_phase for key_updated and key_discarded (#390)
• Removed retry_token from packet_sent and packet_received (#389)
• Updated ALPN handling (#385)
• Added key_unavailable trigger to packet_dropped (#381)
• Updated several uint32 to uint64
• ProtocolEventBody is now called ProtocolEventData (#352)
• Editorial changes (#402, #404, #394, #393)

Since draft-ietf-qlog-quic-events-05:

• SecurityKeyUpdated: the new key is no longer mandatory to log (#294)
• Added ECN related events and metadata (#263)

Since draft-ietf-qlog-quic-events-04:

• Updated guidance on logging events across connections (#279)
• Renamed 'transport' category to 'quic' (#302)
• Added support for multiple packet numbers in 'quic:frames_processed' (#307)
• Added definitions for RFC9287 (QUIC GREASE Bit extension) (#311)
• Added definitions for RFC9221 (QUIC Datagram Frame extension) (#310)
• (Temporarily) removed definitions for connection migration events (#317)
• Editorial and formatting changes (#298, #299, #304, #306, #327)

Since draft-ietf-qlog-quic-events-03:

• Ensured consistent use of RawInfo to indicate raw wire bytes (#243)
• Renamed UnknownFrame:raw_frame_type to :frame_type_value (#54)
• Renamed ConnectionCloseFrame:raw_error_code to :error_code_value (#54)
• Changed triggers for packet_dropped (#278)
• Added entries to TransportError enum (#285)
• Changed minimum_congestion_window to uint64 (#288)

Since draft-ietf-qlog-quic-events-02:

• Renamed key_retired to key_discarded (#185)
• Added fields and events for DPLPMTUD (#135)
• Made packet_number optional in PacketHeader (#244)
• Removed connection_retried event placeholder (#255)
• Changed QuicFrame to a CDDL plug type (#257)
• Moved data definitions out of the appendix into separate sections
• Added overview Table of Contents

Since draft-ietf-qlog-quic-events-01:

• Added Stateless Reset Token type (#122)

Since draft-ietf-qlog-quic-events-00:

• Change the data definition language from TypeScript to CDDL (#143)

Since draft-marx-qlog-event-definitions-quic-h3-02:

• These changes were done in preparation of the adoption of the drafts by the QUIC working group (#137)
• Split QUIC and HTTP/3 events into two separate documents
• Moved RawInfo, Importance, Generic events and Simulation events to the main schema document.
• Changed to/from value options of the data_moved event

Since draft-marx-qlog-event-definitions-quic-h3-01: Major changes:

• Moved data_moved from http to transport. Also made the "from" and "to" fields flexible strings instead of an enum (#111,#65)
• Moved packet_type fields to PacketHeader. Moved packet_size field out of PacketHeader to RawInfo:length (#40)
• Made events that need to log packet_type and packet_number use a header field instead of logging these fields individually
• Added support for logging retry, stateless reset and initial tokens (#94,#86,#117)
• Moved separate general event categories into a single category "generic" (#47)
• Added "transport:connection_closed" event (#43,#85,#78,#49)
• Added version_information and alpn_information events (#85,#75,#28)
• Added parameters_restored events to help clarify 0-RTT behaviour (#88)

Smaller changes:

• Merged loss_timer events into one loss_timer_updated event
• Field data types are now strongly defined (#10,#39,#36,#115)
• Renamed qpack instruction_received and instruction_sent to instruction_created and instruction_parsed (#114)
• Updated qpack:dynamic_table_updated.update_type. It now has the value "inserted" instead of "added" (#113)
• Updated qpack:dynamic_table_updated. It now has an "owner" field to differentiate encoder vs decoder state (#112)
• Removed push_allowed from http:parameters_set (#110)
• Removed explicit trigger field indications from events, since this was moved to be a generic property of the "data" field (#80)
• Updated transport:connection_id_updated to be more in line with other similar events. Also dropped importance from Core to Base (#45)
• Added length property to PaddingFrame (#34)
• Added packet_number field to transport:frames_processed (#74)
• Added a way to generically log packet header flags (first 8 bits) to PacketHeader
• Added additional guidance on which events to log in which situations (#53)
• Added "simulation:scenario" event to help indicate simulation details
• Added "packets_acked" event (#107)
• Added "datagram_ids" to the datagram_X and packet_X events to allow tracking of coalesced QUIC packets (#91)
• Extended connection_state_updated with more fine-grained states (#49)

Since draft-marx-qlog-event-definitions-quic-h3-00:

• Event and category names are now all lowercase
• Added many new events and their definitions
• "type" fields have been made more specific (especially important for PacketType fields, which are now called packet_type instead of type)
• Events are given an importance indicator (issue #22)
• Event names are more consistent and use past tense (issue #21)
• Triggers have been redefined as properties of the "data" field and updated for most events (issue #23)