RELIABLE-CHANNEL-API

FieldValue
NameReliable Channel API definition
Slug173
Statusraw
CategoryStandards Track
Tagsreliability, application, api, sds, segmentation
EditorIvan Folgueira Bande [email protected]
ContributorsJazz Turner-Baggs [email protected], Igor Sirotin [email protected]

Timeline

  • 2026-05-111ac7689 — chore: split ift ts specs (#334)
  • 2026-05-07d3ee034 — docs: add Reliable Channel API spec (#325)

Table of contents

Abstract

This document specifies the Reliable Channel API, an application-level interface that sits between the application layer and the MESSAGING-API plus P2P-RELIABILITY, i.e., application <-> reliable-channel-api <-> messaging-api/p2p-reliability.

It bundles segmentation, end-to-end reliability via Scalable Data Sync (SDS), rate limit management, and a pluggable encryption hook into a single interface for sending and receiving messages reliably.

Motivation

The MESSAGING-API provides peer-to-peer reliability via P2P-RELIABILITY, but does not provide high end-to-end delivery guarantees from sender to recipient.

This API addresses that gap by introducing:

  • SEGMENTATION to handle large messages exceeding network size limits.
  • SDS to provide causal-history-based end-to-end acknowledgement and retransmission.
  • Rate Limit Manager to comply with RLN constraints when sending segmented messages.
  • Encryption Hook to allow upper layers to provide a pluggable encryption mechanism. This enables applications to provide Confidentiality and Integrity if desired.

The separation between Reliable Channels and encryption ensures the API remains agnostic to identity and key management concerns, which are handled by higher layers.

Syntax

The keywords "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.

API design

Architectural position

The Reliable Channel API sits between the application layer and the Messaging API, as follows:

┌────────────────────────────────────────────────────────────┐
│                   Application Layer                        │
└───────────────────────────┬────────────────────────────────┘
                            │
┌───────────────────────────▼────────────────────────────────┐
│                 Reliable Channel API                       │
│  ┌──────────────┐ ┌─────┐ ┌───────────────┐ ┌──────────┐   │
│  │ Segmentation │ │ SDS │ │ Rate Limit Mgr│ │Encryption│   │
│  │              │ │     │ │               │ │   Hook   │   │
│  └──────────────┘ └─────┘ └───────────────┘ └──────────┘   │
└───────────────────────────┬────────────────────────────────┘
                            │
┌───────────────────────────▼────────────────────────────────┐
│                    Messaging API                           │
│      (P2P Reliability, Relay, Filter, Lightpush, Store)    │
└────────────────────────────────────────────────────────────┘

IDL

A custom Interface Definition Language (IDL) in YAML is used, consistent with MESSAGING-API.

Components

Segmentation

A protocol that splits message payloads into smaller units during transmission and reassembles them upon reception. The component is instantiated by supplying the appropriate value to SegmentationConfig.

See SEGMENTATION.

Scalable Data Sync (SDS)

SDS provides end-to-end delivery guarantees using causal history tracking.

  • Each new segment to be sent, requires the following data:

    • MessageId: a keccak-256(message's content) hex string (e.g. 4e03657aea45a94fc7d47ba826c8d667c0d1e6e33a64a036ec44f58fa12d6c45), generated by the Reliable Channel API.
    • ChannelId: the channelId passed to createReliableChannel.
    • Retrieval hint: the transport MessageHash of previous segments, mentioned as message_hash in MESSAGING-API, upon ``MessageSendPropagatedEventreception. The hint provider registered during [Node initialization](#node-initialization) performs thisMessageId → MessageHashlookup. In turn, that mapping MUST be persisted by SDS using thepersistencebackend configured inSdsConfig`.
  • Each sent segment is added to an outgoing buffer.

  • The recipient sends acknowledgements back to the sender upon receiving segments.

  • The sender removes acknowledged segments from the outgoing buffer.

  • Unacknowledged segments are retransmitted after acknowledgementTimeoutMs.

  • SDS state MUST be persisted using the persistence backend configured in SdsConfig.

Rate Limit Manager

The Rate Limit Manager ensures compliance with RLN rate constraints.

  • It tracks how many messages have been sent in the current epoch.
  • When the limit is approached, segment dispatch MUST be delayed to the next epoch.
  • The epoch size MUST match the [epochPeriodSec](https://lip.logos.co/messaging/draft/17/rln-relay.html#epoch-length) configured in RateLimitConfig.

Encryption Hook

The Encryption Hook provides a pluggable interface for upper layers to inject encryption.

  • The hook is optional; when not provided, messages are sent unencrypted.
  • Encryption is applied per segment, after segmentation and SDS.
  • Decryption is applied per segment before being processed by SDS.
  • The Encryption interface MUST be implemented by the caller when the hook is provided.
  • The Reliable Channel API MUST NOT impose any specific encryption scheme.

Procedures

Node initialization

When a node is created via createNode (defined in MESSAGING-API), the implementation MUST perform the following setup before the node is used:

  1. Configure SDS persistence: Supply the Persistence backend from SdsConfig to the SDS module so that causal history and outgoing buffers survive restarts.
  2. Configure SDS hint provider: Register a hint provider with the SDS module. The hint provider converts an SDS MessageId into its corresponding MessageHash.
  3. Configure Segmentation persistence: Supply the Persistence backend from SegmentationConfig to the Segmentation module so that partially reassembled messages survive restarts.
  4. Fetch missed messages: Retrieve messages missed while offline as described in MESSAGING-API — Fetching missed messages on startup.

Outgoing message processing

When send is called, the implementation MUST process message in the following order:

  1. Segment: Split the payload into segments as defined in SEGMENTATION.
  2. Apply SDS: add each sds message to the SDS outgoing buffer (see SDS for parameter bindings).
  3. Encrypt: If an Encryption implementation is provided, encrypt each segment before transmission.
  4. Rate Limit: If RateLimitConfig.enabled is true, delay dispatch as needed to comply with RLN epoch constraints.
  5. Dispatch: Send each segment via the underlying MESSAGING-API.

Incoming message processing

When a segment is received from the network, the implementation MUST process it in the following order:

  1. Decrypt: If an Encryption implementation is provided, decrypt the segment.
  2. Apply SDS: Deliver the segment to the SDS layer, which emits acknowledgements and detects gaps.
    • Detect missing dependencies: If SDS detects a gap in the causal history, it MUST make a best-effort attempt to retrieve the missing message. The Retrieval hint (see Scalable Data Sync (SDS)) carried in each SDS message provides the transport MessageHash needed to query the store; without it, store retrieval is not possible. If the message cannot be retrieved, SDS MAY mark it as lost.
  3. Reassemble: Once all segments for a message have been received, reassemble and emit a reliable:message:received event.

Rate limiting

When RateLimitConfig.enabled is true, the implementation MUST space segment transmissions to comply with the RLN epoch constraints defined in [epochPeriodSec](https://lip.logos.co/messaging/draft/17/rln-relay.html#epoch-length). Segments MUST NOT be sent at a rate that would violate the RLN message rate limit for the active epoch.

Encryption

The encryption parameter in createReliableChannel is intentionally optional. The Reliable Channel API is agnostic to encryption mechanisms.

When an Encryption implementation is provided, it MUST be applied as described in Outgoing message processing and Incoming message processing.

The Reliable Channel API

This API considers the types defined by MESSAGING-API plus the following.

Channel lifecycle

This point assumes that a WakuNode instance is created beforehand. See createNode function in MESSAGING-API.

functions:

  createReliableChannel:
    description: "Creates a reliable channel over the given content topic. Sets up the required SDS state,
    segmentation, and encryption, and subscribes to `contentTopic`."
    parameters:
      - name: node
        type: WakuNode
        description: "The underlying messaging node, as defined in [MESSAGING-API](messaging-api.md).
        Used to send segments and to subscribe/unsubscribe to the content topics."
      - name: channelId
        type: string
        description: "Unique identifier for this channel. Represents the reliable (SDS), segmented, and optionally-encrypted session."
      - name: contentTopic
        type: string
        description: "The topic this channel listens and sends on. This has routing and filtering connotations."
      - name: senderId
        type: string
        description: "An identifier for this sender. SHOULD be unique and persisted between sessions."
      - name: encryption
        type: optional<Encryption>
        default: none
        description: "Optional pluggable encryption implementation. If none, messages are sent unencrypted."
    returns:
      type: result<ReliableChannel, error>

  closeChannel:
    description: "Closes a reliable channel, releases all associated resources and internal state,
    and unsubscribes from its content topic via the underlying [MESSAGING-API](messaging-api.md)."
    parameters:
      - name: channel
        type: ReliableChannel
        description: "The channel handle returned by `createReliableChannel`."
    returns:
      type: result<void, error>

Channel usage

functions:
  send:
    description: "Send a message through a reliable channel. The message is always segmented,
    SDS-tracked, rate-limited (optional), and encrypted (optional)."
    parameters:
      - name: channel
        type: ReliableChannel
        description: "The channel handle returned by `createReliableChannel`."
      - name: message
        type: array<byte>
        description: "The raw message payload to send."
    returns:
      type: result<RequestId, error>
      description: "Returns a `RequestId` that callers can use to correlate subsequent `MessageSentEvent` or `MessageSendErrorEvent` events."

Node configuration

This spec extends NodeConfig, needed to create a node, which is defined in MESSAGING-API, with sds_config and rate_limit_config fields.

NodeConfig:  # Extends NodeConfig defined in MESSAGING-API
  fields:
    sds_config:
      type: SdsConfig
      description: "SDS configuration. See SdsConfig defined in this spec."
    rate_limit_config:
      type: RateLimitConfig
      description: "See RateLimitConfig defined in this spec."
    segmentation_config:
      type: SegmentationConfig
      description: "See SegmentationConfig defined in this spec."

Type definitions

types:

  ReliableChannel:
    type: object
    description: "A handle representing an open reliable channel.
    Returned by `createReliableChannel` and used to send messages and receive events.
    Internal state (SDS, segmentation, encryption) is managed by the implementation."
    events:
      "reliable:message:received":
        type: MessageReceivedEvent
      "reliable:message:sent":
        type: MessageSentEvent
      "reliable:message:delivered":
        type: MessageDeliveredEvent
      "reliable:message:send-error":
        type: MessageSendErrorEvent
      "reliable:message:delivery-error":
        type: MessageDeliveryErrorEvent

  MessageReceivedEvent:
    type: object
    description: "Event emitted when a complete message has been received and reassembled."
    fields:
      message:
        type: array<byte>
        description: "The reassembled message payload."

  MessageSentEvent:
    type: object
    description: "Event emitted when all segments of a message have been transmitted to the network.
    This confirms network-level dispatch only; it does not guarantee the recipient has processed the message.
    For end-to-end confirmation, listen for `MessageDeliveredEvent`."
    fields:
      requestId:
        type: RequestId
        description: "The identifier of the `send` operation whose segments have all been dispatched to the network."

  MessageDeliveredEvent:
    type: object
    description: "Event emitted when the recipient has confirmed end-to-end receipt of a message via SDS acknowledgements.
    This event is fired asynchronously after `MessageSentEvent`, once the SDS layer receives explicit acknowledgements from the recipient."
    fields:
      requestId:
        type: RequestId
        description: "The identifier of the `send` operation confirmed as delivered by the recipient."

  MessageSendErrorEvent:
    type: object
    description: "Event emitted when one or more segments of a message could not be dispatched to the network.
    This indicates a network-level failure; the message was never fully transmitted."
    fields:
      requestId:
        type: RequestId
        description: "The identifier of the `send` operation that failed to dispatch."
      error:
        type: string
        description: "Human-readable description of the dispatch failure."

  MessageDeliveryErrorEvent:
    type: object
    description: "Event emitted when end-to-end delivery could not be confirmed.
    The message reached the network and there's no need to explicit re-send.
    Fired after `maxRetransmissions` attempts have been exhausted without receiving an SDS acknowledgement from the recipient."
    fields:
      requestId:
        type: RequestId
        description: "The identifier of the `send` operation that was not acknowledged by the recipient."
      error:
        type: string
        description: "Human-readable description of the delivery failure."

  RequestId:
    type: string
    description: "Unique identifier for a single `send` operation on a reliable channel.
    It groups all segments produced by segmenting one message, so callers can correlate
    acknowledgement and error events back to the original send call.
    Internally, each segment is dispatched as an independent [MESSAGING-API](messaging-api.md) call,
    producing one `RequestId` (as defined in [MESSAGING-API](messaging-api.md)) per segment.
    A single `RequestId` therefore maps to one or more underlying [MESSAGING-API](messaging-api.md)'s `RequestId` values,
    one per segment sent.
    For example, the `RequestId` `Req_a` yields these MESSAGING-API requests:
    `Req_a:1`, `Req_a:2`, ..., `Req_a:N`, where `Req_a:k` represents the k-th
    MESSAGING-API segment `RequestId`.
    That is, `Req_a` is the `RequestId` from the RELIABLE-CHANNEL-API spec PoV,
    whereas `Req_a:k` is the `RequestId` from the MESSAGING-API spec PoV.
    "

  SegmentationConfig:
    type: object
    fields:
      enableReedSolomon:
        type: bool
        default: false
        description: When enabled, the message sender adds parity (redundant) segments to allow recovery in case of data segment loss. See [SEGMENTATION](./segmentation.md).
      segmentSizeBytes:
        type: uint
        default: 102400  # 100 KiB
        description: "Maximum segment size in bytes.
        Messages larger than this value are split before SDS processing."
      persistence:
        type: Persistence
        description: "Backend for persisting partial reassembly state across restarts.
        Implementations MUST use this backend to store received segments until all segments of a message have arrived and can be reassembled.
        Refer to [SEGMENTATION](./segmentation.md) for the full definition of what state must be persisted."

  SdsConfig:
    type: object
    description: Scalable Data Sync config items.
    fields:
      persistence:
        type: Persistence
        description: "Backend for persisting the SDS local history. Implementations MAY support custom backends."
      acknowledgementTimeoutMs:
        type: uint
        default: 5000
        description: "Time in milliseconds to wait for acknowledgement before retransmitting."
      maxRetransmissions:
        type: uint
        default: 5
        description: "Maximum number of retransmission attempts before considering delivery failed."
      causalHistorySize:
        type: uint
        default: 2
        description: "Number of message IDs to consider in the causal history. With longer value, a stronger correctness is guaranteed but it requires higher bandwidth and memory."

  RateLimitConfig:
    type: object
    description: Rate limiting configuration, containing RLN-specific attributes.
    fields:
      enabled:
        type: bool
        default: false
        description: "Whether rate limiting is enforced. SHOULD be true when RLN is active."
      epochPeriodSec:
        type: uint
        default: 600  # 10 minutes
        description: "The epoch size used by the RLN relay, in seconds."

  Encryption:
    type: object
    description: "Interface for a pluggable encryption mechanism.
    When provided as a parameter to `createReliableChannel`, the API consumer MUST implement both encrypt and decrypt operations.
    Implementations MAY use different signatures than those described below, as long as each operation accepts a byte array and returns a byte array."
    fields:
      encrypt:
        type: function
        description: "Encrypts a byte payload. Returns the encrypted payload."
        parameters:
          - name: content
            type: array<byte>
        returns:
          type: result<array<byte>, error>
      decrypt:
        type: function
        description: "Decrypts a byte payload. Returns the decrypted payload."
        parameters:
          - name: payload
            type: array<byte>
        returns:
          type: result<array<byte>, error>

  Persistence:
    type: object
    description: "Interface for a pluggable SDS persistence backend.
    Implementations MUST provide all functions required to save and retrieve SDS state per channel. Implementations MUST also provide the persistence method of interest, e.g., SQLite, custom encrypted storage, etc.
    Refer to the [SDS spec](https://lip.logos.co/anoncomms/raw/sds.html) for the full definition of what state must be persisted."

Security/Privacy Considerations

  • This API does not provide confidentiality by default. An Encryption implementation MUST be supplied when confidentiality is required.
  • Segment metadata (message ID, segment index, total segments) is visible to network observers unless encrypted by the hook.
  • SDS acknowledgement messages are sent over the same content topic and are subject to the same confidentiality concerns.
  • Rate limiting compliance is required to avoid exclusion from the network by RLN-enforcing relays.

Copyright and related rights waived via CC0.