MSC3886: Simple client rendezvous capability by hughns · Pull Request #3886 · matrix-org/matrix-spec-proposals

hughns · 2022-09-07T08:21:11Z

Servers:
- Standalone Rust server: https://github.com/matrix-org/rust-http-rendezvous-server
- Synapse homeserver module: https://pypi.org/project/matrix-http-rendezvous-synapse/
- Standalone prototype Node.js server: https://github.com/matrix-org/node-http-rendezvous-server
- Synapse experimental feature to delegate rendezvous to external server via an HTTP 307 redirect: Implementation of HTTP 307 response for MSC3886 POST endpoint synapse#14018
Clients:
- Element Web:
  - matrix-js-sdk: Support sign in + E2EE set up using QR code implementing MSC3886, MSC3903 and MSC3906 matrix-js-sdk#2747
  - matrix-react-sdk: Support for login + E2EE set up with QR matrix-react-sdk#9403
- Element iOS:
  - Add rendezvous service (MSC3886) and ECDH X25519 AES 256 based secure… element-hq/element-ios#6806

ara4n · 2022-09-11T20:59:25Z

proposals/3886-simple-rendezvous-capability.md

@@ -0,0 +1,236 @@
+# MSC3886: Simple client rendezvous capability


On one hand, this is a really simple and elegant standalone function. On the other hand, I'm a bit worried that it duplicates the semantics of to-device API (i.e. basic store & forward between devices), albeit with short-polling rather than long-polling.

I wonder how bad it would be if we opened up to-device messages to guests, and used the existing APIs for rendezvous? So a new device would go and /login as a guest to get a temporary access token, and then publish its device ID & HS url in its QR code to let another device rendezvous with it.

My only reason for proposing this is to avoid having two store-and-forward APIs which look suspiciously similar, but have different semantics (short/long poll), and so require more code for client implementors.

Understood. I'll work up an alternative based on to-device messages and see how that feels.

I have started some discussion on the to-device based alternative as part of #3903

I wonder how bad it would be if we opened up to-device messages to guests, and used the existing APIs for rendezvous? So a new device would go and /login as a guest to get a temporary access token, and then publish its device ID & HS url in its QR code to let another device rendezvous with it.

ugh, the complexity of this feels horrible to me.

My only reason for proposing this is to avoid having two store-and-forward APIs which look suspiciously similar, but have different semantics (short/long poll), and so require more code for client implementors.

Sure, having two store-and-forward APIs is rather less than ideal, but this one is so simple and easy to use that I don't really buy that it's a meaningful amount of extra code for clients comparing to have to grab a temporary access token and then start /syncing.

For me, the simplicity of this proposal outweighs the fact it looks a bit like to-device messaging. (Or even matrix rooms, if you squint hard enough and invent "ephemeral rooms".)

The only thing I'd say here is that it would be good if the "Alternatives" section in this MSC said something about this idea (even if it's just a link to MSC3903's alternatives section).

I think I was broadly coming to a similar conclusion to Rich. Adding guest access to to-device feels about as complex as this separate impl.

I've move the section from MSC3903 alternatives section into this proposal as there is much feedback here than on MSC3903 itself.

Given the above, it appears we've settled on using a new channel rather than exposing to-device to guests. @matrix-org/spec-core-team if you disagree then please raise comments :)

Upstream changes: Synapse 1.70.1 (2022-10-28) =========================== (bugfixes) Synapse 1.70.0 (2022-10-26) =========================== Features -------- - Support for [MSC3856](matrix-org/matrix-spec-proposals#3856): threads list API. ([\#13394](matrix-org/synapse#13394), [\#14171](matrix-org/synapse#14171), [\#14175](matrix-org/synapse#14175)) - Support for thread-specific notifications & receipts ([MSC3771](matrix-org/matrix-spec-proposals#3771) and [MSC3773](matrix-org/matrix-spec-proposals#3773)). ([\#13776](matrix-org/synapse#13776), [\#13824](matrix-org/synapse#13824), [\#13877](matrix-org/synapse#13877), [\#13878](matrix-org/synapse#13878), [\#14050](matrix-org/synapse#14050), [\#14140](matrix-org/synapse#14140), [\#14159](matrix-org/synapse#14159), [\#14163](matrix-org/synapse#14163), [\#14174](matrix-org/synapse#14174), [\#14222](matrix-org/synapse#14222)) - Stop fetching missing `prev_events` after we already know their signature is invalid. ([\#13816](matrix-org/synapse#13816)) - Send application service access tokens as a header (and query parameter). Implements [MSC2832](matrix-org/matrix-spec-proposals#2832). ([\#13996](matrix-org/synapse#13996)) - Ignore server ACL changes when generating pushes. Implements [MSC3786](matrix-org/matrix-spec-proposals#3786). ([\#13997](matrix-org/synapse#13997)) - Experimental support for redirecting to an implementation of a [MSC3886](matrix-org/matrix-spec-proposals#3886) HTTP rendezvous service. ([\#14018](matrix-org/synapse#14018)) - The `/relations` endpoint can now be used on workers. ([\#14028](matrix-org/synapse#14028)) - Advertise support for Matrix 1.3 and 1.4 on `/_matrix/client/versions`. ([\#14032](matrix-org/synapse#14032), [\#14184](matrix-org/synapse#14184)) - Improve validation of request bodies for the [Device Management](https://spec.matrix.org/v1.4/client-server-api/#device-management) and [MSC2697 Device Dehyrdation](matrix-org/matrix-spec-proposals#2697) client-server API endpoints. ([\#14054](matrix-org/synapse#14054)) - Experimental support for [MSC3874](matrix-org/matrix-spec-proposals#3874): Filtering threads from the `/messages` endpoint. ([\#14148](matrix-org/synapse#14148)) - Improve the validation of the following PUT endpoints: [`/directory/room/{roomAlias}`](https://spec.matrix.org/v1.4/client-server-api/#put_matrixclientv3directoryroomroomalias), [`/directory/list/room/{roomId}`](https://spec.matrix.org/v1.4/client-server-api/#put_matrixclientv3directorylistroomroomid) and [`/directory/list/appservice/{networkId}/{roomId}`](https://spec.matrix.org/v1.4/application-service-api/#put_matrixclientv3directorylistappservicenetworkidroomid). ([\#14179](matrix-org/synapse#14179)) Deprecations and Removals ------------------------- - Remove the experimental implementation of [MSC3772](matrix-org/matrix-spec-proposals#3772). ([\#14094](matrix-org/synapse#14094)) - Remove the unstable identifier for [MSC3715](matrix-org/matrix-spec-proposals#3715). ([\#14106](matrix-org/synapse#14106), [\#14146](matrix-org/synapse#14146))

turt2live · 2023-03-07T04:31:36Z

This was proposed as ready for FCP by the author, however it appears to have some conflicts with to-device messaging. For now, I'm going to put it into "needs technical design review". This doesn't mean it's not ready for FCP, just that the review's tone should take a slightly different shape.

@matrix-org/spec-core-team please review this for technical design.

Co-authored-by: Richard van der Hoff <1389908+richvdh@users.noreply.github.com>

uhoreg · 2023-04-25T22:28:28Z

proposals/3886-simple-rendezvous-capability.md

+  R->>-B: 202 Accepted 
+
+  Note over A,B: Rendezvous now established
+```


My understanding of this is that A and B take turns writing to the same rendezvous URI until they're done. So when it's B's turn to write, A keeps polling (using the ETag) until the server says the data has changed, and vice versa.

What happens if B tries to write, but gets some sort of network error, or an error from a proxy? If the server got B's data, but B received a network error, then it seems to me what could happen is:

A receives B's data, thinks it's now their turn to send data, so sends their data and gets a new ETag

B retries the request, overwriting A's data (and never receiving it)

A polls for new data, using the new ETag

since B overwrote A's data, the data doesn't match the ETag, so A gets the data B sent, again

So B will miss a message from A, and A will get a duplicate message.

Perhaps we could mitigate against this by using a RFC7232 If-Match on the PUT requests?

ISTM that every PUT should be required to cite a previous ETag so that the rendezvous server can enforce a linear ordering. (The initial ETag is included in the POST and GET response, so both A and B should be fully aware of it.)

58f1e86 does this.

uhoreg · 2023-04-25T22:32:16Z

proposals/3886-simple-rendezvous-capability.md

+HTTP request headers:
+
+- `If-None-Match` - optional, as per [RFC7232](https://httpwg.org/specs/rfc7232.html#header.if-none-match) server will
+only return data if given ETag does not match


Might be nice for servers to have the option to delay responding until it gets content that doesn't match the ETag, so we can do long-polling.

uhoreg · 2023-04-25T22:35:24Z

proposals/3886-simple-rendezvous-capability.md

+Because this is an entirely new set of functionality it should not cause issue with any existing Matrix functions or capabilities.
+
+The proposed protocol requires the devices to have IP connectivity to the server which might not be the case in P2P scenarios.
+


One potential issue here is that if A sends a message to B, then waits for a message from B using the ETag, but the message that B sends to A happens to be exactly the same as the message that A sent, then A will get the 304 Not Modified response, and never realize that B sent a message. So anything built on top of this needs to ensure that a message is never identical to the preceding message.

Is this still a problem? From the current text, it sounds like 304 Not Modified will only be returned when a matching ETag is supplied in a If-None-Match. Given the resolution of this thread, clients will have to supply a previous ETag when doing a PUT, which means we no longer have to rely on the sameness of the content to decide whether the content has been modified. That is, a PUT request that specifies the previous ETag A should be regarded as altering the payload at A, even if the payload is unchanged, and therefore, it should be assigned a new ETag.

turt2live

overall this MSC seems like a good approach for the problem. I've left some comments about things that should be clarified before the MSC can go up for FCP.

turt2live · 2023-06-20T00:56:28Z

proposals/3886-simple-rendezvous-capability.md

+- the user ID
+- facilitation of issuing a new access token
+- device ID for end-to-end encryption
+- device keys for end-to-end encryption


We shouldn't be sending (private) device keys over the wire like this. They should be generated by the new device, which may be the device ID given, but not transmitted over the wire.

dcbbcb0 and 74d9094 clarify that this is an untrusted communications channel.

proposals/3886-simple-rendezvous-capability.md

turt2live · 2023-06-20T00:58:45Z

proposals/3886-simple-rendezvous-capability.md

@@ -0,0 +1,236 @@
+# MSC3886: Simple client rendezvous capability


Given the above, it appears we've settled on using a new channel rather than exposing to-device to guests. @matrix-org/spec-core-team if you disagree then please raise comments :)

turt2live · 2023-06-20T00:59:21Z

proposals/3886-simple-rendezvous-capability.md

+
+- any data up to maximum size allowed by the server
+
+HTTP response codes:


These also need to have Matrix error codes to go with them please

See aee7d81. I have proposed M_DIRTY_WRITE for the HTTP 412 Precondition Failed case where Alice attempts to write but hasn't seen Bob's latest write.

Although I self-bikeshedded a different name in 3fecfcd.

turt2live · 2023-06-20T01:00:17Z

proposals/3886-simple-rendezvous-capability.md

+- `Location` - required, the allocated rendezvous URI which can be on a different server
+- `X-Max-Bytes` - required, the maximum allowed bytes for the payload


why are these headers and not response body parameters?

On reflection I would agree that if it is going to be part of the C-S API then it would make sense to consider consistency with the rest of the C-S API where headers are not used.

Are we concerned with just these two headers? Or do we want all of the response and request headers to be expressed via HTTP bodies?

The response body for GET ought to be just the payload itself. Since the payload data is an arbitrary byte sequence, it would be painful to embed this in JSON. Therefore I would encourage the current GET response headers (Content-Type, ETag, Expires, Last-Modified) to continue to be expressed via headers. For consistency it makes sense to do so in all other resposnes.

For POST this leaves the two highlighted headers: Location and X-Max-Bytes. We could present them as a JSON-encoded body, but it would seem odd to spread the POST response metadata in two places without any meaningful distinction to justify it. My vote would be to leave things as they are. But I neither have a vote, nor any strong opinions.

The response body for GET ought to be just the payload itself. Since the payload data is an arbitrary byte sequence, it would be painful to embed this in JSON.

We could base64 encode the payload, but I tend to agree. Not all things need to be shoehorned through JSON.

turt2live · 2023-06-20T01:01:08Z

proposals/3886-simple-rendezvous-capability.md

+
+- any data up to maximum size allowed by the server
+
+HTTP response codes:


Matrix error codes here too please, and probably a 400 definition to cover missing headers and such

(comment applies throughout remainder of proposal)

90a8b49 defines a 400 response.

DMRobertson · 2023-11-13T14:14:20Z

proposals/3886-simple-rendezvous-capability.md

+
+HTTP response headers for `201 Created`:
+
+- `Location` - required, the allocated rendezvous URI which can be on a different server


Presumably this needs URI needs to be not guessable, to prevent attackers from guessing this and impersonating the intended recipient?

DMRobertson · 2023-11-13T15:08:49Z

proposals/3886-simple-rendezvous-capability.md

+HTTP response headers for `202 Accepted` and `412 Precondition Failed`:
+
+- `ETag` - required, ETag for the current payload at the rendezvous point as per [RFC7232](https://httpwg.org/specs/rfc7232.html#header.etag)
+- `Expires` - required, the expiry time of the rendezvous as per [RFC7233](https://httpwg.org/specs/rfc7234.html#header.expires)


Is the intention that the expiry time is incremented every time the rendezvous payload is updated?

I have assumed so in 65d697c

For establshing an encrypted channel atop this

"Dirty write" refers to a specific SQL phenomenon regarding transaction isolation, see e.g. https://blog.acolyer.org/2016/02/24/a-critique-of-ansi-sql-isolation-levels/ Instead, prefer the word "concurrent".

proposals/3886-simple-rendezvous-capability.md

Co-authored-by: Denis Kasak <dkasak@termina.org.uk>

hughns · 2024-04-09T11:56:09Z

It is proposed that MSC4108 supersedes this MSC.

hughns · 2024-04-30T18:16:16Z

Closing this PR as #4108 is now ready for review.

No significant changes since 1.108.0rc1. - Add a feature that allows clients to query the configured federation whitelist. Disabled by default. ([\#16848](element-hq/synapse#16848), [\#17199](element-hq/synapse#17199)) - Add the ability to allow numeric user IDs with a specific prefix when in the CAS flow. Contributed by Aurélien Grimpard. ([\#17098](element-hq/synapse#17098)) - Fix bug where push rules would be empty in `/sync` for some accounts. Introduced in v1.93.0. ([\#17142](element-hq/synapse#17142)) - Add support for optional whitespace around the Federation API's `Authorization` header's parameter commas. ([\#17145](element-hq/synapse#17145)) - Fix bug where disabling room publication prevented public rooms being created on workers. ([\#17177](element-hq/synapse#17177), [\#17184](element-hq/synapse#17184)) - Document [`/v1/make_knock`](https://spec.matrix.org/v1.10/server-server-api/#get_matrixfederationv1make_knockroomiduserid) and [`/v1/send_knock/`](https://spec.matrix.org/v1.10/server-server-api/#put_matrixfederationv1send_knockroomideventid) federation endpoints as worker-compatible. ([\#17058](element-hq/synapse#17058)) - Update User Admin API with note about prefixing OIDC external_id providers. ([\#17139](element-hq/synapse#17139)) - Clarify the state of the created room when using the `autocreate_auto_join_room_preset` config option. ([\#17150](element-hq/synapse#17150)) - Update the Admin FAQ with the current libjemalloc version for latest Debian stable. Additionally update the name of the "push_rules" stream in the Workers documentation. ([\#17171](element-hq/synapse#17171)) - Add note to reflect that [MSC3886](matrix-org/matrix-spec-proposals#3886) is closed but will remain supported for some time. ([\#17151](element-hq/synapse#17151)) - Update dependency PyO3 to 0.21. ([\#17162](element-hq/synapse#17162)) - Fixes linter errors found in PR #17147. ([\#17166](element-hq/synapse#17166)) - Bump black from 24.2.0 to 24.4.2. ([\#17170](element-hq/synapse#17170)) - Cache literal sync filter validation for performance. ([\#17186](element-hq/synapse#17186)) - Improve performance by fixing a reactor pause. ([\#17192](element-hq/synapse#17192)) - Route `/make_knock` and `/send_knock` federation APIs to the federation reader worker in Complement test runs. ([\#17195](element-hq/synapse#17195)) - Prepare sync handler to be able to return different sync responses (`SyncVersion`). ([\#17200](element-hq/synapse#17200)) - Organize the sync cache key parameter outside of the sync config (separate concerns). ([\#17201](element-hq/synapse#17201)) - Refactor `SyncResultBuilder` assembly to its own function. ([\#17202](element-hq/synapse#17202)) - Rename to be obvious: `joined_rooms` -> `joined_room_ids`. ([\#17203](element-hq/synapse#17203), [\#17208](element-hq/synapse#17208)) - Add a short pause when rate-limiting a request. ([\#17210](element-hq/synapse#17210)) * Bump cryptography from 42.0.5 to 42.0.7. ([\#17180](element-hq/synapse#17180)) * Bump gitpython from 3.1.41 to 3.1.43. ([\#17181](element-hq/synapse#17181)) * Bump immutabledict from 4.1.0 to 4.2.0. ([\#17179](element-hq/synapse#17179)) * Bump sentry-sdk from 1.40.3 to 2.1.1. ([\#17178](element-hq/synapse#17178)) * Bump serde from 1.0.200 to 1.0.201. ([\#17183](element-hq/synapse#17183)) * Bump serde_json from 1.0.116 to 1.0.117. ([\#17182](element-hq/synapse#17182))

Simple rendezvous capability

66cc232

hughns changed the title ~~Simple rendezvous capability~~ MSC3886: Simple rendezvous capability Sep 7, 2022

MSC3886: Simple rendezvous capability

e15fda0

hughns changed the title ~~MSC3886: Simple rendezvous capability~~ MSC3886: Simple client rendezvous capability Sep 7, 2022

hughns added 4 commits September 7, 2022 13:42

Note that it is client rendezvous rather than anything else

0350bf8

Describe protocol + more notes

fcc3270

Use HTTP response headers only instead of JSON for POST /

60f8115

Fix POST response code

5a9c3c7

turt2live added proposal A matrix spec change proposal client-server Client-Server API kind:feature MSC for not-core and not-maintenance stuff needs-implementation This MSC does not have a qualifying implementation for the SCT to review. The MSC cannot enter FCP. labels Sep 7, 2022

hughns and others added 6 commits September 9, 2022 12:19

Add extra emphasis to relative nature of Location header

fac41d1

Use rendezvous URI not IDs

cbffa67

Update CORS with more headers

c67a3d6

Sort HTTP response codes

1ec9ce2

Add requirement for content-length header on POST and PUT

8a0d559

fix typo & clarification

cacae4e

ara4n requested changes Sep 11, 2022

View reviewed changes

hughns added 3 commits September 30, 2022 12:17

Add link to current CORS headers

94ef9dd

Add 307 response for POST

953c4ee

Add unstable prefixes

ff9a373

hughns marked this pull request as ready for review October 4, 2022 17:26

MSCyyyy => MSC3906

6937a86

hughns mentioned this pull request Oct 13, 2022

Support sign in + E2EE set up using QR code implementing MSC3886, MSC3903 and MSC3906 matrix-org/matrix-js-sdk#2747

Merged

3 tasks

turt2live assigned hughns Feb 28, 2023

turt2live requested review from a team and turt2live February 28, 2023 02:31

hughns and others added 5 commits March 15, 2023 16:23

Apply suggestions from code review

4ab59f8

Co-authored-by: Richard van der Hoff <1389908+richvdh@users.noreply.github.com>

Add missing arrows to diagram

5db0af6

Clarify that CORS changes are not global

8a1af85

Authentication + 30x clarifications

97f1709

Import section from MSC3903 about to-device messaging

b5c6c7a

uhoreg reviewed Apr 25, 2023

View reviewed changes

turt2live removed their request for review June 6, 2023 19:48

turt2live reviewed Jun 20, 2023

View reviewed changes

David Robertson added 3 commits November 13, 2023 13:32

Create a rendezvous point via /_matrix/client

931cf07

Define Matrix error codes

aee7d81

Define 400 response for missing required headers

90a8b49

DMRobertson reviewed Nov 13, 2023

View reviewed changes

David Robertson added 8 commits November 13, 2023 15:21

Emphasise that the this is an untrusted channel

dcbbcb0

Refer to MSC3903

74d9094

For establshing an encrypted channel atop this

Add internal hyperlink

4c63493

Require PUT to supply an If-Match

58f1e86

High-level description

65d697c

Forbid complicated etags

43d9871

Rename M_DIRTY_WRITE -> M_CONCURRENT_WRITE

3fecfcd

"Dirty write" refers to a specific SQL phenomenon regarding transaction isolation, see e.g. https://blog.acolyer.org/2016/02/24/a-critique-of-ansi-sql-isolation-levels/ Instead, prefer the word "concurrent".

Explictly reject complicated etags

82fcb44

dkasak reviewed Nov 29, 2023

View reviewed changes

proposals/3886-simple-rendezvous-capability.md Outdated Show resolved Hide resolved

Update proposals/3886-simple-rendezvous-capability.md

a08d14b

Co-authored-by: Denis Kasak <dkasak@termina.org.uk>

hughns closed this Apr 30, 2024

turt2live added the obsolete A proposal which has been overtaken by other proposals label Apr 30, 2024

hughns mentioned this pull request Sep 15, 2025

MSC4108: Mechanism to allow OAuth 2.0 API sign in and E2EE set up via QR code #4108

Open

21 tasks

derritter88 mentioned this pull request Nov 21, 2025

Link new device by QR code is not working element-hq/synapse#18027

Closed

		@@ -0,0 +1,236 @@
		# MSC3886: Simple client rendezvous capability

		Because this is an entirely new set of functionality it should not cause issue with any existing Matrix functions or capabilities.

		The proposed protocol requires the devices to have IP connectivity to the server which might not be the case in P2P scenarios.


		- any data up to maximum size allowed by the server

		HTTP response codes:

		- `Location` - required, the allocated rendezvous URI which can be on a different server
		- `X-Max-Bytes` - required, the maximum allowed bytes for the payload


		HTTP response headers for `201 Created`:

		- `Location` - required, the allocated rendezvous URI which can be on a different server

Conversation

hughns commented Sep 7, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

turt2live commented Mar 7, 2023

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

turt2live left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

hughns commented Apr 9, 2024

Uh oh!

hughns commented Apr 30, 2024

Uh oh!

Reviewers

Assignees

Labels

hughns commented Sep 7, 2022 •

edited

Loading