-
Notifications
You must be signed in to change notification settings - Fork 400
MSC4277: Harmonizing the reporting endpoints #4277
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Changes from 6 commits
0f9c41e
c12e1e9
34309cb
0258d25
8f9de37
eca1255
67c94b7
efa3f23
b3ac83f
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,77 @@ | ||
# MSC4277: Harmonizing the reporting endpoints | ||
|
||
Matrix currently has three reporting endpoints: | ||
|
||
- [`/_matrix/client/v3/rooms/{roomId}/report/{eventId}`] ([added] in Matrix | ||
Johennes marked this conversation as resolved.
Show resolved
Hide resolved
|
||
0.4.0, no MSC known) | ||
- [`/_matrix/client/v3/rooms/{roomId}/report`] ([added][1] in Matrix 1.13 as per | ||
[MSC4151]) | ||
- `/_matrix/client/v3/users/{userId}/report` (to be [added][2] to Matrix as per | ||
[MSC4260]) | ||
|
||
The spec contains a number of subtle differences for these endpoints: | ||
|
||
1. The `reason` parameter on the event reporting endpoint was made optional by | ||
[MSC2414]. The other two reporting endpoints didn't pick up on this, | ||
however, and still require `reason` to be present while allowing it to be | ||
blank. | ||
2. The user reporting endpoint [allows] servers to respond with 200 even if the | ||
user doesn't exist to deny enumerating users. This option is not allowed in | ||
the event and room reporting endpoints. | ||
3. The user and event reporting endpoints allow servers to add a random delay | ||
when generating responses. This is a more sophisticated measure against | ||
enumeration attacks. While the spec doesn't explicit forbid this technique | ||
on the room reporting endpoint it doesn't explicitly mention or recommend | ||
it either. | ||
|
||
These differences seem unnecessary and were likely introduced by accident only. | ||
The present proposal, therefore, seeks to align the three endpoints. | ||
|
||
## Proposal | ||
|
||
On all three endpoints: | ||
|
||
1. The `reason` parameter is made optional. | ||
Johennes marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
||
2. Servers MAY respond with 200 and no content regardless of whether the | ||
reported subject exists or not to combat enumeration attacks. | ||
|
||
3. Servers MAY add a random delay or use constant time functions when | ||
processing responses to combat enumeration attacks. | ||
|
||
All of these changes appear applicable regardless of the reported subject and it | ||
is not clear why the spec should differentiate the endpoints here. | ||
|
||
## Potential issues | ||
|
||
The `reason` field on the room and user reporting endpoints is currently | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Honestly, I find it bizarre that reason is required but may be empty string. I suppose this might not be the MSC to change it, but I'd rather we just made it optional. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I think I understand the value of requiring it to be filled for spam protection. The combination of required but allowed to be empty strikes me as slightly odd, too, but it's hard to judge without knowing T&S's future plans for the field. Generally though, I suppose if we later go from required but possibly empty to required and not empty, it'll be just as much a breaking change for clients as if we went from optional to required and not empty. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. It's hard to say. Even if the field was required people could just keyboard smash the field. I'm not really going to die on this hill though, I just think we should unify it in one direction. My feeling for T&S that any report is a useful report, and that tooling is smart enough these days that we can make use of the room_id / event_id to do preliminary investigations even if the |
||
required. Making it optional, therefore, is a breaking change. Clients should | ||
either act on the servers supported version or blanketly include an empty string | ||
if the user doesn't provide a reason. | ||
|
||
## Alternatives | ||
|
||
None. | ||
|
||
## Security considerations | ||
|
||
Enumeration attacks are likely more severe for users than for rooms or events. | ||
There are still not irrelevant, however. | ||
|
||
## Unstable prefix | ||
|
||
None. | ||
|
||
## Dependencies | ||
|
||
None. | ||
|
||
[`/_matrix/client/v3/rooms/{roomId}/report/{eventId}`]: https://spec.matrix.org/v1.13/client-server-api/#post_matrixclientv3roomsroomidreporteventid | ||
[added]: https://github.com/matrix-org/matrix-spec-proposals/pull/1264 | ||
[`/_matrix/client/v3/rooms/{roomId}/report`]: https://spec.matrix.org/v1.13/client-server-api/#post_matrixclientv3roomsroomidreport | ||
[1]: https://github.com/matrix-org/matrix-spec/pull/1938 | ||
[MSC4151]: https://github.com/matrix-org/matrix-spec-proposals/pull/4151 | ||
[2]: https://github.com/matrix-org/matrix-spec/pull/2093 | ||
[MSC4260]: https://github.com/matrix-org/matrix-spec-proposals/pull/4260 | ||
[MSC2414]: https://github.com/matrix-org/matrix-spec-proposals/pull/2414 | ||
[allows]: https://github.com/matrix-org/matrix-spec-proposals/pull/4260/files#diff-cbb17920e2617e7a20ab0838879675f7aa70e828f0263a3cfa5f4c53913ce5f7R34-R35 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Implementation requirements:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
200 regardless of event / room existence: element-hq/synapse#18263
I'm unsure if the random delay / constant time part requires an implementation. The spec already mentions this for the event reporting endpoint but Synapse doesn't appear to implement it and it's not part of element-hq/synapse#18120 either.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've added the removal of
score
into element-hq/synapse#18263. Synapse didn't do anything else than assert the type anyway.