Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions quest/m0/ci/five-minutes.md
Original file line number Diff line number Diff line change
Expand Up @@ -134,7 +134,9 @@ What remains:
- Admission still makes the measured net route exceed 5 minutes despite its
command taking 227 s. The 2026-10-29 admission decision above remains with
moq-dev/quest.
- Repeat the default FFI route after the Gradle fix lands, without the override.
- Repeat the default FFI route without the override; the Gradle fix
([moq-dev/moq#7632](https://github.com/moq-dev/moq/pull/7632)) merged
2026-11-04.
- Keep the dependency-build and cache work with
[hakari](/quest/m0/ci/hakari.md). Decided 2026-10-30 (user): defer parallel
Rust lint until that quest lands, then measure it. The slow-tests reduction
Expand All @@ -148,8 +150,6 @@ What remains:

- moq-dev/quest: admission must keep the whole gate inside the worker's cache
window; repeat the net measurement after its admission work lands.
- moq-dev/moq: [moq-dev/moq#7632](https://github.com/moq-dev/moq/pull/7632)
must land before repeating the default FFI measurement without `GRADLE_CMD`.

## Related

Expand Down
7 changes: 7 additions & 0 deletions quest/m0/ci/publish-release-only.md
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,13 @@ What remains is the maintainer's, in order:
2. Then drop `main` from `publish`'s deployment policy, and remove
`DISCORD_WEBHOOK` and `CACHIX_AUTH_TOKEN` from `publish`.

Decided 2026-11-04 (maintainer, /quest-unblock): ✅ cut the combined patch
release on `release` soon, once [moq-dev/moq#7596](https://github.com/moq-dev/moq/pull/7596)'s
remaining ports and [moq-dev/moq#7680](https://github.com/moq-dev/moq/pull/7680)
are ready, and merge `main` into `release` for the next minor line
separately, later. Not chosen: one combined cut that merges `main` first, or
neither this week.

Public API: none. Wire: none.

## Related
Expand Down
7 changes: 7 additions & 0 deletions quest/m0/ffi-shutdown-backport.md
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,13 @@ publishes 0.1.1 and exercises the unproven path in
[A Dart tag publishes unattended](/quest/m2/release/dart-publish.md). Decided
2026-10-30 (maintainer): push it during the combined cut, not before.

Decided 2026-11-04 (maintainer, /quest-unblock): ✅ cut the combined patch
release on `release` soon, once [moq-dev/moq#7596](https://github.com/moq-dev/moq/pull/7596)'s
remaining ports and [moq-dev/moq#7680](https://github.com/moq-dev/moq/pull/7680)
are ready, and merge `main` into `release` for the next minor line
separately, later. Not chosen: one combined cut that merges `main` first, or
neither this week.

Public API: none. Wire: none.

## Related
Expand Down
17 changes: 15 additions & 2 deletions quest/m1/bench-ci.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,11 +9,24 @@ runs on the hosted ARM runners, and a deliberately slowed bench fails the job.

## Waiting

- Maintainer: dispatch the Nightly runs on `main` below (A/A and slowed
bench); agents never dispatch Nightly. Check each run's `paired.json`.
- Maintainer: dispatch the slowed-bench Nightly run below. Agents never
dispatch Nightly; the 2026-11-04 authorization covered the two A/A runs only.
- moq-dev/moq: once A/A run
[41921320690](https://github.com/moq-dev/moq/actions/runs/41921320690)
finishes, dispatch the second A/A run under the same authorization
(`gh workflow run nightly.yml --ref main -f bench-base=<main's head>`; as of
2026-11-04 it is not dispatched yet), then read both runs' `paired.json` (the
`bench-rust*` artifacts). Find the second with `gh run list --repo
moq-dev/moq --workflow nightly.yml --event workflow_dispatch`.

## Plan

Decided 2026-11-04 (maintainer, /quest-unblock): ✅ authorize one-time agent
dispatch of the two A/A Nightly runs on `main`, with `bench-base` set to each
run's own commit. Not chosen: the maintainer dispatching them, or skipping.
Nightly's concurrency group cancels an in-progress run, so the second is
dispatched only after the first finishes.

- A/A: dispatch Nightly on `main` with `bench-base` set to the run's own
commit, at least twice. Read each paired group's `paired.json` (the
`bench-rust*` artifacts, or `bench/paired.json` on the site): the spread of
Expand Down
13 changes: 9 additions & 4 deletions quest/m1/cluster-routing/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -41,7 +41,12 @@ Decided with the maintainer, 2026-10-13:
- An origin declares its own routing identity, fresh per process or tab
restart and shared across its live sessions. It is unverified. Decided 2026-11-02 (maintainer, /quest-unblock): no
key proof is planned; if a use case needs a provable identity, moq-lite-07's
route identity becomes a public key, independent of P2P. Receiver-stamped
route identity becomes a public key, independent of P2P. Decided 2026-11-04
(maintainer, /quest-unblock): scoped forgery is that use case. A peer proves
a route key once per announce stream, its own identity derives from it, and
every identity it sends while scoped is qualified under it
([Scoped peers prove a route key](/quest/m1/cluster-routing/routes/proof.md)).
Receiver-stamped
per-session identities were dropped 2026-10-26 (maintainer): in a mesh,
two relays stamping their sessions from one origin forward its routes as
two identities a third relay cannot recognize as one. Keep administrative
Expand All @@ -51,9 +56,9 @@ Decided with the maintainer, 2026-10-13:
qualified when its publish grant is scoped, and stays bare when it is
unscoped, so a forger only damages routes inside its own scope. (Decided
2026-10-14; own identities included 2026-10-26, since they are no longer
proven.) How a scoped peer's identities are keyed waits on the maintainer in
[Route session integration](/quest/m1/cluster-routing/routes/session.md): the
issuer-qualified subject key was dropped 2026-11-02 (maintainer).
proven.) The issuer-qualified subject key was dropped 2026-11-02
(maintainer); since 2026-11-04 a scoped peer's identities are keyed by the
route key it proves ([Route session integration](/quest/m1/cluster-routing/routes/session.md)).
- Matching explicit epochs authorize resume. Epochless changes retain
Restart and fresh resolution; neither node identity nor a healthy
alternate grants permission to splice content.
Expand Down
7 changes: 6 additions & 1 deletion quest/m1/cluster-routing/routes/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -87,7 +87,11 @@ The wire sketch is in [wire](/quest/m1/cluster-routing/routes/wire.md); the bull
shared across its live sessions, decided 2026-10-13. It is unverified.
Decided 2026-11-02 (maintainer, /quest-unblock): peer grants are dropped and no key proof is
planned; a provable identity, if a use case needs one, makes this route
identity a public key, independent of P2P.
identity a public key, independent of P2P. Decided 2026-11-04 (maintainer,
/quest-unblock): scoped forgery is that use case; a peer proves a route key
once per announce stream, its own identity derives from that key, and every
identity it sends while scoped is qualified under it
([Scoped peers prove a route key](/quest/m1/cluster-routing/routes/proof.md)).
Decided 2026-10-26 (maintainer): receivers do not stamp a fresh identity
per session. In a mesh, two relays stamping their sessions from one origin
forward its routes as two identities a third relay cannot recognize as
Expand Down Expand Up @@ -496,6 +500,7 @@ benchmarks.

- [lite-07 route wire](/quest/m1/cluster-routing/routes/wire.md) - lite-07 and the draft carry ROUTE, ANNOUNCE_FLOOR, ROUTE_REQUEST, and path-less ANNOUNCE
- [Route session integration](/quest/m1/cluster-routing/routes/session.md) - one router per origin node drives the origin and every lite-07 announce stream
- [Scoped peers prove a route key](/quest/m1/cluster-routing/routes/proof.md) - a peer proves a route key once per announce stream, and its forwarded identities are qualified by it
- [Announcement availability](/quest/m1/cluster-routing/routes/availability.md) - announcements report availability, and a let-go is offline rather than END, in Rust and JS
- [js/net route layer](/quest/m1/cluster-routing/routes/js.md) - `js/net` encodes, decodes, and resolves the route layer
- [Route layer acceptance](/quest/m1/cluster-routing/routes/acceptance.md) - forwarding fixtures ported and the model deleted, incident fixtures, end-to-end checks, and benchmarks
Expand Down
74 changes: 74 additions & 0 deletions quest/m1/cluster-routing/routes/proof.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,74 @@
# [M] Scoped peers prove a route key

## Goal

A lite-07 peer proves a route key once per announce stream. The requester
sends a nonce in ANNOUNCE_REQUEST, and the publisher's ANNOUNCE_OK carries its
public key and a signature over that nonce, bound to the session. A relay
files every identity a scoped, proven peer sends, its own included (the
2026-10-26 rule in [Routes and announces](/quest/m1/cluster-routing/routes/README.md)),
under H(domain, proven key, id), so a scoped peer that forges or inflates
another origin's identity only affects claims inside its own grant. Relays
that share one key file present an origin under the same qualified key
whichever of them carries it, so a reader of relay l1's own broadcast fails
over through sibling l2 when l1's direct uplink is lost. Unscoped peers stay
bare, and a scoped peer that proves nothing gets a per-session key. This lands
in lite-07 before the cut.

## Plan

Decided 2026-11-04 (maintainer, /quest-unblock), from the keying question in
[Route session integration](/quest/m1/cluster-routing/routes/session.md).
This reverses the 2026-10-25 and 2026-11-02 decisions that no identity proof
gates the lite-07 cut. Options shown for each:

- Proof timing: ✅ before the lite-07 cut. Not chosen: an issuer-named route
label in the grant now with the proof after the cut, or a per-session key.
- Where a relay's key comes from: ✅ a key file. An operator gives all its
relays the same file, so two separate relays (the l1/l2 two-uplink fixture,
`scoped_delivery_through_two_uplinks_reaches_one_key`) fail over as one key.
A relay without a file generates one. The own identity stays fresh per
restart as H(key, per-process salt), as `restart_reusing_identity_is_refused`
requires. Not chosen: a fresh key per process.
- Browsers: ✅ prove with WebTransport's `exportKeyingMaterial(label, context,
outputLength)` where the browser exposes it, detected at session setup (no
runtime fallback); otherwise unproven with a per-session key. As of
2026-11-04 only Firefox ships it (spec-matching from Firefox 160); Chrome has
an Intent to Prototype and Safari nothing. Tabs do not forward routes, so an
unproven tab loses little. Not chosen: browsers never proven, or signing the
nonce with the dialed server name and role.
- Identity on the wire: ✅ ROUTE keeps its u64 node ID, the first 64 bits of
H(key, salt), so the router's `Id` type and ROUTE stay as they are. Not
chosen: the full 32-byte key in each ROUTE.

Own identity, settled in review of
[moq-dev/moq#7919](https://github.com/moq-dev/moq/pull/7919): it is qualified
like any other, not filed bare. Filed bare, l1's own origin would reach c1 as
its bare id but c2 (through l2) as a qualified key, so the two would not fail
over to each other. Extend the two-uplink fixture so l1 serves its own
broadcast and loses its direct uplink, and the reader fails over through l2.

Binding: a bare nonce can be relayed (a peer passes another relay's signed OK
off as its own), so the signature covers the session too. Rust uses the TLS
exporter (`rs/moq-quic` already exposes one); browsers use
`exportKeyingMaterial`. Settle the exact signed transcript (nonce, exporter
output, role, a domain label) and record it in the draft.

Cost, estimated 2026-11-04: about 16 bytes of nonce in ANNOUNCE_REQUEST and
about 90 bytes of key and signature in ANNOUNCE_OK, no extra round trip, and
one Ed25519 sign and verify per announce stream rather than per route update.

Reuse: `rs/moq-crypto`'s `sign::PublicKey` and `SecretKey::load_or_generate`,
backed by `rind` Ed25519, and `js/crypto`'s WebCrypto Ed25519 signing.
moq-crypto's `sign` feature depends on moq-net, so either moq-net calls `rind`
directly or the key type moves to a lower crate. Decide which in the PR.

Public API: moq-net gains the proof on announce streams; moq-relay and
moq-tokio gain a route key file option. Wire: ANNOUNCE_REQUEST and ANNOUNCE_OK
change in lite-07 and its draft.

## Related

- [Route session integration](/quest/m1/cluster-routing/routes/session.md) - files identities under the key this proves
- [lite-07 route wire](/quest/m1/cluster-routing/routes/wire.md) - the announce messages this extends
- [P2P](/quest/m2/p2p/README.md) - its roster id is already planned as a public key
72 changes: 21 additions & 51 deletions quest/m1/cluster-routing/routes/session.md
Original file line number Diff line number Diff line change
Expand Up @@ -96,57 +96,27 @@ epoch. The route layer's own origin-node identity, with its seqno and
revision, stays separate ("keep administrative identity and broadcast epoch
separate", in the epic).

## Waiting

- Maintainer: how a scoped peer's routing identities are keyed. The router
takes the key from the caller (`Scope::qualify`), so nothing else waits on
it. Recommendation: (1), a route label carried in the grant.

What the key is for. Every origin declares a routing identity: an opaque
number nobody verifies. A peer admitted under a scoped grant (say it may
publish only `acme/**`) could declare a victim's identity. One route
serves every claim of an identity, so the forgery would capture the
victim's claims outside `acme/`, and its seqno would outlive it
(`forged_identity_hijacks_claims_outside_the_forgers_scope`). So a relay
files every identity a scoped peer sends under a qualified key, derived
from something about that peer and the identity, and counts claims under
the key only inside the grant. The question is what that something is.
It must stay the same across token refresh, be shared by one operator's
sessions so losing one uplink fails over instead of restarting readers
(`scoped_delivery_through_two_uplinks_reaches_one_key`), differ between
operators, and work for sessions admitted with no token (public rules,
mTLS certificates, or a third-party auth server).

Why not the credential subject: `hash(subject, id)` was decided
2026-10-15 and dropped by the maintainer 2026-11-02, together with
refusing scoped tokens that carry no subject; the implementation is in
[moq-dev/moq@453e1ec](https://github.com/moq-dev/moq/commit/453e1ece7c5435ee19541cfc2e9bbe0b6fbc430e).
It existed only for tokens, so token-less admission had no key. Rejected
earlier: the grant's pattern union, which refresh and withdrawal change
and two uplinks need not share, and the connection root, which two scoped
operators under one root would share.

Example: operator Acme runs a local relay L with uplinks to CDN relays C1
and C2. Each admits L with a token scoped to `acme/**`, refreshed hourly.

1. A route label carried in the grant. Whoever issues it names it: a
token claim such as `route: "acme"`, a public rule's entry, or an mTLS
mapping. The key hashes the label with the identity in its own domain.
Both of Acme's tokens say `acme`, so C1 and C2 file L's origins under
one key, refresh keeps it, and losing C1 fails over through C2.
Another operator's label differs. A grant with no label gets a
per-session key, as in (2). Costs a grant field in `moq-auth`,
`js/auth`, and the docs, and issuers must not hand two operators one
label.
2. A per-session key, with nothing to configure. L's sessions to C1 and
C2 get two keys, so losing C1 restarts the readers it served onto C2's
key instead of failing over, and so does a refresh that reconnects.
Simplest; gives up the two-uplink fixture.
3. A key the relay operator configures per scoped remote, such as C1 and
C2 both mapping admissions under `acme/**` to `acme`. No grant change,
but every CDN relay's configuration must agree, and it names patterns
rather than who connected, so two operators admitted under one pattern
share a key.
Decided 2026-11-04 (maintainer, /quest-unblock): a scoped peer's identities
are keyed by a route key the peer proves, specified in
[Scoped peers prove a route key](/quest/m1/cluster-routing/routes/proof.md).
Every identity a scoped, proven peer sends, its own included, is qualified
under its proven key, so relays sharing a key file fail over as one. A peer that proves no
key gets a per-session key. `Scope::qualify` still takes the key from the
caller, so this quest does not wait on the proof. Options shown:

- ✅ A route key the peer proves once per announce stream, before the lite-07 cut
- A route label carried in the grant (the earlier recommendation), with the
proven key added after the cut
- A per-session key now
- A key the relay operator configures per scoped remote

Research behind it: proving a neighbor's own key cannot remove qualification,
since a public key is public and a scoped peer could still forward a victim's
key with an inflated seqno; origin-signed routes still leave cost lies and
blackholing, at the price of per-update signatures and expiry refreshes. Keyed
by a proven key, every remaining attack stays inside the forger's grant, with
no grant field, no issuer-coordinated labels, and support for token-less
admission.

## Required

Expand Down
11 changes: 7 additions & 4 deletions quest/m1/crates-trusted-publisher.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,10 +9,10 @@ Delete this quest once every crate has one.

## Waiting

- Maintainer: add the trusted publisher in each crate's crates.io settings,
after a first publish by token for the crates not yet on crates.io (listed
under Plan). Settings pages need a crates.io login, so check with a
`release-rs.yml` run that publishes through OIDC.
- Maintainer: after the next release publishes kio-macros, moq-c,
moq-congestion, moq-crypto, moq-ice, moq-quic, and moq-rtsp by token, add
their trusted publishers the same way. Settings pages need a crates.io
login, so check with a `release-rs.yml` run that publishes through OIDC.

## Plan

Expand All @@ -39,3 +39,6 @@ moq-quic, moq-rtsp. moq-crypto became publishable in
[moq-dev/moq#7733](https://github.com/moq-dev/moq/pull/7733). `CARGO_REGISTRY_TOKEN` is revoked only after all of them have one.
Settings pages need a crates.io login, so the check is a `release-rs.yml` run
publishing through OIDC.

Done 2026-11-04 (maintainer, /quest-unblock): all 31 published crates have
the trusted publisher. Only the 7 unpublished crates remain.
6 changes: 5 additions & 1 deletion quest/m1/lite07-finalize.md
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,11 @@ grants and metadata-only subscriptions do not gate it: neither is clustering
stability. Both are additive and, per the freeze policy, target whichever lite
wip is current when they land. Decided 2026-10-25: no routing identity proof
gates the cut, and lite-07 routing identities stay declared and unverified
(2026-11-02: peer grants are dropped and no key proof is planned). Decided 2026-10-26
(2026-11-02: peer grants are dropped and no key proof is planned). Reversed
2026-11-04 (maintainer, /quest-unblock): a peer proves a route key once per
announce stream before the cut, through
[Scoped peers prove a route key](/quest/m1/cluster-routing/routes/proof.md) in
[Routes and announces](/quest/m1/cluster-routing/routes/README.md). Decided 2026-10-26
(maintainer): receivers do not stamp per-session identities either, since
that splits one origin into several identities across a relay mesh.

Expand Down
12 changes: 5 additions & 7 deletions quest/m2/usage-settings-once.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,13 +9,6 @@ declaring every merged flag a second time) are deleted after a bump. Source
presence, precedence, deprecated-name rejection, and provenance keep working.
CLI behavior and wire are unchanged.

## Waiting

- jdx/usage: a usage-rs release ships
[jdx/usage#1608](https://github.com/jdx/usage/issues/1608). Check the issue
and usage-rs releases. If upstream declines, delete this quest and keep the
drift test.

## Plan

Filed 2026-11-02 (maintainer-approved, /quest-unblock), building on the
Expand All @@ -27,6 +20,11 @@ Today `usage::Cli` and `usage::Config` reject each other's attributes, and
Propose inferring those from the field so one declaration serves both. If
upstream declines, delete this quest and keep the drift test.

Shipped 2026-11-03 (checked 2026-11-04, /quest-unblock): usage-lib
[7.2.0](https://github.com/jdx/usage/releases/tag/v7.2.0) lets derive flags
declare their own settings ([jdx/usage#1610](https://github.com/jdx/usage/pull/1610),
closing jdx/usage#1608).

## Related

- [Upstream unknown-env option in usage-rs](/quest/m2/usage-unknown-env.md) - another jdx/usage ask; file both together if both are approved
Loading
Loading