tajniak81andClaude Opus 5
d187ad0d17
CI / Test (ubuntu-latest) (push) Canceled after 0s
CI / Test (windows-latest) (push) Canceled after 0s
CI / Storage tests on PostgreSQL (push) Canceled after 0s
CI / Complement (federation off) (push) Canceled after 0s
CI / golangci-lint (push) Canceled after 0s
CI / Smoke test the binary (push) Canceled after 0s
Implements PLAN.md Phase 4: the read model. /sync, /messages and /context, the filters that narrow them, and the streams a response carries that are not room events — typing, receipts, presence and to-device messages. With this a client can be used against Warp, which is milestone M1. The one architectural choice is that the syncapi has no copy of the rooms: it reads them through the roomserver. Dendrite gives its syncapi its own tables and feeds them from the output stream, which is right for a design where the two can be separate processes, and Dendrite's could be. Warp is one process by decision, and there the copy buys nothing and costs the hardest class of bug a homeserver has — a read model that disagrees with the room, where a client is shown a message that was rejected and the only way to find out is to compare two tables. So the roomserver grew a read model (stream.go, roomsync.go, visibility.go) rather than exporting its store: reading an event means parsing it under its room's version, and a second copy of the version registry, the parser and the client projection would be a second place for them to disagree. syncapi: stream tokens naming a position in each of five independent streams, so a client that has seen receipt 7 but only event 4 can say so in one opaque string. A token with fewer positions than this build writes is accepted and the rest read as zero, which is what lets a token stored before a stream existed go on working. The PDU position comes from MAX(stream_ordering) and not from the counter row the other four use: the counter is advanced inside the writing transaction, so a token taken from it would name a position whose event no reader can see yet, and the event would be skipped for good once that transaction committed. Long polling goes through a notifier that wakes by room and by user rather than globally — an event wakes the listeners of its room, a membership event also wakes the user it names, so an invite reaches a client that is not in the room yet. A wake-up is only ever a hint: the listener re-reads the streams and decides for itself what changed, which is what makes it safe for the notifier to hold no state about what each client has seen. The listener registers before the read, so an event landing in between leaves the channel signalled and the sync reads again rather than blocking on a change it just missed. EventPersisted must not block, because it is called with the room's write lock still held. "Something happened" is not "a token moved". A stream can move without this user being concerned by it, and returning on every such move would turn a long poll back into a busy loop. The same trap caught presence twice over: a client says it is still online by syncing, so ?set_presence= makes every sync a presence write, and a stream that moved each time would wake every peer, which would sync, which would move it again. SetPresence therefore reports whether anything a peer would notice actually changed, and a write that only refreshes the last-active timestamp leaves the stream alone. Timeline windows are named by their two ends and a direction, Low exclusive and High inclusive in both directions, rather than by "from" and "to" as the wire format spells them. Each of those would otherwise be the low bound in one direction and the high bound in the other, which is exactly the off-by-one every pagination bug is made of; the handlers translate once and nothing below them thinks about it again. A sync's timeline is read backwards and reported forwards, because a client wants the newest events of the window it missed — a forward read with a limit hands it the oldest and then calls the rest a gap. visibility: history visibility is now a span rather than a yes or no, which is what Phase 3 deferred here. A user who has left a room is served the room as it was when they left — their state reads, their members and their timeline — instead of being refused, and is bounded above at their own membership event so a kicked client cannot go on streaming. Their span is found from their own member events, so a user who joined, left and rejoined sees their latest span. ?at= on /members is honoured for the same reason. What is left is that the span is computed from the room's current history_visibility where the spec judges each event by the one in force when it was sent; that is a state read per event and belongs with open question 1. Typing is held in memory and nowhere else: notifications expire in seconds, are worthless across a restart, and arrive one per keystroke-burst per room, so persisting them would be the busiest table in the database holding the least valuable data in it. Receipts only ever advance, compared on the position of the event acknowledged rather than on arrival, because two of a user's clients read at different speeds and the slower one must not drag the marker backwards. To-device queues are per device and drained by the next sync rather than by the one that delivered them, so a client that crashes in between gets the batch again — the right way round, since a lost key share leaves a message permanently undecryptable and a duplicate is ignored. A wildcard addressee is expanded when sent, not stored, because "every device" means the ones that existed then. Presence is off by default, which answers PLAN.md open question 3. It is the noisiest thing in Matrix and the one genuinely fan-out notification here; a private homeserver gains little from it. With it off the endpoints refuse rather than disappearing, because a client that gets a 404 from /presence concludes the server is too old to have it and stops asking. storage: one migration per driver adding the receipt, to-device, presence and filter tables with a counter row each, and a stream position on the existing account data table rather than a parallel one — there is exactly one current value per (user, room, type), and a second table would be a second thing to keep in step. Account data is reported in an event list but is not an event, so it gets its own response type: a room event with an empty sender and no timestamp would make a client decide what that meant. Deliberately missing, each named in docs/sync.md: notification counts (they need push rules, Phase 8), device lists (the keyserver, Phase 7), anything remote (Phase 5), threads (Phase 10). device_lists is sent as two empty lists because a client reads a missing one as malformed; the notification counts are left out entirely, because a zero there is a claim and an absent field is not. Tested through HTTP rather than against the builder, because what is worth pinning down is the response a client sees — which section a room lands in, whether a token handed back can be handed in again, whether paging terminates and loses nothing. The storage tests run against both SQL drivers; the notifier has its own tests for the concurrent case. A live server was driven end to end, which is what caught the two bugs the unit tests had missed: a backwards /messages page must be in reverse chronological order, and `start` must name where the page began rather than position zero when ?from= was omitted. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
158 lines
9.1 KiB
Markdown
158 lines
9.1 KiB
Markdown
# Configuration
|
|
|
|
Warp reads one YAML file. Every setting can also come from the
|
|
environment, and the environment wins. With no file at all the defaults
|
|
plus `WARP_*` variables are used, which is how the container image runs.
|
|
|
|
Where the file comes from, in order:
|
|
|
|
1. `--config <path>`
|
|
2. `$WARP_CONFIG`
|
|
3. `./warp.yaml`, if it exists
|
|
4. nothing — defaults and environment only
|
|
|
|
An unknown field is a fatal error. A typo in a security-relevant setting
|
|
must not be ignored.
|
|
|
|
List-valued variables are comma-separated. A single `-` clears a list:
|
|
`WARP_CORS_ALLOWED_ORIGINS=-`.
|
|
|
|
## Settings
|
|
|
|
| YAML | Environment | Default | Notes |
|
|
|---|---|---|---|
|
|
| `global.server_name` | `WARP_SERVER_NAME` | — | Required. Part of every ID; cannot be changed later. |
|
|
| `global.private_key` | `WARP_PRIVATE_KEY` | `./keys/signing.key` | Written by `warp generate-keys`. |
|
|
| `global.well_known.client_base_url` | `WARP_WELL_KNOWN_CLIENT_BASE_URL` | `https://<server_name>` | Advertised to clients. |
|
|
| `global.well_known.identity_server_base_url` | `WARP_WELL_KNOWN_IDENTITY_SERVER_BASE_URL` | empty | Omitted from the document when empty. |
|
|
| `global.well_known.server_address` | `WARP_WELL_KNOWN_SERVER_ADDRESS` | empty | `host:port` for other homeservers. Empty = no delegation, other servers fall back to SRV/A. |
|
|
| `listen.address` | `WARP_LISTEN_ADDRESS` | `:8008` | |
|
|
| `listen.read_header_timeout` | `WARP_READ_HEADER_TIMEOUT` | `10s` | |
|
|
| `listen.read_timeout` | `WARP_READ_TIMEOUT` | `60s` | |
|
|
| `listen.write_timeout` | `WARP_WRITE_TIMEOUT` | `120s` | Must stay above `sync.long_poll_timeout`, or a blocked `/sync` is cut off before it can answer. |
|
|
| `listen.idle_timeout` | `WARP_IDLE_TIMEOUT` | `120s` | |
|
|
| `listen.max_header_bytes` | `WARP_MAX_HEADER_BYTES` | `1048576` | |
|
|
| `listen.shutdown_timeout` | `WARP_SHUTDOWN_TIMEOUT` | `30s` | Grace period for in-flight requests. |
|
|
| `database.driver` | `WARP_DB_DRIVER` | `sqlite` | `sqlite`, `postgres`, `pocketbase` (not implemented). |
|
|
| `database.connection_string` | `WARP_DB_CONNECTION_STRING` | SQLite file in the working directory | |
|
|
| `database.max_open_conns` | `WARP_DB_MAX_OPEN_CONNS` | `10` | Ignored by SQLite: writes are serialised onto one connection. |
|
|
| `database.max_idle_conns` | `WARP_DB_MAX_IDLE_CONNS` | `5` | |
|
|
| `database.conn_max_lifetime` | `WARP_DB_CONN_MAX_LIFETIME` | `1h` | |
|
|
| `database.auto_migrate` | `WARP_DB_AUTO_MIGRATE` | `true` | `false` makes startup fail while migrations are pending. |
|
|
| `client_api.cors_allowed_origins` | `WARP_CORS_ALLOWED_ORIGINS` | `["*"]` | The spec expects any origin. |
|
|
| `client_api.trust_x_forwarded_for` | `WARP_TRUST_X_FORWARDED_FOR` | `false` | Only behind a proxy that overwrites the header; the rate limiter keys on the result. |
|
|
| `client_api.access_token_lifetime` | `WARP_ACCESS_TOKEN_LIFETIME` | `0s` | `0` = never expire. With a lifetime, clients may ask for a refresh token at login. |
|
|
| `client_api.refresh_token_lifetime` | `WARP_REFRESH_TOKEN_LIFETIME` | `0s` | `0` = never expire. |
|
|
| `client_api.rate_limit.enabled` | `WARP_RATE_LIMIT_ENABLED` | `true` | |
|
|
| `client_api.rate_limit.default.burst` / `.interval` | — | `30` / `100ms` | Every rate-limited endpoint without its own bucket. |
|
|
| `client_api.rate_limit.login.burst` / `.interval` | — | `10` / `1s` | `/login` and `/refresh`. |
|
|
| `client_api.rate_limit.register.burst` / `.interval` | — | `5` / `2s` | `/register` and `/register/available`. |
|
|
| `client_api.rate_limit.room_create.burst` / `.interval` | — | `10` / `1s` | `/createRoom`, which writes a dozen events per call. |
|
|
| `client_api.rate_limit.join.burst` / `.interval` | — | `10` / `1s` | The two join endpoints. |
|
|
| `client_api.rate_limit.sync.burst` / `.interval` | — | `100` / `20ms` | `/sync` and the writes behind it: typing, receipts, to-device, presence. |
|
|
| `sync.presence` | `WARP_SYNC_PRESENCE` | `false` | Off by default; see below. |
|
|
| `sync.typing_timeout` | `WARP_SYNC_TYPING_TIMEOUT` | `30s` | How long a typing notification lasts, and the cap on what a client may ask for. |
|
|
| `sync.long_poll_timeout` | `WARP_SYNC_LONG_POLL_TIMEOUT` | `30s` | The cap on a `/sync` `?timeout=`. |
|
|
| `federation.enabled` | `WARP_FEDERATION_ENABLED` | `false` | |
|
|
| `federation.allowlist` | `WARP_FEDERATION_ALLOWLIST` | `[]` | Non-empty = federate only with these. |
|
|
| `federation.denylist` | `WARP_FEDERATION_DENYLIST` | `[]` | Always wins over the allowlist. |
|
|
| `federation.allow_runtime_toggle` | `WARP_FEDERATION_ALLOW_RUNTIME_TOGGLE` | `true` | See below. |
|
|
| `registration.enabled` | `WARP_REGISTRATION_ENABLED` | `false` | |
|
|
| `registration.allow_runtime_toggle` | `WARP_REGISTRATION_ALLOW_RUNTIME_TOGGLE` | `true` | |
|
|
| `registration.tokens` | `WARP_REGISTRATION_TOKENS` | `[]` | Non-empty = sign-up also needs one of these, as the `m.login.registration_token` stage. |
|
|
| `logging.level` | `WARP_LOG_LEVEL` | `info` | `debug`, `info`, `warn`, `error`. |
|
|
| `logging.format` | `WARP_LOG_FORMAT` | `text` | `text` or `json`. |
|
|
| `logging.add_source` | `WARP_LOG_ADD_SOURCE` | `false` | File and line in every record. |
|
|
| `metrics.enabled` | `WARP_METRICS_ENABLED` | `true` | |
|
|
| `metrics.path` | `WARP_METRICS_PATH` | `/metrics` | Served on the same listener. |
|
|
|
|
The rate limit buckets share one Go type, so per-bucket environment
|
|
variables would all collide on the same name; they are YAML only.
|
|
`WARP_RATE_LIMIT_ENABLED=false` turns the whole limiter off, which is what a
|
|
test harness wants.
|
|
|
|
## Rate limiting
|
|
|
|
Each bucket allows `burst` requests back to back and returns one slot every
|
|
`interval`, keyed per client address. Over the limit is
|
|
`429 M_LIMIT_EXCEEDED` with a `retry_after_ms` the client is expected to
|
|
honour.
|
|
|
|
Buckets are per endpoint class, so a client hammering `/login` cannot spend
|
|
the budget of `/profile`. The key is the client address and not the user,
|
|
because the limiter runs before the access token is resolved — a limiter
|
|
that had to query the database to find its own key would be a denial of
|
|
service of its own. Behind a reverse proxy, set
|
|
`client_api.trust_x_forwarded_for: true`, or every client shares one bucket.
|
|
|
|
The admin API under `/_warp/admin/v1` is not rate limited: it needs an admin
|
|
token, and bulk operations from the panel are exactly the traffic a limiter
|
|
would get in the way of.
|
|
|
|
## Presence
|
|
|
|
`sync.presence` is off by default. Presence — who is online, and the status
|
|
message they set — is the noisiest thing in Matrix: every user's state
|
|
change is published to everybody they share a room with, which over
|
|
federation is quadratic in the worst case. Synapse deployments routinely
|
|
turn it off for that reason, and a private homeserver with a dozen users
|
|
gains little from it.
|
|
|
|
With it off:
|
|
|
|
- `GET` and `PUT /presence/{userId}/status` answer `403 M_FORBIDDEN`. They
|
|
are deliberately not unregistered: a client that gets a 404 from
|
|
`/presence` concludes the server is too old to have it and stops asking,
|
|
where a refusal says this one has it switched off.
|
|
- The `presence` section of a `/sync` response is empty.
|
|
- `?set_presence=` on `/sync` is ignored rather than refused — it rides on
|
|
every sync a client sends, and failing them all would make the server
|
|
unusable.
|
|
|
|
It is a boot-time setting, not a runtime one: turning it on starts
|
|
publishing a stream that nothing was tracking before, and a server that
|
|
could do that mid-flight would owe every connected client a resync.
|
|
|
|
## Long polling
|
|
|
|
`/sync` holds a request open until something happens or the timeout
|
|
expires, so `listen.write_timeout` has to stay above
|
|
`sync.long_poll_timeout` — otherwise the listener cuts off a blocked sync
|
|
before it can answer, and the client sees a dropped connection instead of
|
|
an empty response.
|
|
|
|
Raising `sync.long_poll_timeout` does not make a client hear about messages
|
|
sooner: a blocked sync is woken the moment anything concerning it is
|
|
written. All the timeout decides is how often an idle client
|
|
re-establishes its request.
|
|
|
|
## Settings that change at runtime
|
|
|
|
`federation.enabled` and `registration.enabled` are also stored in the
|
|
`server_settings` database table, which is what the admin API writes to.
|
|
|
|
- With `allow_runtime_toggle: true` (the default) the stored value wins
|
|
over the config file at boot, and the admin API may change it. The
|
|
change takes effect on the next request; no restart.
|
|
- With `allow_runtime_toggle: false` the config file is authoritative and
|
|
the admin API refuses to change the setting. Any previously stored value
|
|
is kept but ignored, so re-enabling the toggle restores the
|
|
administrator's last choice.
|
|
|
|
Everything else is read once at startup.
|
|
|
|
## What federation off actually means
|
|
|
|
While `federation.enabled` is false:
|
|
|
|
- `/.well-known/matrix/server` answers `404 M_UNRECOGNIZED`, and so will
|
|
`/_matrix/federation/*` and `/_matrix/key/*` once they exist.
|
|
- Nothing outbound runs — in Phase 0 there is nothing outbound yet; from
|
|
Phase 5 the transaction queue, key fetcher and remote media fetch stay
|
|
stopped.
|
|
- Client-Server endpoints that would need a remote server answer
|
|
`M_FORBIDDEN` with a message saying federation is disabled.
|
|
- Events are still signed with the server key, and room and event IDs are
|
|
unchanged. A room created while federation is off federates normally
|
|
once it is switched on.
|